DEV Community: funpal10-byte

Brand Architecture Framework: The Five Layers That Matter

funpal10-byte — Tue, 01 Sep 2026 18:24:55 +0000

A brand architecture framework isn't a design exercise. It's a capital-allocation decision that determines where your buyer's risk sits. There are five layers — masterbrand, sub-brand, endorsed, independent, invisible — and most Indian groups end up on layer three because it feels safe, not because it's correct.

I've sat through the meeting where this gets decided. It usually runs about ninety minutes, involves a slide with three logo lock-ups, and ends with someone senior saying "let's keep the parent name small in the corner for now." That corner logo then costs the business two years of muddled positioning and a media plan that has to fund two stories instead of one. Nobody ever writes that cost down anywhere.

The reason it goes wrong is that architecture gets treated as a branding question when it is really a question about accountability. Who does the customer blame when the product fails? Whoever they blame is the brand that has to be on the front of the box. Everything else is decoration.

Why most brand architecture decisions get made for the wrong reason

Three bad inputs dominate the room. The first is the org chart — if the businesses report separately, someone argues the brands should look separate. The second is history — an acquired brand has "equity we can't throw away," a claim almost never tested with actual data. The third is politics, which is the honest name for the first two.

None of those inputs is about the buyer. And the buyer's world has become measurably harder to navigate. McKinsey's B2B Pulse survey of nearly 4,000 decision makers across 13 countries found buyers now use an average of ten interaction channels before purchase, up from five in 2016. Every additional brand you put in front of that buyer has to earn its own recall across ten touchpoints. That's not a design cost. That's a media cost, repeated quarterly, forever.

Simplicity has a measurable price attached to it. Siegel+Gale's 2018 World's Simplest Brands study, covering more than 15,000 respondents across nine countries, found 55% of people would pay more for a simpler experience, 64% were more likely to recommend a brand that delivered one, and companies failing to deliver simplicity left an estimated $98 billion on the table. The same study reported that a stock portfolio of the world's simplest brands had outperformed the average of the major indexes by 679% since 2009.

I don't read that as proof that fewer brands is always better. I read it as proof that every brand you add has to pay rent.

What does a brand architecture framework actually decide?

A brand architecture framework decides three things, and only three. Which name the customer buys. Which name carries the warranty in their head. And which name gets the marketing budget when the year gets tight.

That third one is the tell. If you can't say, without hesitation, which brand gets funded first in a bad quarter, you don't have an architecture — you have a logo family. I've watched groups run four brands off one budget line and then act surprised when none of them registered. The question of how much a B2B company should spend on brand is unanswerable until the architecture question is settled, because the denominator keeps changing.

The stakes at group level in India are not small. Brand Finance's India 100 2025 ranking put Tata Group's brand value at USD 31.6 billion, up 10%, with the combined value of the top 100 Indian brands at USD 236.5 billion. Adani Group was the fastest riser at 82%. On a broader methodology, Kantar BrandZ valued India's top 100 brands at USD 523.5 billion in 2025 — roughly 13% of the country's GDP — with HDFC Bank at USD 44.99 billion and Mahindra up 53% to USD 5.5 billion. These are balance-sheet-scale numbers being shaped by a decision most companies make in one afternoon.

The five layers, and what each one costs

Here is the ladder. Layer one is maximum consolidation, layer five is maximum separation. The middle three are where the arguments happen.

| 1. Masterbrand | One name across everything. Descriptors, not brands: Tata Motors, Tata Steel, Tata Power | One promise, one warranty | Lowest per unit of awareness | Buyers across businesses overlap, and the parent already means something specific

| 2. Sub-brand | Parent plus a named product world with its own character: Birla Opus, Tata Neu | Parent's credibility, product's personality | Roughly 1.5x a masterbrand launch | Entering a category where the parent has permission but no product language

| 4. Independent | House of brands. Parent invisible in market: HUL's portfolio | Only the product brand | Full freight per brand, no transfer | Buyer segments genuinely don't overlap, or parent equity would constrain pricing

| 5. Invisible parent | Parent deliberately absent, sometimes contractually | Nothing of the parent | Full freight, plus a governance overhead | Parent association is a liability, or the business is being groomed for sale

The cost column is the part that usually gets left off the slide. It is also the only column a CFO will remember.

Which layer should your business sit on?

Four tests, in order. Run them honestly and the answer usually falls out in twenty minutes.

The blame test. When the product fails at 2am, whose name does the customer say out loud? If it's the parent, you are already on layer one or two whether your brand book says so or not.
The overlap test. What share of your buyers could buy from two or more of your businesses? Above roughly a third, consolidation pays for itself in media efficiency. Below a tenth, you are subsidising a story nobody in either segment needs.
The permission test. Does the parent name give you the right to be believed in the new category, or only the right to be noticed? Noticed is not enough. Aditya Birla had cement and distribution credibility, and it still spent to build a product brand rather than selling paint under a corporate name.
The exit test. Is there a realistic chance this business gets sold in five years? If yes, layer four or five, and accept the cost. Untangling a masterbrand later is far more expensive than running separately from day one.

Notice what isn't on that list: the org chart, the founder's preference, and whether the acquired brand's team will be upset. Those are real political constraints, and you will have to manage them. They are not inputs to the decision.

The endorsement trap

Here's where I part company with the standard advice. David Aaker's brand relationship spectrum treats the endorsed brand as a sensible middle path, and almost every agency deck reaches for it. I think it's the weakest layer on the ladder, and the most expensive.

Endorsement is the most-chosen layer in brand architecture and the least defensible one. It's a hedge, not a strategy.

The logic against it is simple. An endorsement asks the buyer to store two names and one relationship between them. In a category where the Ehrenberg-Bass Institute's 95-5 rule applies — John Dawes' 2021 finding that up to 95% of business buyers aren't in market at any given time — you are trying to build three memory structures instead of one, among people who aren't paying attention. The maths does not work. It's the same reason tracking raw brand awareness in B2B tells you so little: recall of a name is cheap, and recall of a relationship between two names is not.

The honest defence of endorsement is that it buys internal peace. That's a legitimate thing to buy. Just price it, put it in the budget line, and revisit it in eighteen months instead of letting it become permanent by default.

What Tata Neu and Birla Opus got differently

Two of the largest architecture bets in recent Indian business went in opposite directions, and both are instructive.

Tata Neu is a layer-two sub-brand built to sit above an existing set of strong businesses — a masterbrand wrapper applied after the fact. The scale is real: Tata's chairman has cited GMV of â‚¹46,515 crore within four years of launch. So is the cost. Tata Digital reported a net loss of â‚¹4,974 crore in FY26 on revenue of â‚¹35,990 crore, with BigBasket's B2C arm accounting for roughly 64% of the losses. The architecture question that sits underneath is whether customers who already trusted Croma and BigBasket needed a third name in between them and the parent.

Birla Opus went the other way and, so far, more cleanly. The Aditya Birla Group put roughly â‚¹10,000 crore behind a named product brand rather than selling paint as a corporate line extension. By March 2025 it had reached the top three decorative paint brands in India by revenue share, with around 50,000 dealers onboarded, 137 depots and coverage of over 6,600 towns. Outlook Business reported in 2026 that the brand had taken roughly 7% market share, with Asian Paints' share moving from 51% to 47%. The parent name is present, but the product brand carries the promise. Layer two, executed as layer two, not as a hedge.

The counterpoint a smart peer would raise: Birla Opus bought that share with distribution and capital, not architecture. Fair. Grasim's net debt rose from â‚¹4,300 crore in FY22 to â‚¹35,402 crore in FY25. But distribution gets you tried once. The name is what gets you specified the second time.

What the Tata Motors demerger says about structure versus brand

The cleanest recent proof that architecture and corporate structure are different questions came in 2025. Tata Motors' commercial vehicle business demerged effective 1 October 2025 and listed separately on 12 November 2025, splitting a company into two listed entities with different customers, technology platforms and business models.

The brand didn't split. Both entities still lead with Tata Motors. A fleet operator buying trucks and a family buying an SUV are about as far apart as two buyer sets get, and the group still concluded that one name serves both better than two. That's the blame test and the permission test answering the same way, independently of who owns the P&L.

If you take one thing from this: legal separation is a tax and governance decision. Brand separation is a memory decision. They are allowed to disagree, and when they do, the memory decision should usually win.

What this means for you on Monday

A working sequence, in the order I'd actually run it inside an engineering-led business:

Write down every name you currently spend money on. Product brands, sub-brands, programme names, internal initiative names that leaked outside. Most groups find twice what they expected.
Put a rupee figure against each. Media, collateral, trade, events, digital. If a name has no budget, it isn't a brand; delete it from the architecture and stop protecting it.
Run the four tests on each one. Blame, overlap, permission, exit. Record the answer in one sentence each, not a slide.
Kill or fold anything that fails three of four. This is where the savings are, and it's the only part of a rebrand that pays for itself inside a year.
Fix the naming rule before the logos. A one-page rule for how new products get named prevents the next ten arguments. Designing the lock-ups first guarantees you'll redo them.
Sequence the rollout by revenue, not by visibility. Signage and stationery are the last 10% of the work and the first thing everyone wants to discuss.

Globally, the direction of travel favours clarity. Interbrand's Best Global Brands 2025 put the total value of the top 100 at USD 3.6 trillion, up 4.4%, with Nvidia rising 116% and Tesla falling 35% in the same year. The gainers are, almost without exception, single-name businesses where the corporate brand and the product promise are the same thing. That's not a coincidence, and it isn't only a tech story — the same pattern shows up in how the strongest supply chain and logistics brands have consolidated over the last decade.

None of this gets easier with better tools. AI has changed a lot about brand work, but it can't tell you who your customer blames at 2am. That answer only comes from asking them.

So the question I'd put to anyone sitting on layer three right now: if you had to remove one name from your architecture this quarter, and the budget stayed the same, which one would you remove — and what has stopped you so far?

What AI Actually Changed in Brand Work, and What It Didn't

funpal10-byte — Tue, 01 Sep 2026 18:24:43 +0000

Eighteen months into using AI tools inside a brand function, the honest inventory looks nothing like the vendor pitch decks. Some things genuinely changed â€” first-draft speed, research synthesis, localisation across markets. Most of what actually matters in brand work â€” judgement, category insight, knowing which stat to cut because it's not quite right â€” didn't move at all.

I started using generative AI tools seriously for brand and communication work in 2024, first for first drafts, then for competitive research synthesis, then reluctantly for creative concepting because everyone on the team was already doing it quietly. The gap between what I expected and what actually happened is the most useful thing I can report back, because most of what's written about "AI in marketing" right now is either breathless or dismissive, and the reality sits in a much less exciting middle.

What genuinely got faster

First-draft speed on long-form content is the clearest win, and it isn't close. A brief that used to take a writer three to four hours to turn into a rough draft now takes 30 to 45 minutes of AI-assisted drafting followed by an hour of substantive editing â€” the editing hasn't gotten shorter, but the blank-page time has nearly disappeared. Competitive research synthesis is the second real win: pulling together what ten competitors said in their last four quarters of public communication used to be a day of manual reading; it's now a couple of hours of directed research plus verification.

McKinsey's 2024 research on generative AI in marketing functions found meaningful productivity gains concentrated specifically in content generation and summarisation tasks, which matches what I've seen directly â€” the gains are real, but they cluster tightly around a specific set of tasks rather than spreading evenly across the function.

What didn't change, no matter how good the tool got

Positioning didn't get easier. Deciding what a brand should actually stand for, in a specific category, against specific competitors, for a specific buyer who has specific frustrations â€” that's still slow, still requires sitting with customers and sales teams, and AI tools are actively bad at it because they default to generic, defensible-sounding positioning that could apply to almost any company in the category. I've had to explicitly fight a tool's own instinct toward blandness more times than I can count.

Category judgement is the second thing that didn't move. Knowing that a stat is technically accurate but will land wrong with a specific board, or that a campaign concept that tested well will read as tone-deaf in a market that just went through layoffs â€” that's pattern recognition built from years of being in rooms, and no tool I've used gets close to it. Gartner's research on generative AI in marketing organisations has flagged this specific gap â€” the tools accelerate execution but don't substitute for strategic judgement, which sounds obvious written down and is routinely ignored in practice by teams under budget pressure.

The AI slop problem is real, and it's a brand risk, not just a quality complaint

The visible failure mode across the industry right now is generic, interchangeable content â€” every company's LinkedIn post reading like it came from the same three prompts, because it largely did. The Content Marketing Institute's research on AI-assisted content has flagged the same pattern industry-wide: audiences are getting faster at pattern-matching AI-typical phrasing and disengaging from it within a sentence or two. HubSpot's own State of Marketing research has found a similar split â€” marketers report AI saves time on production, but differentiation and originality remain the top-cited concerns about over-relying on it. For a brand function specifically, that's not a productivity problem, it's a differentiation problem: if your content sounds like everyone else's, AI hasn't helped your brand, it's actively eroded the one thing brand work is supposed to build.

| Task | What changed | What still requires a human

| Long-form drafting | Blank-page time cut roughly 60â€“70% | Substantive edit, voice, judgement on what to cut

| Competitive research | Synthesis time cut significantly | Verification, interpretation, deciding what actually matters

| Positioning strategy | Marginal â€” good for stress-testing an existing idea | Original strategic direction; category insight

| Creative concepting | Faster volume of raw ideas | Judging which idea fits the specific brand and moment

Where this leaves brand teams in practice

The teams doing this well aren't the ones using AI the most â€” they're the ones using it for a narrowly scoped set of tasks and holding the line hard everywhere else. Draft acceleration, research synthesis, localisation across markets where a human reviewer still signs off â€” yes. Original positioning, brand strategy, anything that needs to sound distinctly like a specific company rather than a category-generic version of it â€” no, or at minimum, AI-assisted-then-heavily-rewritten rather than AI-generated. I wrote about the underlying visibility mechanics that make distinctive, fact-dense content more valuable now, not less, in the future of marketing and navigating the digital landscape.

Does this mean brand teams need fewer people?

Not in my experience so far â€” it's changed what the team's time goes toward, not the size of the team. The hours saved on first drafts and research synthesis have mostly been redirected into more original strategic work and more actual customer conversation, which was already the scarcest resource on most brand teams before any of this started. Headcount pressure in marketing right now is a budget story more than an AI-substitution story â€” LinkedIn's Economic Graph data on marketing job postings tracks that pressure moving in line with broader marketing budget cycles, not with AI adoption curves specifically.

What this means for you

Audit your own team's AI usage against the same split: draft speed and research synthesis are safe to hand over broadly; positioning, judgement calls, and anything meant to sound distinctly like your brand should stay firmly human, AI-assisted at most. The teams that blur this line are the ones producing the interchangeable content that's starting to actively hurt brand differentiation rather than help it â€” a risk I think is under-discussed relative to how much airtime "AI productivity" gets. I've made a related argument about protecting what's genuinely differentiated in data-driven marketing and unlocking insights.

Frequently asked questions

Has AI actually made brand marketing teams more productive?

Yes, specifically for first-draft content generation and competitive research synthesis, where blank-page and reading time drop substantially. It has not meaningfully accelerated positioning strategy or original creative judgement, which remain the bottleneck tasks.

What is "AI slop" and why does it matter for brand?

It's the industry term for generic, interchangeable AI-generated content that readers increasingly pattern-match and disengage from. For a brand function, it's a differentiation risk, not just a quality issue â€” if your content sounds like every competitor's, AI has quietly worked against the brand's core job.

Should brand teams stop using AI tools for creative work?

Not stop â€” but treat AI output as a starting point for ideation volume, never a finished creative concept. The judgement on which idea genuinely fits the brand and moment still needs to be human, and skipping that step is exactly what produces generic output.

Has AI reduced the need for brand marketing headcount?

In my experience, no â€” it's redirected time saved on drafting and research toward more original strategic work and customer conversation, rather than shrinking the team itself. Headcount pressure in marketing right now tracks budget cycles more than AI substitution.

What's the biggest mistake brand teams make with AI tools right now?

Using them for positioning and strategic direction rather than execution support. AI tools default toward generic, category-safe answers, which is close to the opposite of what distinctive positioning requires.

Where has AI actually changed your team's day-to-day work, versus where does it just sound like it should have? I'd like to compare notes with other marketers who've been in it for more than a quarter.

How Much Should a B2B Company Spend on Brand?

funpal10-byte — Tue, 01 Sep 2026 18:07:09 +0000

Most B2B companies land their brand budget somewhere between 5% and 10% of revenue, then defend that number with almost nothing â€” no benchmark, no logic, just what finance approved last year plus or minus a percent. The number that survives a board conversation isn't a guess. It's a range, backed by category context, and tied to a growth objective.

I've built brand budgets in freight, agricultural machinery, and automotive â€” three categories with completely different sales cycles and completely different tolerances for "why does marketing need this much money." The honest answer to "how much should we spend" is never a single figure. It's a formula with three or four real inputs, and most companies never write the formula down; they just inherit last year's number.

What the published benchmarks actually say

Gartner's 2026 CMO Spend Survey, covering 401 senior marketers at companies with over $1 billion in revenue, put average marketing budgets at 7.8% of company revenue in 2026 â€” and within that, brand-building spend (as opposed to demand generation and retention marketing) has historically run around a third of the total marketing budget in Gartner's category breakdowns, though the exact split moves with the economic cycle. That puts B2B brand spend, for a company at the Gartner average, somewhere near 2.5â€“3% of revenue â€” lower than most marketers assume, and lower than most B2B founders fear.

The IPA's long-running "Marketing in the Downturn" research, led by Peter Field and Les Binet, found something that should reframe this conversation entirely: brands that maintained or increased brand-building investment through a downturn came out of it with materially stronger market share than those that cut. Their broader 60:40 rule â€” roughly 60% brand building to 40% activation, for maximum long-run effectiveness â€” was built on B2C data but the B2B replication studies the IPA and LinkedIn's B2B Institute have published since land in a similar range: 46:54 brand-to-activation is the commonly cited B2B adaptation, closer to parity than the consumer split.

Why the "right" number depends on category maturity, not revenue alone

A percentage-of-revenue rule treats every company the same, and no two B2B categories behave the same way. A company entering a new market or launching a new category needs to spend well above category average on brand â€” you're building recognition and category entry points from zero, and demand generation has nothing to convert if nobody knows you exist. A company with 20+ years of category leadership can often spend below average and still hold position, because the compounding effect of a decade of consistent brand investment means each incremental rupee does less work than it did in year one.

The category's consideration-set length also matters. Categories with short lists (three to five vendors, per TrustRadius' 2024 buyer research) reward brand investment more than commoditised categories with ten-plus interchangeable vendors, where price and availability dominate the decision regardless of brand strength.

| Company situation | Reasonable brand spend as % of revenue | Why

| New entrant / new category | Above category average, often 8â€“12% of total marketing spend on brand-specific activity | No existing mental availability to compound; starting from zero

| Established mid-market player | Near category average, roughly 40â€“50% of total marketing budget on brand vs demand gen | Maintaining share of mind against active competitors

| Category leader, mature market | Below category average is often defensible | Compounding effect of years of consistent investment

| Commoditised, price-led category | Lower brand allocation, heavier demand-gen weighting | Consideration set is driven by price/availability more than brand

The mistake I've watched most often

Category maturity isn't a one-time assessment either. A company can move from "new entrant" to "established mid-market player" over three or four years without anyone formally revisiting the brand budget logic that was set at launch â€” the percentage just gets carried forward, quarter after quarter, until someone finally asks why spend hasn't moved even though the business, and the competitive set around it, clearly has.

The single most common budgeting mistake I've seen â€” including one I made early in my own career â€” is treating brand and demand generation as a zero-sum fight for the same pool of money, argued line-item by line-item every quarter. That structure guarantees brand loses, because demand-gen has a shorter, easier-to-prove attribution story even when its long-run effectiveness is weaker. Binet and Field's research is explicit on this: over-indexing on short-term activation produces short-term results that decay, while under-indexing on brand produces a category where competitors with stronger mental availability win the deals that were never going to be won on price or speed alone.

How do you defend a specific number to a CFO?

Don't lead with a percentage. Lead with the objective: are you defending share in a mature category, or building category entry points in a new one? Then attach the percentage as the mechanism, cite the category benchmark (Gartner, IPA/LinkedIn 46:54), and pair it with a leading indicator you'll report against â€” category entry point coverage, share of search, or branded search volume â€” rather than promising a revenue number brand spend structurally can't deliver on its own timeline. I've written more on the specific leading indicators that hold up in that conversation in why B2B brand awareness is the wrong metric to track.

What this means for you

If your brand budget is a percentage inherited from last year with no logic attached, that's the actual risk â€” not the number itself. Work out where your category sits (new entrant, established, leader, commoditised), benchmark against the Gartner and IPA/LinkedIn figures above, and build a one-page rationale you can defend in a five-minute conversation. It's the same discipline I've argued for in data-driven marketing and unlocking insights â€” a budget without a stated logic is the first thing cut, regardless of whether the number was actually right.

Frequently asked questions

What percentage of revenue should a B2B company spend on marketing overall?

Gartner's 2026 CMO Spend Survey puts the average at 7.8% of company revenue across large B2B and B2C companies. B2B-specific and smaller-company figures vary more widely, but this is the most current cross-industry benchmark available.

What's the difference between brand spend and demand generation spend?

Brand spend builds long-term mental availability and category recognition; demand generation converts existing intent into pipeline. The IPA/LinkedIn B2B research suggests a roughly 46:54 brand-to-activation split performs best for long-run B2B growth.

Should a startup spend more or less on brand than an established company?

Generally more, as a share of its marketing budget â€” a new entrant has no existing brand recognition to draw on, so demand generation has less to convert without simultaneous brand investment building awareness and trust.

Is it ever right to cut brand spend during a downturn?

The IPA's downturn research consistently finds companies that protect brand investment through a downturn emerge with stronger relative market share than those that cut it, because competitors who do cut create an opening for share gain at lower cost.

How do you know if your current brand budget is too low?

Track category entry point coverage and share of branded search over two to three quarters. If both are flat or declining while competitors are visibly investing, that's a stronger signal than any single percentage-of-revenue benchmark.

What number does your brand budget currently sit at, and could you defend the logic behind it in one slide? I'd like to hear how other B2B marketers are framing this conversation right now.

{"@context": "https://schema.org", "@type": "BlogPosting", "headline": "How Much Should a B2B Company Spend on Brand?", "description": "Most B2B brand budgets are inherited, not calculated. Here's the actual benchmark data and a framework for defending your number to a CFO.", "datePublished": "2026-07-30T09:30:00", "dateModified": "2026-08-05T18:13:32.620005+00:00", "author": {"@type": "Person", "@id": "https://www.rachitmishra.in/#rachit-mishra", "name": "Rachit Mishra", "url": "https://www.rachitmishra.in/profile/"}, "publisher": {"@type": "Person", "@id": "https://www.rachitmishra.in/#rachit-mishra", "name": "Rachit Mishra"}, "mainEntityOfPage": {"@type": "WebPage", "@id": "https://www.rachitmishra.in/b2b-brand-marketing-budget-percentage/"}, "articleSection": "Marketing Leadership", "keywords": "marketing-budget, marketing-roi, brand-positioning, b2b-marketing", "inLanguage": "en-IN"}
{"@context": "https://schema.org", "@type": "FAQPage", "mainEntity": [{"@type": "Question", "name": "What percentage of revenue should a B2B company spend on marketing overall?", "acceptedAnswer": {"@type": "Answer", "text": "Gartner's 2026 CMO Spend Survey puts the average at 7.8% of company revenue across large B2B and B2C companies. B2B-specific and smaller-company figures vary more widely, but this is the most current cross-industry benchmark available."}}, {"@type": "Question", "name": "What's the difference between brand spend and demand generation spend?", "acceptedAnswer": {"@type": "Answer", "text": "Brand spend builds long-term mental availability and category recognition; demand generation converts existing intent into pipeline. IPA/LinkedIn B2B research suggests a roughly 46:54 brand-to-activation split performs best for long-run B2B growth."}}, {"@type": "Question", "name": "Should a startup spend more or less on brand than an established company?", "acceptedAnswer": {"@type": "Answer", "text": "Generally more, as a share of its marketing budget \u00e2\u20ac\u201d a new entrant has no existing brand recognition to draw on, so demand generation has less to convert without simultaneous brand investment building awareness and trust."}}, {"@type": "Question", "name": "Is it ever right to cut brand spend during a downturn?", "acceptedAnswer": {"@type": "Answer", "text": "The IPA's downturn research consistently finds companies that protect brand investment through a downturn emerge with stronger relative market share than those that cut it, as competitors who do cut create room for share gain."}}, {"@type": "Question", "name": "How do you know if your current brand budget is too low?", "acceptedAnswer": {"@type": "Answer", "text": "Track category entry point coverage and share of branded search over two to three quarters. If both are flat or declining while competitors are visibly investing, that's a stronger signal than any single percentage-of-revenue benchmark."}}]}

Related reading: marketing in a manufacturing company.

The CX Score Trap: What NPS Never Tells You

funpal10-byte — Tue, 01 Sep 2026 18:06:59 +0000

A Net Promoter Score of 62 looks like a trophy on a slide. It told nobody in the room whether the account it came from was about to renew, expand, or quietly let the contract lapse at the next procurement cycle. NPS was never built to answer that question, and in B2B it mostly doesn't.

I sat through a QBR a few years ago where the CX team opened with a regional NPS of 58 â€” "world-class," someone said, and a few people nodded. Two months later we lost a mid-sized account that had scored a 9 on the same survey. The champion who filled it in had moved teams; the new procurement lead had never been asked a single question. The score was real. It was also measuring a person who no longer had a vote.

What NPS was actually built to measure

Fred Reichheld introduced the Net Promoter Score in a 2003 Harvard Business Review article, built on consumer research: one person, one relationship, one purchase decision, asked whether they'd recommend a company to a friend. That's a coherent question for a bank customer or a streaming subscriber. It is a much stranger question to ask a procurement manager who didn't choose your company, doesn't personally use your product, and whose job is explicitly not to be swayed by likability.

Why B2B breaks the model

Gartner's research on B2B buying groups puts the average number of stakeholders involved in a purchase decision at six to ten people, spanning procurement, end users, finance, and an executive sponsor who may never touch the product. NPS surveys typically reach one of those people â€” usually whoever's email is in the CRM, usually the friendliest one. You are not measuring the account. You are measuring your best relationship inside the account, and calling it representative.

The academic critique of NPS's predictive power isn't new. A 2007 paper by Timothy Keiningham and colleagues, published in the Journal of Marketing, tested the claim that NPS is the single best predictor of growth against a broader set of satisfaction metrics and found no evidence it outperformed them. That paper is nearly two decades old and the debate still hasn't settled â€” which itself tells you the "one number" claim was always a marketing line, not a measurement finding.

What the score misses in a renewal-driven business

In a B2B relationship, the commercial outcome you actually care about is renewal, expansion, or churn â€” and none of those live on a single detractor-to-promoter scale filled in by one contact. Forrester's CX Index research has consistently found that customer experience quality correlates with revenue growth at the account level, tracked over time, not at the level of a single quarterly survey score. Qualtrics' XM Institute has published similar findings: the organisations that outperform on customer-led growth are the ones tracking outcome metrics tied to the account, not sentiment metrics tied to an individual.

Zendesk's CX Trends research adds a B2B-specific data point worth citing directly: a majority of business buyers now say experience is as important as the product itself in their purchase and renewal decisions â€” which makes it even more important that whatever metric represents "experience" on a dashboard actually reflects the account, not one respondent's mood on a Tuesday.

Here's the honest problem with NPS in B2B: it is cheap to collect, easy to put on a slide, and comparable across quarters, so it survives long after people stop trusting it. I've watched three separate organisations keep reporting NPS at the board level specifically because the alternative â€” building a proper multi-stakeholder health score â€” takes real work, and nobody wanted to own that project.

| Metric | What it captures | What it misses in B2B

| NPS | Sentiment of one respondent, one moment | Whether the whole buying group is aligned; whether the account is commercially healthy

| CSAT | Satisfaction with a specific interaction | Long-term account trajectory; only measures a moment, not a relationship

| Net Revenue Retention | Actual account-level commercial outcome | Lagging â€” tells you what already happened, not why

| Executive sponsor engagement | Whether the account has a champion above the day-to-day contact | Harder to quantify; requires CRM discipline most teams don't have

What to track instead

None of this means throw out satisfaction data â€” it means stop treating one score as if it were a commercial forecast. A more honest account-health model combines: net revenue retention (the actual number), expansion or contraction trend over the last two renewal cycles, breadth of stakeholder engagement (how many distinct people in the account have you had a real conversation with in 90 days), and a qualitative signal from the executive sponsor specifically, not just whoever answers the survey. This is close to the account-health scoring that customer success teams at PLG software companies have been building for years â€” B2B services and manufacturing businesses have mostly not caught up.

I wrote about a related measurement trap in why B2B brand awareness is the wrong metric to track â€” the pattern repeats: a metric built for consumer, single-decision-maker contexts gets imported into B2B wholesale, survives because it's simple to report, and quietly stops meaning what everyone assumes it means.

How do you retire NPS without the room feeling like you're hiding bad news?

This is the real objection, and it's fair â€” NPS has boardroom credibility precisely because everyone already understands it. The honest approach isn't to kill it overnight; it's to keep reporting it for context while introducing the account-health metrics alongside it, and let the gap between the two tell the story. When NPS says 58 and net revenue retention says 91%, that gap is more informative than either number alone, and it's a much harder story to wave away with "we're doing fine."

What this means for you

If NPS is the headline metric in your CX reporting, don't remove it â€” recontextualise it. Add net revenue retention, stakeholder breadth, and executive sponsor engagement as companion metrics for every account, not just the ones flagged at risk. I've argued for a similar shift on the data side in data-driven marketing and unlocking insights â€” the organisations that win aren't the ones with the fanciest single number, they're the ones willing to track something less flattering and more true.

Frequently asked questions

Is NPS completely useless in B2B?

No â€” it's a reasonable pulse-check on individual relationships and it's cheap to run consistently. The problem is treating it as a proxy for account health or a predictor of renewal, which the underlying research doesn't support in multi-stakeholder buying contexts.

What should replace NPS as the primary CX metric in B2B?

There isn't a single replacement â€” that's the point. Net revenue retention, stakeholder breadth, and executive sponsor sentiment together give a more honest picture than any single score, including NPS itself.

Why do B2B companies keep using NPS if it's flawed?

It's simple to collect, easy to benchmark against competitors, and boards already understand it. Building a proper multi-stakeholder account-health model takes real operational work that most CX teams aren't resourced to do.

How many stakeholders are typically involved in a B2B buying decision?

Gartner's research puts the average buying group at six to ten people spanning procurement, end users, finance and an executive sponsor â€” which is exactly why a single-respondent survey score struggles to represent the account.

Does a high NPS mean an account is safe from churn?

Not reliably. A high score from one satisfied day-to-day contact says nothing about whether the executive sponsor is engaged or whether procurement is quietly evaluating alternatives â€” which is where most enterprise churn actually starts.

If your CX dashboard leads with NPS, what's the metric sitting quietly next to it that actually predicted your last churn? I'd like to know what other teams have found works.

{"@context": "https://schema.org", "@type": "BlogPosting", "headline": "The CX Score Trap: What NPS Never Tells You", "description": "NPS predicts almost nothing about B2B account health on its own. Here's what actually correlates with renewal and expansion, and how to track it.", "datePublished": "2026-07-27T09:30:00", "dateModified": "2026-08-05T18:13:08.924159+00:00", "author": {"@type": "Person", "@id": "https://www.rachitmishra.in/#rachit-mishra", "name": "Rachit Mishra", "url": "https://www.rachitmishra.in/profile/"}, "publisher": {"@type": "Person", "@id": "https://www.rachitmishra.in/#rachit-mishra", "name": "Rachit Mishra"}, "mainEntityOfPage": {"@type": "WebPage", "@id": "https://www.rachitmishra.in/nps-limitations-b2b/"}, "articleSection": "Customer Experience", "keywords": "cx-strategy, customer-journey, customer-loyalty, brand-measurement", "inLanguage": "en-IN"}
{"@context": "https://schema.org", "@type": "FAQPage", "mainEntity": [{"@type": "Question", "name": "Is NPS completely useless in B2B?", "acceptedAnswer": {"@type": "Answer", "text": "No \u00e2\u20ac\u201d it's a reasonable pulse-check on individual relationships and it's cheap to run consistently. The problem is treating it as a proxy for account health or a predictor of renewal, which the underlying research doesn't support in multi-stakeholder buying contexts."}}, {"@type": "Question", "name": "What should replace NPS as the primary CX metric in B2B?", "acceptedAnswer": {"@type": "Answer", "text": "There isn't a single replacement \u00e2\u20ac\u201d that's the point. Net revenue retention, stakeholder breadth, and executive sponsor sentiment together give a more honest picture than any single score, including NPS itself."}}, {"@type": "Question", "name": "Why do B2B companies keep using NPS if it's flawed?", "acceptedAnswer": {"@type": "Answer", "text": "It's simple to collect, easy to benchmark against competitors, and boards already understand it. Building a proper multi-stakeholder account-health model takes real operational work most CX teams aren't resourced to do."}}, {"@type": "Question", "name": "How many stakeholders are typically involved in a B2B buying decision?", "acceptedAnswer": {"@type": "Answer", "text": "Gartner's research puts the average buying group at six to ten people spanning procurement, end users, finance and an executive sponsor \u00e2\u20ac\u201d exactly why a single-respondent survey score struggles to represent the account."}}, {"@type": "Question", "name": "Does a high NPS mean an account is safe from churn?", "acceptedAnswer": {"@type": "Answer", "text": "Not reliably. A high score from one satisfied day-to-day contact says nothing about whether the executive sponsor is engaged or whether procurement is quietly evaluating alternatives \u00e2\u20ac\u201d where most enterprise churn actually starts."}}]}

When to Rebrand a Company: The Case for Not Doing It

funpal10-byte — Sat, 29 Aug 2026 12:27:08 +0000

Rebrand when the name is legally, structurally or reputationally unusable â€” a joint venture unwinds, a merger creates two competing identities, or the name carries damage you cannot outrun. Everything else is a refresh. Changing a working name resets recognition you have already paid for, and most companies underestimate how long it takes to earn back.

I've run brand and communication functions for fifteen years across India, APAC and MEA, in categories where nobody buys because the logo looks modern. Steel, freight, tractors, printers. I've been in the room for a handful of serious rebrand proposals. The ones with a structural trigger were straightforward. The ones that opened with a slide reading "our identity feels dated" were the expensive ones â€” and the hardest to argue against, because nobody could prove the current brand was working either.

Brand equity is invisible until you switch it off. That's the whole problem.

What a rebrand actually costs, beyond the design fee

The agency invoice is the smallest line. The real bill is the recognition you delete on the day you launch.

Bain and Google surveyed 1,208 people involved in B2B purchasing across software, hosting, hardware, telecom, logistics and industrial equipment. Writing up the research in Harvard Business Review in 2022, they reported that 90% of buyers ultimately choose a vendor that was already on their shortlist on day one. Not the vendor with the best proposal. The vendor they could already name before the process started.

Now put that next to the 95:5 rule. Professor John Dawes at the Ehrenberg-Bass Institute established that roughly 5% of B2B buyers are in-market at any moment, and 95% won't buy for months or years. The memory structure you're building today gets used by someone in 2028.

A name change hits both mechanisms at once. It removes you from day-one lists that took a decade to get onto, and it invalidates the memory you've been depositing with the 95% who weren't listening closely enough to notice the announcement. This is the same reason tracking brand awareness as a headline metric misleads people â€” awareness surveys measure whether people recognise a name when prompted, not whether they can retrieve it unprompted when a purchase actually starts.

Consumer categories give the cleanest evidence because sales data arrives fast. Tropicana redesigned its Pure Premium packaging in January 2009. According to a case study archived at AgEcon Search, University of Minnesota, sales fell 20% year-on-year between January and February, at an estimated cost of $27 million. The old packaging was back by the end of February. Nothing about the juice changed. Shoppers simply stopped finding it.

Corporate rebrands correct more slowly because the damage is harder to see. Royal Mail became Consignia in January 2001; Campaign reported the group abandoned the name on 22 May 2002. Sixteen months, a full rollout, a reversal.

A rebrand is the only marketing project that begins by destroying an asset you already own and spent years paying for.

Why does every new leader want a new logo?

Because it's the fastest visible evidence that something is happening.

A new CMO or CEO arrives with eighteen months to demonstrate change. Fixing distribution takes three years. Fixing product quality takes longer. A new identity ships in six months, appears on every truck and reception desk, and photographs well in the annual report. It's change you can point at.

There's a second, quieter driver: internal fatigue. Marketing teams look at their own logo every day for years. Customers see it for four seconds a quarter and feel nothing of the sort. I've written before about marketing inside a product-led manufacturing business, and the gap is sharper there than anywhere: the people who own the brand are the only ones over-exposed to it.

Kantar's 2025 analysis of its BrandZ rankings found that brands with strong Meaningful Difference were twice as likely to still be in the rankings twenty years later, and that the average brand in its Global Luxury Top 10 carries a 132-year history. Longevity is not a design problem. It's an accumulation problem, and accumulation only works if you stop resetting the counter.

The four reasons that justify a rebrand, and the four that don't

Here's the test I use before a rebrand gets budget. If the trigger isn't in the left column, it's a refresh, not a rebrand.

|
Justified â€” change the name | Not justified â€” refresh instead

| **Legal or structural force.** A JV dissolves, a licence expires, a demerger splits the entity, or you lose the right to the name. | **The identity feels dated.** Typography, palette and photography can all be modernised without touching the name. 

| **The name misdescribes the business.** You are now materially something else, and the name actively blocks buyers from considering you for what you sell. | **New leadership wants a marker.** Legitimate need, wrong instrument. Pick something customers can feel. 

| **Reputational damage you cannot outrun.** The name is permanently attached to a failure, and remediation has already happened. | **A competitor rebranded.** Their trigger is not your trigger. 

| **Merger of equals with two live identities.** Two names competing in the same category cost more to maintain than one new one. | **Internal fatigue.** You are bored of it. Your buyers have barely registered it.

Most proposals I've seen sit squarely in the right-hand column while being argued with left-hand-column urgency. One question settles it: what specifically becomes possible after the change that is impossible today?

What do Jaguar's numbers actually prove?

Less than the internet claims, and the honest version is more useful.

Jaguar registered 49 new cars across the EU, EFTA and the UK in April 2025, against 1,961 in April 2024 â€” a 97.5% fall, in a market that grew 1.3% year-on-year, per ACEA registration data. That figure has been passed around as proof that a controversial rebrand destroyed a car company.

It isn't. Jaguar had deliberately stopped selling its existing petrol range in Europe ahead of an all-electric relaunch. With almost nothing to sell, registrations were always going to collapse. Anyone citing that number as rebrand damage is reading a production decision as a marketing one.

The real lesson sits elsewhere. Jaguar changed identity, product strategy and price position at once, so nobody â€” including Jaguar â€” can isolate which decision moved which number. That's the underrated risk of a large rebrand: not that it fails, but that it makes attribution impossible for years, so you never learn anything reusable.

The X case is cleaner, because only the name changed. Brand Finance valued Twitter at USD 5.7 billion in January 2022; by its September 2024 assessment, X was worth USD 673.3 million. Other factors were in play, but abandoning a name that had become a verb was a decision taken freely, and there was no structural trigger requiring it.

Which brands got this right, and what did they actually change?

The good examples share a pattern: they changed the layer that was broken and left the layer that was working alone.

Eternal. In February 2025, Zomato renamed its parent company Eternal to reflect a portfolio that now includes Blinkit, Hyperpure and District. The consumer app stayed Zomato. Customers experienced no change at all. The rename solved a real problem â€” a group whose largest value driver was no longer food delivery being read entirely as a food delivery company â€” without spending a rupee of consumer recognition. That is a brand architecture decision, not a rebrand, and the distinction is worth defending in a boardroom.

Hero. When the Hero Honda joint venture ended, there was no choice. Forbes India reported that royalties to Honda had reached 3.68% of sales, around â‚¹500 crore a year by 2010, and that the split was decided in December 2010 with the new brand launched in August 2011. Hero MotoCorp's two-wheeler share moved from 45% in 2010-11 to 39% by 2015-16, while Honda's own India arm went from 13% to 26%. A necessary rebrand, executed well, still cost six points of share. That's the price even when you're right.

Air India. The identity change unveiled on 10 August 2023 replaced a decades-old logo with "The Vista" but kept the name. It arrived alongside a $400 million cabin refit of 43 widebody aircraft and orders for 470 new planes. The visual work was the smallest part of what was being communicated.

Notice what none of these did: change a well-known consumer name because it felt tired.

What this means for you

If a rebrand is on your table this year, work through this before the agency briefing:

Name the trigger in one sentence, without adjectives. If the sentence needs the words "modern", "fresh" or "aligned", you don't have a trigger.
Separate the four layers. Name, logo and distinctive assets, visual system, and messaging. Most problems live in layers three and four. Only layer one is a rebrand.
Audit your distinctive assets first. Which colour, shape, sound or phrase do buyers already attribute to you? Anything with high recognition is an asset on your balance sheet that no design review should casually delete.
Price the recovery, not the rollout. Budget the media weight needed to rebuild recognition to its current level, and put that number in the same paper as the design cost. This single line kills more unnecessary rebrands than any argument I've ever made.
Set a measurement window before launch. Baseline unprompted recall, inbound enquiry volume and share of search now. Without a pre-read you'll be arguing from opinion in eighteen months.
Change one thing at a time. If product, pricing and identity all move together, you lose the ability to learn from any of it.

And if the honest answer is that the brand is fine but the business isn't growing, say so out loud. A new logo has never fixed a distribution gap.

Frequently asked questions

When should a company rebrand?

Rebrand when there's a structural trigger: a joint venture or licence ends, a merger leaves two competing identities, the name misdescribes what you now sell, or reputational damage is permanently attached to it. Aesthetic fatigue, new leadership and competitor activity are not triggers. Those situations call for a visual refresh instead.

How much does a rebrand really cost?

The design and rollout invoice is usually the smaller half. The larger cost is rebuilding recognition you already had, which takes sustained media spend over several years. Tropicana's 2009 packaging change alone was estimated at $27 million in lost sales within two months, before any recovery spending.

What's the difference between a rebrand and a brand refresh?

A rebrand changes the name and therefore the memory structure buyers use to retrieve you. A refresh updates typography, colour, photography, tone and messaging while keeping the name and the distinctive assets intact. Most companies that believe they need a rebrand actually need a refresh, which costs far less and risks nothing.

Does changing a company name hurt SEO and search visibility?

Yes, temporarily and sometimes permanently. Branded search volume for the old name doesn't transfer automatically, and links, citations and third-party mentions accumulated over years point to the previous entity. Proper 301 redirects and consistent structured data reduce the loss, but expect a dip in branded queries lasting several quarters.

Can you rebrand the parent company without changing the customer-facing brand?

Yes, and it's often the correct answer. Zomato renamed its parent entity Eternal in February 2025 while leaving the consumer app untouched, solving a corporate identity problem without spending customer recognition. This works whenever the issue is how investors and analysts read the group rather than how buyers choose.

The uncomfortable version

Most rebrands are a legitimate need answered with the wrong instrument. Leadership wants proof of momentum, and identity is the only lever marketing controls outright, so identity gets pulled. The companies that resist have one thing in common: marketing has enough standing to say the awkward thing out loud â€” the brand isn't the problem.

What's the strongest rebrand case you've had to argue against â€” and did the number for rebuilding recognition ever make it into the paper? I'd like to hear how that conversation went. Find me on LinkedIn.

Generative Engine Optimization: How Brands Actually Get Cited

funpal10-byte — Sat, 29 Aug 2026 12:15:23 +0000

Generative engine optimization is the work of getting your brand named and cited inside AI answers rather than ranked inside a list of blue links. The evidence so far says it is won less by rewriting pages and more by being mentioned, consistently, across the sources these engines already read.

Earlier this year I spent an afternoon doing something I would recommend to any brand head: I opened four AI assistants, typed in the fifteen questions a serious buyer in my category would actually ask, and wrote down every company name that came back. It took under an hour. The same handful of names kept appearing, and the ranking bore almost no relationship to who is actually largest by revenue. Several companies I would have called market leaders never surfaced once.

That is the part worth sitting with. A discovery layer now exists that a lot of well-run brands are simply absent from, and most of them have not checked.

Why does generative engine optimization matter now rather than later?

Because the traffic maths has already changed, and the numbers are not subtle.

Pew Research Center tracked 68,879 Google searches from 900 US adults in March 2025. On visits where an AI summary appeared, users clicked a traditional search result 8% of the time. Where no summary appeared, they clicked 15% of the time â€” nearly twice as often. Clicks on the sources cited inside the summary itself happened on 1% of visits.

Ahrefs studied 300,000 keywords and found position-one click-through rates fell 34.5% for terms where an AI Overview appeared, comparing March 2024 with March 2025. It also found that 99.2% of the keywords triggering AI Overviews are informational in intent â€” which is to say, exactly the queries that thought leadership content is written to win.

And the audience is not a Western phenomenon. Sam Altman confirmed in February 2026 that India has 100 million weekly active ChatGPT users, making it OpenAI's second-largest market after the United States.

The traffic that does arrive this way behaves better. Semrush research published in June 2025 put the average AI-search visitor at 4.4 times the value of an average organic search visitor. Adobe's April 2026 analysis of US retail found AI-sourced traffic up 393% year over year in Q1 2026, converting 42% better, staying 48% longer, and viewing 13% more pages than non-AI traffic.

Fewer visitors, better visitors. That is a trade most B2B marketers would take.

What do AI engines actually cite?

Not what most guides assume. Semrush analysed 230,000 prompts and over 100 million citations across ChatGPT, Google AI Mode and Perplexity over thirteen weeks between July and October 2025. Reddit and Wikipedia dominated overall â€” but the composition moved violently. On ChatGPT specifically, Reddit's citation frequency fell from roughly 60% to around 10% in mid-September, and Wikipedia from about 55% to under 20%, with LinkedIn and Forbes picking up the slack. Google's AI Mode showed a different mix entirely, with LinkedIn at around 15% and Wikipedia appearing in roughly 2% of responses.

Two things follow from that. First, the citation surface is not one surface; it is three or four with different tastes. Second, it is unstable enough that any tactic tuned to last quarter's mix is depreciating while you implement it.

Then there is the finding that reframed this whole thing for me. Ahrefs ran a correlation study across 75,000 brands to see which signals track with being mentioned in AI Overviews. Branded web mentions came out strongest at 0.664. Branded search volume followed at 0.392, Domain Rating at 0.326, and the number of backlinks trailed at 0.218. About 26% of the brands studied had no AI Overview mentions at all. The authors are careful to note these are correlations, not causes, and that most sit in the moderate-to-weak band.

Still, the ordering is instructive. The thing SEO has spent twenty years accumulating â€” links â€” correlates least. The thing brand marketing has always been about â€” being talked about by name â€” correlates most.

Generative engine optimization is not a new discipline. It is brand building, measured for the first time by a machine that reads.

The Citation Ladder: four rungs from invisible to cited

I use a four-rung model when I plan this work, because it forces an honest answer to the question of which rung you are actually stuck on. Most teams are not stuck where they think.

|
Rung | The question it answers | What it takes | How long

| **1. Resolvable** | Can the engine tell who you are, distinct from everyone with a similar name? | Person and Organisation schema with a stable identifier, one bio used verbatim everywhere, a complete set of linked profiles | Days 

| **2. Readable** | Can a passage be lifted from your page and used as an answer? | Answer-first openings, self-contained sections, tables, dated facts, FAQ markup | Weeks 

| **3. Referenced** | Are you named on the sources these engines already read? | Trade press, LinkedIn, community forums, podcasts, industry directories, analyst mentions | Months 

| **4. Repeated** | Does the mention recur across many sources over time? | Consistent naming, sustained publishing cadence, repeated third-party coverage | Quarters

Rung one is cheap and almost everyone skips it. If an AI system cannot resolve your brand or your executives to a single entity with defined expertise, everything above it leaks. In my own case there are at least three other people online who share my name, in entirely different professions, and until the schema and profiles said otherwise the engines had no reason to prefer one of us.

Rung two is where nearly all published generative engine optimization advice lives, and it is genuinely worth doing. The original GEO paper from Aggarwal and colleagues at Princeton, presented at KDD 2024, showed that black-box content changes â€” adding citations, quotations and statistics â€” lifted visibility in generative engine responses by up to 40%, while noting the effect varies by domain. Adobe's readability scoring found retail product pages averaging only 66% machine-readability against a homepage average of 75%, so there is real, unclaimed ground here.

But rung two is a page-level advantage, and page-level advantages are the ones that get arbitraged away first. Everyone reads the same guides.

Rung three is where the difficulty, and therefore the durability, lives. Being named in trade press, on LinkedIn, in a community thread, on a podcast transcript, in an analyst note â€” those are the branded web mentions that topped the Ahrefs correlation table. They are also the hardest to fake, which is the point.

Rung four is simply rung three sustained long enough to survive a model update. Given how fast the ChatGPT citation mix moved in a single quarter, a brand that appears in one source once is a rounding error. A brand that appears across many sources repeatedly is a fact about the category.

Why is most GEO advice aimed at the wrong rung?

Because rung two is the only one an agency can invoice for in ninety days. Schema, headings, FAQ blocks and answer-first paragraphs are legible, auditable and fast. Getting your company named in eleven credible places over six months is slower, less attributable and harder to sell.

It is the same asymmetry that has always pushed marketing budgets toward the measurable half of the job. I have argued before that B2B brand awareness is the wrong metric to track, and this is that argument arriving from a new direction: the machines are now rewarding mental availability, they are just calling it citation frequency.

There is also a quieter cost to over-indexing on rung two. If every brand in a category publishes structurally identical, answer-first, statistics-dense content, the format stops being a differentiator and the engine falls back on which name it has seen most. Which returns you to rung three.

What this means for you

A practical sequence, in the order I would run it:

Measure before you optimise. Write down fifteen questions a buyer in your category would ask an assistant. Run them across ChatGPT, Gemini, Perplexity and Claude once a month. Log which brands appear and whether you are one of them. This costs twenty minutes and it is the only baseline you will get, since there is no Search Console for AI citations.
Fix entity resolution first. Organisation and Person schema, one bio used verbatim across every profile, consistent job titles, and every owned profile linking back to your site. This is a one-week job that most teams have never done.
Make the top twenty pages extractable. First 40 to 60 words answer the question in the title. Sections that make sense pulled out on their own. One table per page. A dated statistic every few hundred words. FAQ markup.
Budget for rung three explicitly. Two or three bylined pieces in genuine trade publications a year, a consistent LinkedIn presence from named executives rather than the logo, and honest participation in the communities where your buyers already argue. Treat it as brand spend, because it is â€” the same conversation as how much a B2B company should spend on brand.
Hold the cadence. Recency is a filter, not a tiebreaker. Publishing quarterly and expecting citations is like running one flight of advertising and expecting salience.

None of this replaces the rest of the job. The AI layer sits on top of a buying process that still runs through people, and if the customer journey underneath it is broken, being cited more often just delivers better-informed prospects to a worse experience. It is also worth being clear-eyed about what has and has not changed, which is more or less what I argued about AI in brand management more broadly.

Frequently asked questions

What is generative engine optimization?

Generative engine optimization is the practice of increasing how often a brand is named and cited inside AI-generated answers from systems like ChatGPT, Perplexity, Gemini and Google AI Overviews. It combines entity clarity, machine-readable page structure and third-party mentions, rather than the ranking-position focus of traditional search optimisation.

How is GEO different from SEO?

SEO competes for a position in a ranked list of links. GEO competes to be the source an answer is built from, where there is no list and often no click. SEO rewards links and rankings; the available evidence suggests generative engines lean more heavily on branded mentions, entity clarity and extractable passages.

Does generative engine optimization actually work?

The Princeton GEO study presented at KDD 2024 found that content changes such as adding citations, quotations and statistics raised visibility in generative engine responses by up to 40%, with effects varying by domain. That is a page-level gain. Broader visibility appears to track branded mentions across third-party sources more strongly.

How do you measure AI citations without Search Console?

Run a fixed set of buyer questions across the major assistants monthly and log whether your brand appears and whether your site is cited. Check server logs for crawlers such as GPTBot, PerplexityBot and ClaudeBot. Paid tools from Ahrefs and Semrush automate this, but a spreadsheet establishes the trend line.

Should B2B companies block AI crawlers?

For most B2B brands, no. Blocking crawlers removes you from the retrieval layer entirely, which forfeits the citations rather than protecting anything. Publishers with paywalled inventory face a genuine trade-off. A company whose content exists to build category authority does not.

The part I am still unsure about

Whether any of this stays true for more than a year. The citation mix on one platform moved by fifty percentage points in a single quarter, and the correlations we are all quoting are moderate at best. The four rungs feel durable to me because they describe how reputation has always worked, not how one model currently retrieves. But I would rather be told I am wrong now than discover it in 2027.

So a question for anyone running a brand function right now: have you actually run the query set for your own category, and did your company come back? I would like to know how many people find what I found. Tell me on LinkedIn.

Generative Engine Optimization: How Brands Actually Get Cited

funpal10-byte — Mon, 24 Aug 2026 05:21:56 +0000

That is the part worth sitting with. A discovery layer now exists that a lot of well-run brands are simply absent from, and most of them have not checked.

Why does generative engine optimization matter now rather than later?

Because the traffic maths has already changed, and the numbers are not subtle.

Fewer visitors, better visitors. That is a trade most B2B marketers would take.

What do AI engines actually cite?

Generative engine optimization is not a new discipline. It is brand building, measured for the first time by a machine that reads.

The Citation Ladder: four rungs from invisible to cited

I use a four-rung model when I plan this work, because it forces an honest answer to the question of which rung you are actually stuck on. Most teams are not stuck where they think.

|
Rung | The question it answers | What it takes | How long

| **1. Resolvable** | Can the engine tell who you are, distinct from everyone with a similar name? | Person and Organisation schema with a stable identifier, one bio used verbatim everywhere, a complete set of linked profiles | Days 

| **2. Readable** | Can a passage be lifted from your page and used as an answer? | Answer-first openings, self-contained sections, tables, dated facts, FAQ markup | Weeks 

| **3. Referenced** | Are you named on the sources these engines already read? | Trade press, LinkedIn, community forums, podcasts, industry directories, analyst mentions | Months 

| **4. Repeated** | Does the mention recur across many sources over time? | Consistent naming, sustained publishing cadence, repeated third-party coverage | Quarters

But rung two is a page-level advantage, and page-level advantages are the ones that get arbitraged away first. Everyone reads the same guides.

Why is most GEO advice aimed at the wrong rung?

What this means for you

A practical sequence, in the order I would run it:

Measure before you optimise. Write down fifteen questions a buyer in your category would ask an assistant. Run them across ChatGPT, Gemini, Perplexity and Claude once a month. Log which brands appear and whether you are one of them. This costs twenty minutes and it is the only baseline you will get, since there is no Search Console for AI citations.
Fix entity resolution first. Organisation and Person schema, one bio used verbatim across every profile, consistent job titles, and every owned profile linking back to your site. This is a one-week job that most teams have never done.
Make the top twenty pages extractable. First 40 to 60 words answer the question in the title. Sections that make sense pulled out on their own. One table per page. A dated statistic every few hundred words. FAQ markup.
Budget for rung three explicitly. Two or three bylined pieces in genuine trade publications a year, a consistent LinkedIn presence from named executives rather than the logo, and honest participation in the communities where your buyers already argue. Treat it as brand spend, because it is â€” the same conversation as how much a B2B company should spend on brand.
Hold the cadence. Recency is a filter, not a tiebreaker. Publishing quarterly and expecting citations is like running one flight of advertising and expecting salience.

Frequently asked questions

What is generative engine optimization?

How is GEO different from SEO?

Does generative engine optimization actually work?

How do you measure AI citations without Search Console?

Should B2B companies block AI crawlers?

The part I am still unsure about

Customer Journey Mapping That Survives Contact With Reality

funpal10-byte — Wed, 19 Aug 2026 07:59:00 +0000

A B2B customer journey map survives contact with reality when it is built on three things: the real buying group rather than a single persona, evidence from your own systems rather than a workshop whiteboard, and a named owner for every stage. Most maps fail because they have none of the three.

I once commissioned a journey mapping exercise across four Asian markets. Three days of workshops, an external facilitator, sticky notes in five colours, and an A0 poster at the end of it. Eighteen months later I found that poster rolled up behind a filing cabinet. Nobody had opened it since the week it was printed.

The thinking in it wasn't wrong. It was unowned, undated and unfalsifiable, and those three properties reliably get a document ignored inside a business with quarterly targets. This is a guide to doing it the other way.

Why do most B2B journey maps fail?

Because they map a person, and B2B buying is done by a committee. Forrester's 2026 buyer research puts the typical business buying decision at 13 internal stakeholders plus nine external influencers. Its 2024 edition found 89% of purchases involve two or more departments. A map with one smiling persona at the top cannot explain the thing a map exists to explain: why deals stop.

And they do stop. That same 2024 Forrester study found 86% of B2B purchases stall at some point in the process, and 81% of buyers were dissatisfied with the provider they eventually chose. Procurement professionals are decision-makers in 53% of business buying cycles, according to the 2026 study. If your journey map has no procurement lane, it is missing the single most common reason your deal is sitting still.

The second failure is quieter. A journey map presumes the customer is on a journey. Research by Professor John Dawes at the Ehrenberg-Bass Institute, conducted with the LinkedIn B2B Institute, established that roughly 95% of business buyers are not in the market for a given category at any moment. Your map describes the 5%. Most of your budget is spent reaching the rest, which is a separate job and one I've argued about before in why brand awareness is the wrong metric to track.

Start with the buying group, not the buyer

Gartner's B2B buying research is blunt about how little of this involves you. Three-quarters of B2B buyers say they prefer a rep-free sales experience. Ninety-nine percent of B2B purchases are triggered by an organisational change rather than by a campaign. And buyers who use supplier-provided digital tools alongside a sales rep are 1.8 times more likely to complete a high-quality deal, which tells you the answer is not to remove the rep but to arm the buyer.

McKinsey's B2B Pulse work, published in April 2023, found buyers now move across about ten channels during a purchase, double the five they used in 2016, and that around 70% of decision makers are willing to spend up to $500,000 through a self-serve digital channel.

So the working unit of a B2B journey map is not "the customer". It is a role, at a stage, on a channel. Three coordinates, not one. The moment you write it that way, the map stops being a poster and starts being a grid you can populate with evidence.

Five checks that decide whether a map survives

I now run every journey map, mine or an agency's, through the same five checks before anyone is allowed to present it. A map that fails two of them goes back.

| Check | The question it answers | What fails it

| Ownership | Whose quarterly objective moves if this stage improves? | Stages owned by "the business" or by a committee

| Evidence | Which system did this stage's timing come from? | Durations estimated in a workshop and never checked

| Falsifiability | What would we observe if this map were wrong? | Maps written so generally that no data could contradict them

| Group coverage | Does every role in the buying group appear, including procurement and the sceptic? | One persona, usually the friendliest one

| Expiry | What is the review date, and who called it? | No date, which in practice means never

The expiry check is the one people resist and the one that matters most. A journey map that nobody has to update is not a management tool. It is a poster.

What data should a B2B journey map be built on?

Five sources, in descending order of how much I trust them.

CRM stage-change timestamps. Not the pipeline value, the timestamps. The gap between "proposal sent" and "commercial review" is the truest thing your CRM knows about your journey.
Win/loss interviews, done by someone who did not own the deal. Fifteen conversations will teach you more than any survey panel.
Service and support tickets from the first 90 days post-sale. Onboarding failures are journey failures that happen after the map usually ends.
Self-serve behaviour. Pricing page visits, spec-sheet downloads, quote-tool abandonment. Since most buyers now prefer to do this unaccompanied, it is often the only visible evidence of an active deal.
Tender and procurement documents. Tedious, and the most honest statement of evaluation criteria you will ever get.

Notice what is absent from that list. Relationship survey scores sit near the bottom for journey work, for reasons I set out in what NPS never tells you about a B2B account. A score tells you the temperature. It does not tell you which stage caused it.

The one-page format that replaced my poster

Columns are buying-group roles: economic buyer, technical evaluator, operational user, procurement, and the internal sceptic every deal has. Rows are stages, in your language, not a textbook's. Each cell holds three lines and no more: what that role needs at that stage, who owns supplying it, and the evidence that the need is real.

That constraint is deliberate. Anything that doesn't fit on one page doesn't get read in a review meeting, and a journey map that isn't reviewed in a meeting has no mechanism to change anything.

The gap this format exposes fastest is coordination. Salesforce's sixth State of the Connected Customer study, covering 3,300 business buyers in 2023, found 73% say most sales interactions feel transactional and 55% feel they are engaging with separate departments rather than one company. Those are not messaging problems. They are handover problems, and handovers only become visible when you draw the map by role.

Does journey mapping still work at all?

It works, but the practice has a poor record and it is worth being honest about that. Gartner reported in July 2018 that almost one-third of organisations still had difficulty incorporating journey maps into their CX efforts at all. Eight years on, I don't see that number improving in the rooms I sit in.

Outcomes suggest the same. In Forrester's 2025 Global CX Index, built on more than 275,000 customer perceptions across 469 brands in 13 countries, 21% of brands declined and only 6% improved. In Asia Pacific, 37% of brands' scores fell. An enormous amount of journey mapping happened during those years. Very little of it moved a score.

My reading is that mapping is a diagnostic, and organisations keep treating it as an intervention. Drawing the journey is the cheap part. Assigning the stage to somebody with a target attached is the part that changes anything.

What this means for you

If you are about to commission a journey mapping exercise, do four things differently.

Budget for the second version. The first map is a hypothesis. Book the review before you book the workshop.
Put a name in every stage cell. Not a function. A person who will be in the room at the next quarterly review.
Map the loss, not the win. Build the first version from ten deals you lost and one you won. The failure path is where the information is.
Keep it in the operating rhythm. A map reviewed monthly alongside the pipeline stays alive. A map reviewed annually is a poster with a date on it.

The industry context matters here too. India's logistics sector employs over 22 million people and the country is working to bring logistics costs below 10% of GDP by 2030, according to a Press Information Bureau paper from August 2025. In sectors moving that fast, a journey map built in March is describing a different business by September. That is an argument for shorter maps reviewed more often, not longer ones reviewed less. I've written about the related problem of selling brand work inside engineering-led firms in marketing inside a manufacturing company, and about a CX programme that only half-worked in this account-level post-mortem.

Frequently asked questions

What is customer journey mapping in B2B?

A B2B customer journey map documents how a buying group moves from trigger to purchase to renewal. Forrester's 2026 research puts that group at 13 internal stakeholders plus nine external influencers. A usable map records what each role needs at each stage, who owns supplying it, and the evidence behind the claim.

Why do B2B journey maps fail so often?

Three reasons dominate. They describe a single persona instead of a buying committee, so they cannot explain why a deal stalls. They use workshop estimates rather than system timestamps, so no data can contradict them. And they carry no owner and no review date, which means nobody ever updates them.

What data should a B2B journey map be built on?

Rank your sources. CRM stage-change timestamps first, then win/loss interviews run by someone outside the deal, then support tickets from the first ninety days after signing, then self-serve behaviour such as pricing-page visits and quote abandonment, then procurement and tender documents. Relationship survey scores rank last for journey work.

How often should a B2B journey map be reviewed?

Quarterly at minimum, and monthly if your category is moving quickly. Set the review date when you commission the map rather than afterwards, and name the person who calls it. A map without an expiry date is never revisited, which is how most of them end up behind a filing cabinet.

Is journey mapping worth the cost for a mid-sized B2B company?

Yes, if you treat it as a diagnostic rather than a project. A one-page map built from ten lost deals costs a fortnight of internal effort and usually exposes at least one broken handover between sales, marketing and service. A three-day facilitated workshop that produces a poster rarely repays itself.

The uncomfortable part of all this is that journey mapping mostly fails for organisational reasons, not analytical ones. The map is usually right. Nobody owns the stage it points at.

So here is the question worth taking into your next planning session: if you pulled up your current journey map today, could you name the person whose quarterly objective moves when stage three improves? If you can't, that gap is the work, not the map. I'd like to hear how you've solved it â€” tell me on LinkedIn.

Marketing in a Manufacturing Company: How to Win Engineers

funpal10-byte — Sat, 15 Aug 2026 10:17:19 +0000

Marketing in a manufacturing company fails for one reason, and it isn't the work. It's the currency. Engineers accept evidence, tolerance and repeatability; most brand decks offer adjectives. Reframe brand as demand-risk reduction and the argument changes, because 70% of technical buyers pick the better-known brand when two solutions are technically equivalent.

The first brand review I presented inside an engineering-led business ran eleven minutes before someone asked, politely, what the payback period was. Not the ROI — the payback period, in months, the way you'd ask about a conveyor upgrade. I didn't have one. I had reach, share of voice, and a nicely typeset positioning statement. The meeting ended early and I deserved it.

Fourteen years later I've run brand and communication inside tractors, trucks, printers, freight and now steel. The pattern repeats in every one of them, and so does the fix. What follows is the version I'd give my younger self before that eleven-minute meeting.

Why does marketing lose the argument inside engineering-led companies?

Because it asks to be believed rather than measured, in a building where nothing else is.

An engineering-led company runs on an evidentiary culture. A design change carries a test report. A supplier change carries a qualification trial. A capex request carries a payback in months. Then marketing arrives with a campaign proposal whose supporting evidence is a benchmark from a different industry and a slide titled "why brand matters."

This isn't anti-marketing prejudice. It's consistency. The same room would reject an unvalidated bearing spec with exactly the same politeness. The mistake marketers make is diagnosing it as hostility and responding with more persuasion, when the actual gap is procedural.

There's a scale problem underneath it too. India's manufacturing sector grew 8.4% in the first half of FY26 and holds a roughly 17–18% share of the economy, according to the Economic Survey 2025-26. These are large, capital-heavy businesses where a marketing budget looks like a rounding error next to a plant expansion — and is scrutinised harder precisely because nobody can point to the machine it bought.

How much of the decision is made before sales gets a meeting?

Most of it. The 6sense 2025 B2B Buyer Experience Report, based on more than 4,000 buyers, found that 94% of buying groups had already ranked their preferred vendors before first contact with a seller — and that the pre-contact favourite goes on to win the deal roughly 80% of the time. Sellers confirm decisions now. They rarely create them.

The technical-buyer data says the same thing from the other side. The 2026 State of Marketing to Engineers study from TREW Marketing and GlobalSpec found engineers spend 62% of the buying journey researching online, 53% say brand familiarity influenced their most recent purchase, and 70% are more likely to choose the better-known brand when comparing technically similar options.

That last figure is the whole argument, and it is the one to put on the first slide. It says the specification sheet stops being the tiebreaker at the exact moment two suppliers can both meet spec — which, in a mature industrial category, is most of the time.

Translating brand into a language engineering already accepts

The reframe that worked for me: stop selling brand as awareness and start selling it as procurement-risk reduction. A plant head choosing a supplier is not buying a feeling. They are buying the probability that the line doesn't stop. Familiarity is a proxy for that probability, and it is priced into the decision whether or not anyone admits it.

Brand in a manufacturing company isn't a communications budget. It's a procurement-risk discount you pay for once and collect on for years.

Here is the translation table I now use in the first meeting with any new engineering stakeholder.

| "We need to build awareness" | Unfalsifiable spend | "We need to be on the shortlist before the RFQ is written" | Share of shortlist appearances per quarter

| "This campaign drove engagement" | Vanity metric | "Enquiries from unqualified accounts fell; enquiries from target accounts rose" | Enquiry mix by target-account list

| "Our brand is trusted" | Assertion without test | "Buyers who knew us before the RFQ closed at a higher rate" | Win rate split by prior familiarity

| "We should invest in thought leadership" | Content for its own sake | "Technical buyers research for months without contacting sales; we should be what they find" | Inbound RFQs citing published material

| "Rebranding will modernise us" | Cost with no output | "Nine sub-brands are splitting the recognition one name should be earning" | Recognition per name, cost per name maintained

None of these are new metrics. They're existing commercial metrics, cut by a brand variable. That distinction matters more than it sounds: you're not asking the business to adopt your dashboard, you're asking to add one column to theirs.

Four moves that earned the function credibility

1. Borrow the engineering evidence standard, publicly. Every claim I made carried a source, a year and a sample size, the way a test report does. When I couldn't source something, I said so in the meeting rather than rounding it into a confident sentence. Credibility in that room is built by the things you decline to claim.

2. Start with the aftermarket, not the flagship. Deloitte's 2026 Manufacturing Industry Outlook notes that aftermarket services carry margins more than twice those of equipment sales. Aftermarket is also where marketing can show a short payback period, because the buying cycle is weeks rather than years. Win there first. The flagship budget gets easier to defend once you've already produced a number.

3. Put engineers in the content, not marketers. The 2025 Edelman-LinkedIn B2B Thought Leadership Impact Report, fielded among 1,934 executives, found 71% of hidden decision-makers have little or no interaction with sales, and 79% are more likely to advocate for a proposal from a company that consistently produces high-quality thought leadership. It also found 86% want perspectives that challenge their assumptions. Your application engineers already hold those perspectives. My job stopped being to write and became to extract, structure and publish.

4. Show up where the buying group actually is. McKinsey's 2024 B2B Pulse, covering nearly 4,000 decision-makers across 13 countries, found buyers now use an average of ten interaction channels, up from five in 2016, and that e-commerce has become the top revenue-generating channel at 34% of revenue. A plant buyer and a procurement head do not use the same three channels, and neither of them reads the corporate brochure.

The uncomfortable part of move three is that it makes marketing look smaller. You become the function that makes other people credible. That is exactly the trade that buys you the budget conversation eighteen months later.

What should a manufacturing marketing budget actually look like?

Lower than the cross-industry benchmark, and that's fine. Gartner's 2026 CMO Spend Survey of 401 marketing leaders puts average marketing budgets at 7.8% of company revenue, with 56% of CMOs saying they don't have enough to deliver the strategy. Quoting that 7.8% inside a heavy-industry business is a mistake — it's a blended figure dominated by categories with far higher gross margins and far shorter purchase cycles.

Bring the shape of the argument instead of the number. I've written separately on how much a B2B company should actually spend on brand, and the short version holds here: argue from category economics and purchase cycle length, never from a benchmark table.

Purchase cycle length is the argument that lands hardest in manufacturing. Professor John Dawes at the Ehrenberg-Bass Institute established the 95-5 rule in 2021: at any moment up to 95% of business buyers are not in the market, because firms change major suppliers roughly every five years. In capital equipment the replacement cycle is longer still. Every rupee of in-quarter lead-gen is fishing in a pond containing a small fraction of your buyers — which is a sentence a finance director understands immediately, because it's an inventory-turns argument wearing different clothes.

The mistake I'd warn you against

Don't fight for the word "brand" in year one. I did, twice, and both times I won the vocabulary and lost the budget.

In a product-led company, marketing should give up the word "brand" for the first year and earn the right to use it back. Talk about shortlist presence, enquiry quality, win rate by familiarity, cost per qualified RFQ. These are the same activities. They are simply named in a way that survives a plant review. Once two consecutive quarters show the pattern, you can put the word back on the slide and nobody objects, because by then it refers to something with a history.

The related trap is over-claiming attribution. Manufacturing buying groups are large, slow and multi-touch. A model that confidently attributes a ₹4 crore order to a LinkedIn campaign will be tested by someone who models mechanical systems for a living, and it will not survive the test. Under-claim. Show correlation, name it as correlation, and let the pattern accumulate.

What this means for you

If you're the first or only marketer inside an engineering-led business, here is the sequence I'd run.

Week one: get the last twelve months of won and lost deals, and tag each by whether the buyer had prior familiarity with you. This single cut is usually the most persuasive slide you will ever build.
Month one: pick one aftermarket or spares category with a short cycle and run a measurable campaign against it. Publish the result including what didn't work.
Month two: interview four application engineers and publish four technical pieces under their names. Not case studies — genuinely useful answers to questions buyers are already searching.
Month three: build the shortlist-presence metric and report it monthly alongside the sales pipeline, in the same deck, in the same format.
Quarter two: only now propose the brand investment. You'll be arguing from your own data instead of someone else's benchmark.

Two related pieces if you're building this function from scratch: why B2B brand awareness is the wrong metric to track covers what to measure instead of recall, and marketing in the supply chain deals with the specific problem of selling an invisible service. If your business runs multiple product names, the five-layer brand architecture framework is the place to start before you spend anything on campaigns.

One more thing worth holding onto. Deloitte's 2026 outlook estimates that more than 81% of task hours in manufacturing will remain human-driven even as smart-manufacturing investment accelerates. The buying decisions in this sector are still made by people who visit plants, remember bad shutdowns, and carry supplier reputations in their heads for a decade. That is a brand argument. It just needs to be said in a language the building already speaks.

So the question I'd put to any marketer inside a product-led company: if your CFO asked tomorrow for the payback period on brand, in months, what would you actually say — and what data would you need to have started collecting six months ago to say it? I'd like to hear how you'd answer that, on LinkedIn.

Why B2B Brand Awareness Is the Wrong Metric to Track

funpal10-byte — Thu, 06 Aug 2026 19:26:48 +0000

Brand awareness is the easiest number in marketing to move and the least useful one to defend a budget with. If your KPI dashboard leads with an awareness score, you are very likely measuring the wrong thing — and the CFO reading it over your shoulder can tell, even if they can't say exactly why.

I found this out the expensive way. Early in a stint running regional brand for an industrial manufacturer, I walked into a quarterly review with an awareness slide showing a 6-point lift across three markets. It landed flat. The sales head across the table asked one question: "Did we sell more trucks." I didn't have a good answer, and the campaign budget for the following quarter was cut by a third. The awareness number was real. It just wasn't connected to anything the room cared about.

What awareness actually measures

Unprompted and prompted brand awareness tell you whether people recognise your name when asked. That's it. They don't tell you whether your brand comes to mind unprompted, at the moment a buyer actually starts a purchase process, in the specific situation that triggers a search for a vendor like yours. That distinction — recognition versus retrieval at the moment of need — is the entire argument against using awareness as a primary KPI.

The Ehrenberg-Bass Institute's research on mental availability, built largely by Jenni Romaniuk, reframes this well: what predicts purchase isn't whether someone has heard of you, it's whether your brand is linked in memory to the specific triggers — the "category entry points" — that send a buyer looking for a solution. A logistics buyer doesn't wake up thinking about freight brands in the abstract. They think about it when a shipment is delayed, when a contract is up for renewal, or when a new plant needs a supply chain built from scratch. Those are entry points. Awareness scores don't touch them.

Why the 95-5 rule makes awareness even less useful

LinkedIn's B2B Institute has been pushing a number that should reset how most B2B marketers think about targeting: at any given time, roughly 95% of your buyers aren't in-market, and only about 5% are actively ready to buy. An awareness campaign optimised for broad reach is, by definition, mostly talking to people who cannot act on it this quarter. That's not wasted spend in the way performance marketers mean it — brand building for the 95% is real and it compounds — but it means the awareness number you're reporting this quarter has almost no relationship to revenue this quarter, and reporting it as if it does sets you up for exactly the conversation I had with that sales head.

Harvard Business Review's research on B2B buying puts a sharper point on where the real battle happens: 80 to 90% of buyers arrive with a self-researched shortlist already built before they ever talk to a vendor, and roughly 90% of the time the buyer ends up choosing from that Day-1 list. TrustRadius found something consistent with this in 2024 — 78% of buyers building a shortlist pick products they already knew of, and that number climbs to 86% among enterprise buyers. The list has also gotten shorter. Where consideration sets used to run five to seven vendors, they've compressed to three to five, sometimes one.

Put those together and the implication is uncomfortable: you are not fighting to be recognised. You are fighting for one of three to five slots on a list that gets built almost entirely before your sales team is in the room, based on associations that formed months or years earlier. Awareness tracking cannot tell you if you're winning that fight.

What to track instead

Category entry points give you something an awareness score never will: a specific, actionable list of the situations your brand needs to own. Ehrenberg-Bass research attributed to Romaniuk found that each additional category entry point a customer links to your brand reduces the risk of them defecting to a competitor by roughly 5%. Salesforce, by way of illustrating how far this can be taken, reportedly tracks 33 distinct entry points across its business. Most B2B brands don't need that many — the practical guidance from category entry point practitioners is to identify five to eight that matter most and track brand performance against each one specifically, rather than against a single undifferentiated awareness number. This is a close cousin of the storytelling problem I wrote about in brand storytelling for industrial categories — you can't tell a memorable story about a trigger you haven't identified.

Building this list isn't complicated, but it does require actual customer conversations rather than a survey template. You're listing the situations, moments, and problems that would make a buyer start thinking about your category at all — "our current 3PL missed a peak-season SLA," "we're setting up a new distribution hub," "the board asked why logistics costs are up 12%" — and then asking, for each one, whether your brand comes to mind and how strongly.

| Metric | What it tells you | What it misses

| Unprompted brand awareness | Whether people can recall your name when asked directly | Whether they'd think of you at the actual moment of need

| Prompted brand awareness | Whether people recognise your name from a list | Almost everything about purchase intent or timing

| Category entry point coverage | Which specific buying triggers link to your brand, and how strongly | Requires more setup; not a single tidy top-line number for a slide

| Shortlist inclusion rate | Whether you make the Day-1 list buyers build before contacting anyone | Hard to measure directly without buyer research or win/loss interviews

How do you present this to a board that only understands revenue?

This is the actual objection I hear most, and it's fair. A category entry point dashboard is a harder story to tell in one slide than "awareness up 6 points." The honest answer is that you tell a two-part story: leading indicators (CEP coverage, shortlist inclusion signals from win/loss interviews) paired with lagging ones (pipeline sourced from brand-attributed channels, win rate against known competitors). You do not replace awareness with nothing — you replace it with something that has a visible line, even an imperfect one, back to whether you're winning deals.

Where marketing budgets sit right now makes this more urgent

Gartner's 2026 CMO Spend Survey, based on 401 senior marketers at companies with over $1 billion in revenue across the US and Europe, put average marketing budgets at 7.8% of company revenue in 2026 — up marginally from 7.7% in 2025, but still roughly 18% lower than the average four years earlier. Budgets have not recovered to where they were, and every line item on that reduced budget gets scrutinised harder than it used to. A metric that can't explain itself to a CFO is the first one cut when the next review comes around — something I wrote about from the data-driven side in data-driven marketing and unlocking insights. I've watched it happen. It's not usually personal or even really about the number — it's that nobody in the room can connect the metric to a decision they'd make differently.

What this means for you

If your current brand tracker leads with an awareness score, don't rip it out — awareness data still has a role in understanding reach and category size. But stop leading with it in front of anyone who controls budget. Build a shortlist of five to eight category entry points specific to your business, start tracking brand association against each one, and pair that with whatever lagging revenue signal you can credibly attribute to brand activity. This is the same shift I've argued for on the supply-chain side in marketing alchemy and supply chain innovation — categories that don't think of themselves as brand-led still run on the same buyer psychology. It's a harder story to build. It's also the only one that survives the second question in the room.

Frequently asked questions

What's the difference between brand awareness and mental availability?

Brand awareness measures whether someone recognises your name when asked. Mental availability, the concept behind category entry points, measures whether your brand comes to mind unprompted at the actual moment a buying trigger occurs. The second is a much stronger predictor of whether you'll be considered when it matters.

How many category entry points should a B2B brand track?

Most practitioners recommend five to eight, built from real buyer conversations rather than a generic template. Salesforce reportedly tracks 33, but that reflects a business with dozens of distinct product lines and buying scenarios — most B2B companies don't need anywhere near that many to get useful signal.

Is the 95-5 rule exact, or a rough guide?

It's a rough guide popularised by LinkedIn's B2B Institute, not a precise measurement for every category. The number will vary by purchase cycle length and category, but the underlying point — that most of your audience isn't ready to buy at any given moment — holds directionally across almost every B2B category I've worked in.

Should marketing stop measuring awareness entirely?

No. Awareness still has a role in tracking category reach and long-term brand health. The argument here is against using it as the primary metric you defend a budget with, not against measuring it at all.

How do you get category entry points into a board presentation without losing the room?

Pair a small number of entry points (three to five, not all eight) with a lagging business metric — win rate against named competitors, or pipeline sourced through brand-attributed channels. Lead with the business number, use the entry point data to explain why it moved.

What's the metric your board actually trusts right now, and does it survive the "did we sell more" question? I'd genuinely like to know what's working in other categories.