DEV Community: Artefact Ventures

GTM engineering: why your go-to-market problem is a systems problem

Alex Boissonneault — Tue, 15 Sep 2026 21:39:25 +0000

The first thing I do on a new engagement is export the CRM property list and sort it by fill rate. It takes ten minutes, and it usually settles the argument about whether the company has a marketing problem.

It almost never does. It has an engineering problem that nobody was hired to solve.

What is GTM engineering?

GTM engineering is the practice of designing, building and maintaining the systems that run a company's go-to-market motion, with the same rigour an engineering team applies to production software. The data model, the pipeline architecture, the automations, the feedback loops.

It treats revenue as a system with inputs, transformations, outputs and failure modes. And it treats the stack the way you would treat a codebase: versioned, tested, documented, owned by someone with a name.

If you write software for a living, here is the translation table I use with engineering leaders:

Software engineering	GTM engineering
Schema design	CRM object model
Data pipeline	Lead to opportunity to revenue flow
Observability	Pipeline metrics, leading indicators
Tech debt	Property sprawl, duplicate records, dead automations
Incident postmortem	Lost-deal and churn analysis
CI/CD	Campaign launch process

The mapping is literal, and it explains why so many GTM stacks fail in ways any engineer would recognise on sight.

The failure mode that has no name on the revenue side

Run the export I described above on your own CRM. Sort by fill rate. Count how many custom properties sit under five percent. Then ask three people on three different teams what "qualified" means, and write down the three answers.

You now have the two measurements that describe most B2B revenue stacks I open. A schema that was never designed, only accumulated, one campaign and one departed employee at a time. Deal stages named years ago by someone who no longer works there. Definitions that live in people's heads and disagree.

Engineers have a word for this. Tech debt. Revenue teams do not have a word for it, so they buy another tool, which is roughly like fixing a broken schema by adding a caching layer.

The cost shows up as leakage at the joints. Handoffs that drop leads, stages that hide stalled deals, attribution gaps that send budget toward channels that are not working. I wrote a longer breakdown of where those leaks hide in revenue leakage: how architecture gaps cost B2B companies growth. The short version is that the money disappears where the system has no seam to catch it, which is never where you go looking.

Three layers, in order

When I rebuild a go-to-market motion, I work in three layers, strictly in order. Skipping ahead is the most common mistake and the most expensive one.

Layer 1: the blueprint, schema before software

Before touching any tool, you need a designed data model. What objects exist, what states each can be in, what event moves it from one to the next. This is schema design, and everything downstream inherits its quality. A CRM built on a bad object model is a database with a UI, and every report you run on it is fiction with formatting.

Layer 2: the execution system, pipelines rather than heroics

Once the model is right, automate the transitions. Routing, enrichment, stage progression rules, handoff triggers. The goal is boring: no revenue-critical step should depend on someone remembering to do it. This is where you recover most of the manual work tax, and it is also where automating chaos becomes tempting. Automating a design and automating a mess look identical in a project plan.

Layer 3: growth R&D, observability and iteration

Then instrumentation. Leading indicators, not only lagging ones. Cohort views, not only snapshots. The test is simple: when a metric moves, can you answer "what changed and why" within an hour? If not, you have a dashboard, and a dashboard is decoration.

Why AI made this urgent

Everyone is bolting language models onto their GTM stack right now. Agents that draft outreach, score leads, summarise calls. The results are bimodal in a way that has nothing to do with the model anyone chose.

Companies with a clean data model and defined processes get compounding leverage, because the agent has real state to read and real actions to take. Companies with undesigned stacks get confident nonsense at volume, because the agent is reasoning over sediment.

An agent reading a CRM with three definitions of "qualified" will pick one, apply it at machine speed, and never mention that it chose.

Which makes this discipline the prerequisite for the whole AI-in-revenue wave rather than a nice-to-have for scale-ups.

Where you stand

Five questions. One point for each confident yes.

Could a new hire learn your pipeline stages and their exit criteria from documentation rather than from a colleague?
Do marketing, sales and customer success agree in writing on the definition of a qualified lead?
If a deal has been stuck for 30 days, does the system flag it, or does a human have to notice?
Can you trace any closed deal back to its first touch without opening a spreadsheet?
Has anyone deleted an automation or a property in the last quarter? Deletion is a sign of ownership. Sprawl is a sign of abandonment.

Zero to two: you are running on heroics, and your growth ceiling is your team's stamina. Three to four: you have a system carrying debt, which is normal and fixable. Five: your next lever is AI leverage, not headcount.

The structured version of that diagnostic is free at artefactventures.com, along with the frameworks I use in client work. Most of it has no email wall, because I would rather you use it than bounce off a form.

The part that is uncomfortable

The companies winning right now are the ones whose revenue motion behaves like well-run software. Designed schema, automated pipelines, observable state, and agents operating on top of all three.

That is GTM engineering. It is learnable, and if you already understand systems and how companies make money, it may be the most underpriced skill set of the next five years. I say that as someone who spent a long time being told this work was administrative.

Over the next three months I am publishing a series here that goes layer by layer. The CRM data model that ends data silos, what agents actually need from your stack including the MCP server I built for it, attribution as data engineering, forecast accuracy at the schema level, and finally how to simulate a go-to-market motion before spending on it.

Follow along if that intersection is your thing. And if you are staring at a stack that matches the failure mode above, that is the work I do.

I am a revenue operations consultant at Artefact Ventures. I write about the engineering underneath revenue.

RFM Segmentation for B2B SaaS: An 11-Segment Model to Cut Churn (with Python)

Alex Boissonneault — Wed, 15 Jul 2026 22:27:28 +0000

Most retention teams treat their customer base like a single audience. One email sequence. One renewal push. One "we miss you" campaign.

Then they wonder why churn stays stubbornly flat.

The problem isn't the message. It's the model. They're targeting a distribution as if it were a point.

RFM segmentation fixes this, but the textbook version was built for e-commerce. B2B SaaS has different buying patterns, longer sales cycles, multi-seat dynamics, and expansion revenue that the classic model completely ignores.

After running this analysis across a dozen B2B clients ranging from $1.6M to $70M in revenue, here's the adapted 11-segment model we use, how we score it, and what retention plays actually move the needle for each segment.

Why Classic RFM Breaks Down for B2B SaaS

The original RFM framework (Recency, Frequency, Monetary) was designed for transactional businesses: retail, e-commerce, subscription boxes. The logic is simple: customers who bought recently, bought often, and spent the most are your best customers.

In B2B SaaS, this falls apart for three reasons:

1. Frequency isn't purchase frequency, it's engagement frequency.
A B2B SaaS customer might renew once a year but log in daily. Or they might have bought three seats but nobody's touched the product in 90 days. Transaction frequency tells you almost nothing about health.

2. Monetary value is backward-looking without expansion signals.
A customer at $500/month MRR might be your fastest path to $2,000/month if they have three more teams that could use the product. Or they might be on a legacy plan that will churn at renewal. MRR alone doesn't tell you which.

3. Recency is a lagging indicator.
By the time "recency" signals a problem (low login frequency, no new users added), it's usually 60–90 days too late to save the account cost-effectively.

The fix: redefine each dimension for product-led and sales-led SaaS, then score them at the account level, not the individual user level.

Redefining R, F, M for B2B SaaS

Before we segment, we need metrics that actually predict retention and expansion.

Recency → Last Meaningful Engagement

Don't just track last login. Track last meaningful action:

Last time a core workflow was completed (not just a page view)
Last time a new user was added or a new feature was activated
Last time the customer engaged with your team (support ticket, QBR, check-in call)

Score 1–5, where 5 = meaningful engagement within 14 days, 1 = no meaningful engagement in 90+ days.

Frequency → Breadth × Depth of Usage

In B2B, you want to measure two dimensions of usage:

Breadth: % of licensed seats that are active monthly
Depth: How many core features or workflows are in regular use

Combine them: Frequency Score = (Active Seats / Licensed Seats) × (Active Features / Total Features), normalized to 1–5.

A company using 90% of seats but only 1 feature is a different risk profile than one using 3 seats but 8 features. Both matter.

Monetary → Expansion Potential Score

This is where B2B diverges most from classic RFM. Current MRR is useful but insufficient. We score Monetary on:

Current MRR as % of estimated account capacity (total addressable seats × price)
Expansion history: has the account added seats or upgraded in the last 12 months?
Contract type: month-to-month vs. annual (annual = higher score baseline)

A $200/month account at 10% of capacity scores higher on expansion potential than a $2,000/month account that's fully saturated.

The 11-Segment Model

Once you have R, F, M scores (each 1–5), you can compute a composite score and map accounts to segments. Here's the full model:

#	Segment	R	F	M	Profile
1	Champions	5	5	5	Power users, likely advocates, expanding
2	Loyal Expanders	4–5	4–5	4–5	Strong users, actively adding seats/features
3	High-Value Sleepers	1–2	4–5	4–5	Disengaged recently but historically strong
4	Potential Loyalists	4–5	3–4	3–4	Good engagement, room to deepen value
5	Recent Big Bets	5	1–2	4–5	New or renewed, high contract, low usage yet
6	Promising Actives	4–5	3–4	1–2	Engaged product users, low spend so far
7	Needs Activation	3–4	1–2	3–4	Paying well but not using the product
8	At Risk	2–3	2–3	2–3	Declining across all dimensions
9	Can't Lose	1–2	1–2	4–5	High-value accounts going dark (emergency)
10	Hibernating	1–2	1–2	2–3	Low across all dimensions, renewal risk
11	Lost (Pre-Churn)	1	1	1–2	Effectively churned, may not have cancelled yet

Segment-by-Segment Retention Playbooks

Segmentation only matters if your plays are different per segment. Here's what actually works:

Segment 1: Champions

Risk: Low. Opportunity: Advocacy, case studies, expansion.

Don't leave these accounts alone; engage them intentionally. This is your referral engine. Invite them to beta programs. Ask for testimonials. Connect them to your product team. Champions who feel seen expand faster and refer more.

Play: Executive sponsor check-in + referral ask. Offer co-marketing opportunity. Tag for beta access to new features.

Segment 2: Loyal Expanders

Risk: Low. Opportunity: Upsell, multi-year lock-in.

These accounts are on a growth trajectory. They're your strongest candidates for annual-to-multi-year upgrades, tier expansions, or enterprise plan migrations.

Play: QBR focused on ROI proof points. Present expansion roadmap. Propose multi-year discount.

Segment 3: High-Value Sleepers

Risk: Medium-High (fast-moving). Opportunity: Reactivation before renewal window.

These are your scariest accounts. High historical value, but something changed recently: often a champion left, a new admin was assigned, or a workflow broke. You have a short window.

Play: Immediate CSM outreach, not templated, personal. Reference their historical usage. Book a re-onboarding session. Assign an internal champion as DRI.

Segment 4: Potential Loyalists

Risk: Low-Medium. Opportunity: Deepening product stickiness.

Good engagement, but they haven't unlocked the product's full value. One power feature adoption away from becoming a Champion.

Play: Feature spotlight email sequence. Offer a workflow audit call. Surface one underused feature that maps directly to their use case.

Segment 5: Recent Big Bets

Risk: Medium (early stage). Opportunity: Fast time-to-value → prevents buyer's remorse.

These accounts just signed or renewed a large contract but haven't fully adopted yet. The gap between sales promise and product reality is widest here.

Play: Structured onboarding with milestones. Weekly check-in for 60 days. Define 90-day success metrics with the customer explicitly. Assign a dedicated CSM.

Segment 6: Promising Actives

Risk: Low. Opportunity: Monetize the engagement.

These users love the product but are on a starter plan or underinvested relative to their engagement. They're growing into the product; help them grow into a larger plan.

Play: In-app nudge when they hit feature limits. Personalized upgrade offer tied to a specific feature they're hitting limits on. One call from sales (not CS).

Segment 7: Needs Activation

Risk: High (silent churn). Opportunity: Activation play can unlock product value fast.

The most dangerous segment that looks safe because MRR is healthy. They're paying, but they're not using. When renewal comes, they'll ask "what did we actually get from this?"

Play: Activation audit: which users are licensed vs. active? Run a targeted user activation campaign internally with the admin contact. Offer a free implementation session.

Segment 8: At Risk

Risk: High. Opportunity: Save rate varies.

These accounts are declining across all three dimensions. They haven't fully churned yet, but the trajectory is clear.

Play: CSM-led save call. Understand the real reason for disengagement (not "how can we help" but "what would need to change for this to be worth it to you?"). Consider a plan adjustment or pause option to retain relationship.

Segment 9: Can't Lose

Risk: Critical. Opportunity: High-value accounts are still winnable with speed.

This is your fire alarm segment. High MRR, but engagement has fallen off a cliff recently. Every day without outreach reduces save probability.

Play: Executive-to-executive outreach within 24 hours. Skip the template. Propose a success review with your CEO or VP. Consider a contract adjustment to buy back engagement.

Segment 10: Hibernating

Risk: High. Opportunity: Low, but worth testing.

Low across the board, but not fully lost. These are often small accounts that never fully onboarded. Not worth high-touch CS resources, but worth a structured win-back sequence.

Play: Automated 3-email win-back sequence over 30 days. Offer a 30-minute re-onboarding call. If no response, flag for churned treatment at next renewal.

Segment 11: Lost (Pre-Churn)

Risk: Near-certain churn. Opportunity: Clean data, future re-acquisition.

They're technically still a customer, but they've mentally left. Don't throw expensive resources at saves that won't close. Focus on clean offboarding, exit survey data, and tagging for re-acquisition at a future date.

Play: Short exit survey. Document reason for churn in CRM. Set a re-engagement automation for 6–12 months post-churn if they don't formally cancel.

How to Build This in Practice

You don't need a data warehouse to start. Here's a pragmatic three-step approach:

Step 1: Define your RFM proxies

Map your product's available data to R, F, M:

Recency: last login date, last workflow completion, last support ticket (pick the highest-signal one)
Frequency: DAU/MAU per account, feature adoption breadth
Monetary: MRR, contract value, expansion history

If you're on HubSpot + a product analytics tool (Mixpanel, Amplitude, PostHog), most of this data is already available, it just needs to be joined at the account level.

Step 2: Score and segment monthly

Run the scoring monthly, not in real time. RFM is a strategic tool, not an alert system. Monthly cadence matches your CS team's capacity to act.

def score_rfm(recency_days, seat_adoption_pct, feature_adoption_pct, 
              mrr, account_capacity_mrr, has_expanded):

    # Recency score (1-5)
    if recency_days <= 7: r = 5
    elif recency_days <= 14: r = 4
    elif recency_days <= 30: r = 3
    elif recency_days <= 60: r = 2
    else: r = 1

    # Frequency score (1-5): breadth x depth composite
    breadth = seat_adoption_pct  # 0.0 to 1.0
    depth = feature_adoption_pct  # 0.0 to 1.0
    f_raw = (breadth + depth) / 2
    f = max(1, round(f_raw * 5))

    # Monetary score (1-5): expansion potential weighted
    capacity_utilization = mrr / account_capacity_mrr if account_capacity_mrr > 0 else 1
    expansion_bonus = 1 if has_expanded else 0
    m_raw = (1 - capacity_utilization) + (expansion_bonus * 0.2)
    # Blend with absolute MRR tier
    mrr_tier = min(5, max(1, round(mrr / 500)))  # $500 = score 1, $2500+ = score 5
    m = round((m_raw * 3 + mrr_tier * 2) / 5)
    m = max(1, min(5, m))

    return r, f, m

def assign_segment(r, f, m):
    score = r + f + m
    if r >= 5 and f >= 5 and m >= 5: return "Champions"
    if r >= 4 and f >= 4 and m >= 4: return "Loyal Expanders"
    if r <= 2 and f >= 4 and m >= 4: return "High-Value Sleepers"
    if r >= 4 and f >= 3 and m >= 3: return "Potential Loyalists"
    if r >= 5 and f <= 2 and m >= 4: return "Recent Big Bets"
    if r >= 4 and f >= 3 and m <= 2: return "Promising Actives"
    if r >= 3 and f <= 2 and m >= 3: return "Needs Activation"
    if r <= 2 and f <= 2 and m >= 4: return "Can't Lose"
    if 2 <= r <= 3 and 2 <= f <= 3 and 2 <= m <= 3: return "At Risk"
    if r <= 2 and f <= 2 and m <= 3: return "Hibernating"
    return "Lost"

Step 3: Route to the right play

Map each segment to a specific CS action, owner, and SLA:

Segment	Owner	SLA	Action
Champions	CS Lead	30 days	Advocacy program invite
Can't Lose	VP Customer Success	24 hours	Exec escalation call
Needs Activation	CSM	7 days	Activation audit
At Risk	CSM	3 days	Save call
Lost	Automation	30 days	Win-back sequence

What Changed When Clients Actually Used This

Across the clients where we implemented this model, a few consistent patterns emerged:

The "Needs Activation" segment was almost always bigger than expected. Most teams thought 5–10% of accounts were dormant. The actual number was typically 20–30%. These were accounts paying on time (so they never triggered a churn alert) but not using the product. Activating them extended LTV significantly. We saw 40–60% improvement in renewal rates for this segment specifically when CSMs intervened 90 days before renewal.

"Can't Lose" accounts were being handled too slowly. Without explicit segmentation, high-value at-risk accounts often went 3–4 weeks before a CSM noticed something was wrong. With the segment flagged and a 24-hour SLA, save rates for this group more than doubled.

Champions weren't being asked for anything. Most CS teams were so focused on at-risk accounts that they completely neglected their best customers. When we added structured advocacy outreach for Champions, referral-sourced pipeline increased from near-zero to a meaningful percentage of new business.

The Honest Limitations

RFM is a diagnostic tool, not a prediction engine. A few caveats worth naming:

It doesn't capture stakeholder risk. If your champion just left the company, no RFM model will flag it until the engagement drop shows up 30–60 days later. Layer in relationship tracking (CRM contact health) separately.
The segment boundaries are arbitrary. The thresholds in the model above are starting points, not laws. Calibrate them against your actual churn data. A "score 3 on recency" might mean different things for a daily-use tool vs. a quarterly planning tool.
It's only as good as your data. If you don't have product usage data at the account level, you're flying blind on the Frequency dimension. This is a reason to invest in the data plumbing, not a reason to skip segmentation.

Getting Started Today

You don't need to build all 11 segments on day one. Start with three:

Flag "Can't Lose": your highest-MRR, lowest-engagement accounts
Flag "Needs Activation": accounts paying but not using
Flag "Champions": your best accounts to activate as advocates

Run it manually in a spreadsheet if you have to. The insight that comes from seeing those three buckets clearly for the first time is usually enough to justify the investment in making it automatic.

Retention isn't about catching churn. It's about understanding the distribution you're actually working with, and giving each segment the attention it deserves.

If you're building this kind of revenue intelligence tooling and want to see how the segmentation integrates with HubSpot and a full CRM workflow, I'm happy to share more details in the comments.