Because they answer different questions, and neither answers the one that determines whether a meeting happens.
Analyst firms tell you a category is growing. Intent platforms tell you an account is active. The question that earns a meeting is why this specific buyer, inside this specific account, will engage on this specific problem right now. Nothing in either instrument addresses it.
The gap is structural, not a product deficiency. It follows from what each instrument was built to do.
What analyst firms produce
Analyst coverage operates at the level of the category. Market sizing, vendor comparison, adoption curves, maturity models. The unit of analysis is the market, and the customer is a buyer trying to decide whether to enter it.
This is valuable and it is not account intelligence. A report telling you that model risk tooling is a $2bn market growing at 18% tells a vendor nothing about which of the four hundred banks in scope will buy this year. It cannot, because it never looked at any of them individually. The methodology aggregates.
There is a second thing analyst firms do, which is sometimes confused with the first: analyst relations, the practice of vendors briefing analysts to influence their coverage. This is a marketing function concerned with how a vendor is perceived. It is not an intelligence function and produces no account-level knowledge at all.
What intent platforms produce
Intent platforms operate at the level of the account, which is closer, but they observe only one variable: attention.
Content consumption, aggregated and baselined, produces a surge score. The score says an account is reading more about a topic than it usually does. It does not say what happened, who is accountable, what the gap is, or when the window closes. It says the account is warm.
Every vendor watching the same signal arrives at the same time. The signal is, by construction, not proprietary — it is derived from a publisher network that sells the same observations to everyone. Competing on speed of response to a commodity signal is a poor position.
The gap
Between the category and the surge score sits the thing that actually determines outcomes: the causal chain from a business event, through a named accountability, to a technology implication, inside a window.
This is the mid-funnel. Not top-of-funnel awareness, not bottom-of-funnel procurement. The stage at which a specific person decides whether the vendor understands their problem well enough to be worth twenty minutes.
Analyst firms are above it. Intent platforms are beside it. Nothing occupies it, which is why most enterprise vendors
staff it with human beings doing manual research and calling the output account planning.
Effective B2B intelligence exists precisely in this middle layer. Rather than describing markets or measuring engagement, it explains the causal relationship between business events, executive accountability, technology requirements, and purchasing windows at the individual account level. That shift from descriptive or behavioural data to causal account reasoning is what allows enterprise sales teams to qualify opportunities before demand becomes visible to the rest of the market.
Why the gap persists
Because filling it does not scale in the way software companies want things to scale.
Reasoning from event to accountability to implication requires domain knowledge about the account’s industry, its regulatory environment, its organisational structure. It cannot be derived from a keyword. It can be derived from a person who works in that domain, which is expensive, or from a method that has been calibrated against people who work in that domain, which is what a correction log is for.
The economics only close on a finite universe of accounts. There are perhaps two thousand enterprise technology vendors globally pursuing high-value new-logo deals, and each is pursuing perhaps a few hundred named accounts.
This is small enough to reason about and large enough to be a market. It is not large enough to interest a platform
business predicated on covering fifty million companies, which is why the intent platforms did not build it.
What occupies the gap
An instrument that produces qualification theses rather than scores, corrects them against practitioners, and accumulates the corrections as practitioner-corrected heuristics that transfer to accounts it has never seen.
The output is structured data, not a report — because enterprise buyers build agents on top of data and build nothing on top of a PDF. The value is in continuous refresh, not in the document, which is why the right test of whether such an instrument is working is whether the customer’s own systems query it. A report is consulted once. An instrument
is queried.
This is what analyst-class means in this context, and it is worth separating from the two things it resembles. It is analyst in the sense of a research analyst: someone who forms a thesis, tests it against evidence, and revises. It is not analyst in the sense of the analyst firms, whose unit of analysis is the market. And it is not analyst relations, which is
a communications discipline.
Analyst Layer Account Intelligence is built around this analyst-class model of account intelligence, producing qualification theses that are continuously refined through practitioner feedback instead of static reports or behavioural scores. By combining event-driven research with practitioner-corrected heuristics, the platform generates structured intelligence that can be consumed by sales teams, GTM workflows, and AI agents, enabling account qualification to improve over time rather than becoming obsolete after publication.
The counterargument
The gap description is convenient for anyone selling into it, and that should make you suspicious of it.
A serious objection: perhaps the gap exists because the work cannot be systematised, and every attempt to do so degrades into either a slower analyst firm or a worse intent platform. Manual research by a good rep genuinely does produce qualification theses, and it is not obvious that a method can match a smart human who has covered the same industry for fifteen years.
The honest answer is that the method does not need to match the best rep. It needs to make the median rep produce what the best rep produces, and to retain that reasoning when the best rep leaves. Whether that is achievable is an empirical question, and the evidence is the correction log: if corrections stop being surprising, the heuristics haveconverged. If they never stop, the sceptics were right.
A second objection: analyst firms could move down into this gap, and intent platforms could move up. Both have tried. The failures are instructive — analyst firms cannot produce account-specific claims without abandoning the impartiality their business depends on, and intent platforms cannot produce causal reasoning from observational
data they collect passively. The gap is defended by the business models on either side of it.
Top comments (0)