When any organisation can synthesize all published market data in seconds, that intelligence no longer creates competitive advantage — it is a shared baseline. The only intelligence that changes decisions now is the kind AI cannot access: found by reaching the right people in your target market and asking about choices they have already made.
38 in every 100 mandates — primary evidence from verified market participants directly contradicts what published intelligence says about the most important strategic assumption.
Reports across 50+ industry verticals — scanned before every engagement to establish the published baseline and identify precisely where primary human evidence is needed.
From agreed scope to research findings — fielded with verified human respondents in the target market. Not a synthesis report. Evidence collected from real people.
AI has made secondary market intelligence free, fast, and universally available. The same published forecasts, the same synthesized reports, the same competitive signals — available to every organisation through the same tools in the same seconds.
Intelligence that every competitor holds simultaneously is not an advantage. It is a shared baseline. And that baseline is wrong with surprising regularity: in 38 of every 100 primary research mandates, the evidence gathered directly from people in the target market directly contradicts what published intelligence says about the most critical assumption in the engagement.
The gap between what the published market believes and what the right people in that market have actually done is where competitive intelligence lives. Closing it requires something no AI can do: reaching the right human beings — the decision-makers, evaluators, gatekeepers, and practitioners in your target market — and asking about decisions they have already made, not intentions they hold for the future.
That is what primary research is for. Not describing the market. Returning evidence from it — evidence that does not exist in any published source and can only be found by asking the right people the right questions directly.
The more powerful AI becomes at secondary synthesis, the more value migrates to the one layer it cannot reach — evidence that exists only in the minds of real people who have already lived the decision you are about to make.
As AI democratizes synthesis, the value of market intelligence migrates entirely to the third layer. The only intelligence with competitive value in the AI age is intelligence that has to be found — not synthesized. That means reaching the right people in the target market and asking about decisions they have already made. Everything else is shared baseline.
One leads to intelligence every competitor already holds. The other reaches evidence that exists only inside real human beings — and only surfaces when the right question reaches the right person directly.
Intelligence every competitor holds in identical form, from identical sources, derived in identical seconds.
Evidence that did not exist before it was collected. Only your organisation has seen it.
The question design determines whether a finding reflects aspiration or behaviour. These produce structurally different answers — and different outcomes when decisions follow from them.
"Are you planning to adopt this technology in the next 12 months?"
"How likely are you to specify this material in your next product cycle?"
"Would you pay a premium for a sustainable alternative at this price point?"
"What proportion of next year's budget will go toward this category?"
"In the last 18 months, which technologies in this category has your committee approved for active deployment?"
"When did your organisation last commit budget to this category, and what drove that allocation?"
"Has your committee approved any premium specification in this category in the last two years? What did the approval path require?"
"What line item does this category appear under in the current cycle? Who authorises it at what threshold?"
Asking people what they plan to do captures aspiration, social desirability, and expressed alignment — not what committees approve at budget review. This is the mechanism behind the 38% divergence: the stated market view says one thing, documented decisions by the relevant people reveal another.
A committee that has approved a specification before will do so again under similar conditions. One that has consistently rejected it will continue to. Past decisions by the right people are a more reliable predictor of future commitments than any statement of future intention — because they are not subject to social desirability or aspiration bias.
A strategy team was approaching a major capital allocation review with a market entry recommendation already shaped by secondary intelligence. The published corpus, AI-synthesized intelligence, and category-level survey data all pointed toward strong adoption momentum in the target geography. The internal narrative around the opportunity had been building for over a year.
A 14-day primary research mandate was commissioned to answer one question: had the relevant decision-making authorities in the target market actually allocated budget for this category in the current cycle — and if so, at what scale and under what conditions?
Not one of the 34 verified respondents — each confirmed as holding actual authority over the relevant decision — had allocated budget for the category in the current cycle. The adoption momentum was real at the awareness and interest level. The commitment had not followed.
The layer blocking adoption existed below the level that any published source or stated intent survey reaches. It was a structural budget classification issue that had not been surfaced through standard intelligence channels. The capital commitment was restructured. The entry strategy was redesigned around the actual decision-making pathway, not the assumed one.
Every verified respondent held confirmed decision-making authority. The stated interest was genuine. The commitment had not been approved.
Fixed fee. 14-day delivery. All respondents verified before their response entered analysis.
The difference between the original staged plan and the original full commitment — changed by one primary finding before deployment.
The budget classification barrier existed only inside the minds of the people who managed it. It was not reported, not published, and not visible to any secondary source.
The target audience for primary research is defined by the research question — not by a commercial role. It may be procurement authorities, clinical evaluators, technical specifiers, formulary committees, regulatory practitioners, or channel gatekeepers.
The difference between a respondent who claims to hold relevant authority and one who has actually exercised it is the difference between evidence and noise. Standard panel verification was built for a world where fraudulent completions were manual.
Our screening process operates across seven independent layers before any contribution is accepted into analysis. Each layer catches a different category of error. A response that fails any single layer is removed from the dataset.
The result is a dataset composed entirely of verified human beings who hold the actual decision-making or evaluative authority the engagement requires.
Each addressing a distinct failure mode — identity fraud, role misrepresentation, AI-generated responses, and duplicate contributions
Completions removed after full screening — respondents who pass standard checks but do not pass ours
Profile and company domain cross-referenced to confirm the respondent is who they claim to be.
Title is insufficient. Actual decision-making or evaluative role in the relevant category is confirmed independently — not self-reported.
Sector, geography, revenue, and organisational profile confirmed against the target audience specification for the engagement.
The same individual contributing under different identities is identified and removed before the dataset is assembled.
Open-text responses are analysed for linguistic signatures consistent with AI generation — responses not written from direct professional experience.
Responses that contradict each other within the same engagement are identified and excluded before analysis.
Completion time, response depth, and open-text authenticity — contributions showing automated completion or disengagement are removed.
13,241-report corpus scanned across relevant verticals. Establishes what secondary intelligence already knows — and precisely where it cannot answer the specific question.
Behavioral-anchored research with the verified target audience. Every respondent cleared through 7-layer screening. Questions ask about past decisions — not future intentions.
Primary findings mapped against the published baseline. Where human behavioral evidence diverges from secondary intelligence, the gap is surfaced explicitly and quantified.
A directional finding — grounded in behavioral evidence from verified human respondents. Delivered with a structured review session.
Primary research is commissioned to answer a specific question about a specific assumption. The engagement returns a directional finding — a clear statement of what behavioral evidence reveals about that assumption.
Behavioral evidence from verified participants in the target market aligns with the assumption. The relevant decision-makers have actually approved, committed to, or acted on the relevant category in ways consistent with the strategic assumption.
The internal view is strong but has never been tested against primary evidence from the actual market.
Behavioral evidence partially supports the assumption — in some segments and not in others. The conditions are specific: a geography where adoption is documented, a company size below which budget authority has not been established.
The strategic question involves geographic expansion, segmentation decisions, or phased deployment.
Behavioral evidence from verified participants in the target market contradicts the assumption that secondary intelligence supports. What the corpus predicts is not what the relevant decision-makers have actually done.
Received before a significant capital commitment. The earlier a Contradicted finding arrives, the more it reshapes the decision.
Primary research is commissioned when a capital decision depends on an assumption that cannot be resolved from published sources. The fee is fixed at engagement start — determined by the scope of the question, not by time spent.
Fixed fee against agreed scope. No change orders. No time-based billing. The fee does not change once the engagement begins.
The scale and complexity of the strategic question — not the size of your organisation or the seniority of the decision-maker.
If the question can be resolved from existing published sources, we say so in the 30-minute scoping call — before any fee is agreed.
A single, precisely bounded assumption that requires primary evidence before a specific decision is made.
Typical decision value: $1M–$30M. Research represents less than 2%.
A complete primary research engagement for decisions where the strategic question is complex enough that a single method would leave meaningful uncertainty.
Typical decision value: $20M–$200M. Research represents less than 0.45%.
One governing question, pursued across multiple geographies or audience segments through sequenced mandates.
Typical decision value: $100M+. Research represents less than 0.1%.
All engagements are fixed-fee. Scope and fee are agreed before fieldwork begins. The scoping call is 30 minutes and carries no commitment.
We scope the engagement in 30 minutes. If behavioral evidence from verified participants in your target market would change what you decide — we can have that evidence in your hands within 14 days.
The scoping call has one purpose: to determine whether primary research would produce findings that are decision-relevant for your specific question. If it would not, we say so directly in that call — before any fee is agreed.
Your brief and decision context are confidential and not shared with any third party.
30-minute scoping call within 2 business days. We tell you in that call whether primary research would change your picture.
Fixed fee agreed before fieldwork begins. No open-ended billing, no scope creep.
If the question can be resolved from existing published sources, we will tell you — and point you toward them.
