Competitive positioning research explained

Competitive positioning research has to distinguish the position you hold from the position you could hold after a change. A brand tracker measures current perceptions over time. A proposed repositioning or a rival’s next move asks a conditional question: how would the audience respond to a different set of offers? We would define that change before choosing what to measure, rather than treat a current brand ranking as an answer about a future market.

Confidentiality also affects the design. In our transport-policy study, showing draft narratives to the intended audience would have changed the debate the client wanted to understand. Simulation allowed the team to test those narratives without putting them into circulation.

Our survey benchmark gives you a basis for assessing the underlying method: 86% distribution accuracy across 1,000 surveys, against a 91% human self-replication ceiling, according to our Survey Evaluation Report, January 2026. Distribution accuracy concerns the pattern of survey answers. The evaluation measures survey responses, so we use it to support the method without presenting it as accuracy on a competitive brand-comparison task.

A competitive positioning study starts with an audience grounded in diverse observations, from public signals or your first-party data. The proposed test then presents your offering alongside real rivals and entrants described by their positioning. The design examines how personas weigh attributes and price alongside trust and positioning. We would structure the research around the following decisions:

1. 1.

   Define whose preference matters and identify the observations available to ground that audience.

2. 2.

   Set out your offering and the competitive alternatives, distinguishing existing rivals from hypothetical entrants.

3. 3.

   Examine preference and perceived differentiation, including the reasons an audience member would switch.

4. 4.

   Read the results by segment and inspect the reasoning behind the selections before deciding which positioning to pursue.

Our consumer-goods case study shows what grounding an audience involved. We worked with social listening partners to gather over 100,000 anonymised profiles from platforms like Instagram, Facebook, TikTok, Reddit and X, alongside health forum discussions and reviews of health-adjacent products. We combined similar profiles into enriched lookalike consumers and de-identified their behavioural traces. Survey questions requested a verbatim reason for each selection, so the team could examine why a concept appealed.

That study compared a client’s own concepts. It supports asking for reasons alongside choices; it does not establish a switching rate between rival brands.

A hypothetical competitor can enter the research design through a description of its positioning. You can therefore examine a possible rival move before choosing your response, or test your own proposed position against that entrant. We would use this for war-gaming a strategic decision. The result belongs to the scenario you tested; a plausible description does not establish what a rival will actually launch.

Our transport-policy case study shows how hypothetical scenarios worked in practice. For a global transportation leader, we built the Washington D.C. Opinion Leaders society: 1,500 AI personas spanning more than 360 organisations and 800 distinct job titles, according to the transport-policy case study. The study began with baseline attitudes, tested two policy narratives and their proof points blind, then introduced damaging coverage and hypothetical scenarios. The team could examine which arguments survived a hostile context before committing to a narrative.

The transport-policy case study records 250,000 responses to more than 170 questions across three phases, and identifies a credibility gap alongside the strongest narrative and proof points. The persona and response totals are rounded figures. That is direct evidence for testing positioning under hypothetical conditions. The alternatives were the client’s own narratives; hypothetical rivals remain a design application, with no published client result claimed here.

Our published positioning work has informed choices between narratives and product concepts. The table distinguishes the alternatives tested from the decision the research supported. For a strategy lead, the useful comparison is between these research designs and the decision you face, while keeping the competitive set explicit.

Sources: Artificial Societies’ transport-policy, Teneo and consumer-goods case studies. Teneo’s engagement took place in late 2025.

Teneo received a written report and an interactive platform for examining approval scores, emotional sentiment and verbatim reactions. Its late-2025 engagement generated 189,756 unique responses from 30 deep-dive questions, according to our Teneo case study. The Teneo study names the audiences: Washington D.C., with 1,364 personas; Tech Leaders, with 1,526; and General Population, with 2,381. We identified an overall narrative, then refined the messaging for each audience segment.

For a choice among your own proposed products, our AI concept testing guide develops the consumer-goods design. For a competitive positioning decision, use these cases to assess the experimental approach and the depth of explanation you should request, while retaining rival brands as a distinct study requirement.

Brand tracking follows current perceptions among reachable audiences over time. Competitive positioning simulation examines how perceptions would change after a proposed repositioning or competitor move, including among hard-to-reach audiences. We would keep the tracker as the record of current perception and use simulation to investigate a specified strategic change. Our Teneo work illustrates the forward-looking question: in late 2025 we supported Teneo on a project for a major US company preparing to launch a new technology strategy, as the Teneo case study records.

The transport-policy engagement shows why that distinction matters. The client needed to compare sensitive draft messages with an audience it could not safely approach about those messages. Our published account describes an advisory report on positioning, proof points and risk, plus access to examine segment cuts and individual persona reactions. Those outputs served a decision about what to say before saying it publicly.

Competitive positioning simulation can examine the perceptions behind preference within a stated competitive set. Our published studies establish experience with confidential narratives, concept choices and hypothetical scenarios. We describe rival-brand testing as a research design: none of our five published case studies reports a competitive set of rival brands, and none reports a share-of-preference, switching or win-rate figure.

Audience size follows the decision. Our hyperscaler earnings-call study included 14 Sell-Side Analysts, representing those who had asked questions on the company’s last eight calls, alongside Buy-Side Investors, Corporate Ecosystem and Financial Media societies. For a positioning study, we would start by identifying whose judgement you need to understand.

The audience also sets a firm boundary. Where diverse observations do not exist, there is nothing to ground a persona in. Public signals and client first-party data are the starting conditions for this work. We would judge a proposed study by the audience evidence and the strategic choice it must inform, then keep the eventual recommendation within the set of alternatives actually tested.