Competitive Research
Understand your market with confident insights and data.
When you need a number you can defend in a board meeting, you need conclusive research: representative samples, sound instruments, real fieldwork, and analysis that holds up.
This is measurement. What price maximizes revenue. How the brand is really performing. Whether the concept is ready. What customers actually think — at a scale that settles the argument.
What we help with
Customer experience
What customers do, what they think of it, and what happens after the sale — at a scale that settles the argument.
Market & competitive intelligence
The field as your buyers see it: who is in it, where the positions are, and where the opening is.
Product & concept
Whether it is ready — before the spend, not after. Concepts, packs, claims, interfaces and advertising.
Brand
Where the brand stands, where it is losing people, and what it would take to move it.
Pricing
What the market will bear, and which price maximizes revenue rather than volume.
Satisfaction & NPS
What customers think, tracked at a scale that settles the argument.
Implicit association testing
Reaction-time tasks that surface associations a direct question would let a respondent edit before answering.
Dial testing for advertising and video
Continuous, second-by-second response captured while an audience watches an ad or video, pinpointing exactly where attention rises or drops.
Range and assortment optimisation
Which combination of products or SKUs to offer, and in what range, to maximise revenue without one cannibalising another.
Claims and messages testing
Which product claim or marketing message lands hardest with a given audience, tested head-to-head rather than judged in a room.
Ranking and trade-off exercises
Forces a real choice between options rather than a rating of each in isolation, surfacing what a buyer would actually give up for what.
Segmentation typing
A short tool that assigns any respondent to an existing segment from a handful of questions, so a segmentation keeps paying off long after the original study.
Eye-tracking studies
Where attention actually goes on a pack, an ad or a shelf — measured, not inferred from what someone says they noticed.
Central location and in-home usage tests
A product placed in a controlled venue or in someone's home for a defined period, then evaluated on real use rather than a first impression.
What happens to the data
The analysis is where the answer actually comes from.
Fieldwork produces data. What you are paying for is what happens next, and it goes five rungs further than most buyers are shown.
Run in-house, not subcontracted. Instrument design and programming, data processing and respondent recruitment happen on our own tooling — which is why the person who designed the questionnaire is the person who can tell you what the cross-tab means.
1 Description
Frequencies, cross-tabulations and significance testing — what happened, cut by every sub-group you need to read separately.
2 Relationships
Bivariate analysis: what moves with what, and how strongly. The point at which one number starts to explain another.
3 Structure
Multivariate work — multiple regression, factor and correspondence analysis, cluster and discriminant analysis — to find the segments and drivers a cross-tab cannot show.
4 Causation
Experimental and quasi-experimental designs: controlled comparisons that let you claim a cause instead of a correlation.
5 Prediction
Price and demand modelling, time series and predictive scoring — where the analysis stops describing the market and starts forecasting it.
Methods, by name
- Van Westendorp
- A price-sensitivity meter. Four questions establish the range buyers consider too cheap, a bargain, expensive, and too expensive — mapping an acceptable price band.
- Gabor-Granger
- Tests purchase intent at a series of specific price points to estimate a demand curve and the revenue-maximizing price.
- Brand-Price Trade-Off
- Puts competing brands at moving prices in front of a buyer and records what they actually choose, so price is read against the competitive set rather than in isolation. It answers what happens to your share when a rival discounts, which a single-brand price test cannot.
- Conjoint analysis
- Presents realistic trade-offs between features and price to isolate what buyers actually value, rather than what they say they value. It answers which features carry real value, what buyers will pay, which concept wins against which, where the business would come from, and what a “what if” scenario does to share.
- Brand funnel
- Measures the path from awareness through consideration and preference to use and loyalty, showing exactly where the brand is losing people.
- Segmentation and clustering
- Groups a market by what genuinely differentiates behaviour rather than by demographics alone, using cluster and discriminant analysis on real response data. The output is a set of segments you can find and reach again, not a set of personas.
Book a call with a senior researcher
Tell us the decision you're facing. A senior researcher will help you diagnose it — no sales pitch, and no obligation to run a study.
Questions we get
About competitive research
Which pricing method should we use?
It depends on the decision. Van Westendorp maps an acceptable range, Gabor-Granger estimates the revenue-maximizing point, and conjoint isolates what buyers trade off. We pick the method after we understand the decision — that is the diagnosis.
How large does the sample need to be?
Large enough to support the claim you need to make, and no larger. Sample size follows from the decision and the sub-groups you need to read, not from a default.
Do you run the fieldwork yourselves?
Yes. Fieldwork is run in-market and in-language with real quality control, rather than subcontracted and passed through.
Which of these have you actually run?
Every figure on this page comes from our own studies ledger, and the ledger is published on this site rather than summarised. Where something is a capability rather than a track record it sits in the platform section and says so plainly. If a line has no number next to it, that is deliberate.
What is a tracking study, and when is one worth it?
The same instrument and the same sample design, repeated on a fixed cadence, so a change in the number means a change in the market rather than a change in the method. It is worth it when you need direction rather than level, and it is the largest single block of work we have done: 180 of 339 studies.
Tell us the number you need to defend.
A senior researcher will work out what a confident answer actually requires.
