Brand research: match each question to the cheapest method

Brand research is any study of what people in your market know and believe about your brand: whether they have heard of you, what they think you stand for, whether you come to mind when they need something, and how much your name changes their response to an offer. Each brand research method answers one of those questions well and the others badly, so pick the question first and the method second.

For a small B2B company, most of this costs time more than money. A few interviews, a short survey and a filter in Search Console cover the first three questions. The fourth, brand equity, needs a small experiment, and it is the one most teams skip.

Brand research measures four things, and two academics named most of them

Kevin Lane Keller's 1993 paper in the Journal of Marketing splits what buyers know about a brand into two parts. Brand awareness is recall (naming the brand when given only the category) plus recognition (confirming you have seen it before when shown the name). Brand image is the set of associations people hold about the brand, which Keller says vary in how favorable, strong and unique they are.

Keller then defines customer-based brand equity as the difference that knowledge makes. A brand has positive equity when people respond more favorably to its marketing than to the same marketing from a fictitiously named or unnamed version of the product. In plain terms: would they react the same way if your logo were missing?

David Aaker's model, as summarized by WARC, groups brand equity into five categories of assets: brand loyalty, brand awareness, perceived quality, brand associations and other proprietary assets such as patents and trademarks. Aaker's list is useful as a checklist. Keller's definition is more useful for research, because it tells you what to compare.

Here is how the questions map to methods.

A survey can tell you how many buyers recognize your name. It cannot tell you that many of them think you sell payroll software.

1. Start with interviews to learn what your brand means to buyers

Run interviews before any survey, because you cannot write good answer options for associations you have not heard yet. Keller recommends free association tasks where people describe what a brand means to them in their own words, probed with who, what, when, where, why and how questions. He also suggests projective techniques such as sentence completion for people who cannot or will not say what they feel.

Talk to eight to twelve people who buy in your category, and include people who have never bought from you. Ask about the last time they chose a vendor before you mention your company. Then ask what they know about you, what kind of company they think uses you, and to finish a sentence such as "Teams that use [brand] are usually..." (It feels like a party game, which is the point: Keller suggests these techniques for people unwilling to say what they think directly.)

Write down their exact words. You will reuse them in step 2, and buyer phrasing beats your team's phrasing for every later survey. If you already run win-loss interviews, the recordings probably contain brand perception data nobody has coded yet.

2. Turn interview language into buying situations, then count them

Awareness alone misses the point in B2B, because buyers rarely shop for a brand. They shop for a fix to a situation. Jenni Romaniuk of the Ehrenberg-Bass Institute calls these situations category entry points: the cues buyers use to search their memory when a buying need comes up. For a finance software buyer, one might be "need to close the books faster at year end" and another "has to work with our existing accounting system". Her paper for LinkedIn's B2B Institute says brands that come to mind from memory are the starting point for the buying process, and buyers consult search engines or colleagues mainly when that list falls short.

Two steps of her process matter here. First, collect entry points using her prompts about when, where, why, with what and with whom the category gets bought, which is where your interview notes come in. Second, survey category buyers with each entry point and a list of brands underneath, plus a "none" option, and let them tick every brand they link to it.

The same paper defines three numbers you can calculate from that survey:

• Mental penetration: the share of buyers who link your brand to at least one entry point. In its UK business banking example, 472 of 609 customers linked Barclays to at least one, or 78%.
• Network size: the average number of entry points linked to your brand among those people.
• Mental market share: your share of all brand links across all entry points and brands.

Expect modest numbers. John Dawes of the same institute writes that many well-established brands reach only 20% to 30% of respondents linking them to an entry point, and market leaders often only about 50%.

3. Track awareness on a fixed schedule, because buyer memory moves slowly

A tracking survey asks the same questions, in the same order, to the same kind of audience, wave after wave. Our brand awareness survey guide covers the question wording, order and sample sizes, so it is not repeated here. Add the entry point grid from step 2 to that survey once you have it.

Run it once or twice a year. Dawes estimates in his 95-5 rule paper that companies change providers such as their main bank or law firm about once every five years, so only about 20% of business buyers are in the market in a year and something like 5% in a quarter. He says the 95% figure is a heuristic, not a precise rule, and that building mental availability to double digits is a multi-year task. A monthly tracker in a slow category mostly measures noise and your patience.

Dawes also gives a cheap way to size your own category: ask buyers how often they purchase it. In his example, a two-year buying cycle means about 50% of buyers are in the market over a year and around 13% in any quarter.

4. Measure brand equity with a blind test before a pricing or naming decision

Keller's direct approach is an experiment. One group responds to an offer attributed to your brand, a second group responds to the identical offer from a fictitiously named or unnamed product, and the gap between them is the brand's effect. He notes that written concept statements work when people cannot see the real product without noticing the brand, and that conjoint analysis can measure how a brand name interacts with price and features.

Here is a made-up example. Ledgerline sells accounts payable software to mid-sized companies and wants to know whether its name is worth anything at its planned price of $450 a month. It shows 60 finance managers the same one-paragraph description at that price. Thirty see it as "Ledgerline". Thirty see it as "an accounts payable tool". Both groups rate how likely they are to book a demo on the 0 to 10 Juster probability scale, which the Romaniuk paper uses to ask business buyers how likely they are to switch providers. The branded group averages 6.1 and the unbranded group 4.9, so the name adds 1.2 points at that price.

Thirty people per group is small, so treat the gap as a direction to check, and run it again before betting the price list on it. And if almost nobody in the sample has heard of you, both groups will answer about the same. That is a real finding: outside your customer base, the brand is not yet doing much work, and the message is carrying the sale. Message testing is the better next study in that case.

5. The small B2B version uses tools you already have

You do not need a research agency to start. A workable first year looks like this:

1. Watch branded search. Google's branded queries filter in Search Console, announced in November 2025, splits search traffic into queries that include your brand name and those that do not, using an AI-assisted classifier. Google says it helps measure brand recognition. It only works for top-level properties with enough search volume, and Google warns that some queries may be misclassified.
2. Interview ten category buyers using step 1, with at least half of them not customers.
3. Build an entry point list from those interviews and add it to a short survey of buyers in your category.
4. Run one blind test before your next pricing or naming decision.

Brands that sell to millions of everyday consumers have a different job, with panels and much bigger samples. Consumer research covers how to start that one.

Your next step

Write down which of the four questions you need answered this quarter, then book interviews with category buyers who have never bought from you. Instant Expert finds people who match a description you write, such as "finance managers at mid-sized companies who chose accounts payable software in the last two years", and sends them paid invitations for a short call or a written or voice answer; you pay only if someone books or answers. For help designing the study itself, see the directory pages for brand consultants and B2B marketing consultants.