If customers do not use your chosen category name, first anchor the product in the problem, familiar alternatives and outcome they already recognise. Teach the new term only when it captures a genuinely useful distinction, buyers can learn it without losing the immediate meaning, and the business can sustain the education. Do not assume low use means rejection, and do not assume a clever label creates a market.

The decision requires evidence from what people do as well as what they say. Search data, interviews, sales calls and product comprehension each reveal a different part of the picture. A good category strategy makes the product easier to place and compare; it does not require the buyer to solve a vocabulary puzzle before understanding the value.

Diagnose the mismatch before changing the name

Several different problems can produce the same observation:

  • Customers recognise the problem but describe it with another term.
  • Different buying roles use different names for the same work.
  • People search by symptom, task or incumbent rather than by category.
  • The category exists in another market but not this segment or country.
  • The proposed category combines jobs buyers do not consider connected.
  • Buyers understand the phrase but see no reason for it to change the decision.
  • The product team invented a term mainly to avoid an unfavourable comparison.

Each diagnosis suggests a different response. Replacing a label will not fix an incoherent product scope. Teaching a term can be worthwhile when it helps buyers name an emerging problem or compare previously fragmented options. Familiar language is safer when the new term adds no decision value.

The US Small Business Administration's guidance on market research and competitive analysis advises examining demand, alternatives, saturation and pricing. Category-language research belongs inside that larger market inquiry. Vocabulary alone is not evidence of willingness to buy.

Observe how the work is described in context

Ask people to show or reconstruct a recent occasion when the problem occurred. Listen for the names of tasks, documents, teams, tools, failures and outcomes. Note what they call the current alternative and which words appear only after the researcher introduces them.

GOV.UK guidance on contextual research and observation recommends observing activities in the environment where they happen, using real tools, documents and devices where appropriate. This reduces dependence on abstract answers. A buyer may agree that a label sounds sensible while never using it to find, budget or explain a solution.

Capture exact language with context and consent. One striking quotation is not a category decision. Look for repeated patterns across the relevant segment and include non-buyers or lost opportunities so the vocabulary is not limited to people already educated by your sales process.

Build the category-language evidence ladder

The ladder ranks evidence by how closely it connects the term to a real decision. Higher levels do not automatically invalidate lower ones; they answer stronger questions.

LevelEvidenceWhat it can tell youWhat it cannot prove
1Internal preferenceThe term fits the team's intended storyBuyer recognition or demand
2Industry and competitor usageThe term appears in the marketThat your segment uses or trusts it
3Unprompted customer languagePeople independently use the termSearch frequency or purchase intent
4Prompted comprehensionPeople can explain the term accurately after seeing itThat they would use it unaided
5Search and navigation behaviourThe term or related phrases help people find relevant informationMarket size or causation
6Buying-process evidenceThe term appears in briefs, budgets, shortlists or referralsThat the label alone caused the purchase
7Controlled message behaviourA matched test changes qualified action without increasing misunderstandingLong-term category ownership

Do not convert the ladder into a universal scoring model. Its purpose is to show the distance between "we like this phrase" and "buyers use this frame in a consequential decision".

Use interviews that do not plant the answer

Begin with open prompts:

  • Tell me about the last time this problem happened.
  • What did you call the work internally?
  • What did you search for or ask other people about?
  • Which options did you compare?
  • What budget or project did the purchase belong to?
  • How did you explain the decision to a colleague?

Only after capturing unprompted language should you show the proposed term. Then ask what the person thinks it includes, excludes and replaces. Avoid "Do you like this category?" because liking is not recognition or purchase evidence.

Pew Research Center's guidance on writing survey questions explains that wording, question order and response options can influence answers. Although positioning interviews are not representative surveys, the warning is directly relevant: introduce your category term too early and you may manufacture the vocabulary you intended to measure.

If you later use a survey, pretest the wording and report the sample, recruitment, geography and question order. Do not present a convenient customer email poll as population evidence.

Read search data as a partial signal

Search behaviour can reveal the language people type, but every tool has scope and limitations. Google Search Central's guide to using Search Console and Google Analytics data explains how site owners can examine queries, pages, countries, clicks and impressions alongside on-site behaviour. Search Console reflects a property's visibility in Google Search, not all market demand, and a new category page cannot reveal demand it has never had a chance to earn.

The official Google News Initiative lesson on the basics of Google Trends explains how to compare topics or terms, read location and related-query data, and interpret indexed and normalised values. It also explains that Trends draws on a sample rather than providing exact query totals and excludes some data. Treat it as one directional signal, not a scientific poll or an estimate of absolute market size.

Use search evidence to form and challenge hypotheses:

  • Compare the new term with problem, task and alternative language.
  • Segment by country and time where sufficient data exists.
  • Inspect related queries and the pages currently shown.
  • Check whether impressions lead to qualified behaviour, not only visits.
  • Record tool settings and date so the comparison can be repeated.

Do not translate relative trend interest into an absolute market-size claim. Low volume can mean a small market, a new term, fragmented phrasing or poor tool coverage.

Choose one of four language strategies

Once the evidence is assembled, choose deliberately:

  1. 1 Use the familiar category. Best when buyers recognise the term and the product's difference can be expressed inside it.
  2. 2 Use a familiar anchor plus a modifier. Describe the known category, then name the specific difference: "X for Y" or "X with Z", provided the modifier is accurate.
  3. 3 Use a bridge phrase. Lead with the problem or known alternative, then introduce the proposed category as a concise organising term.
  4. 4 Teach a new category. Use when the term captures a coherent new comparison, materially helps the buyer and the business can invest in education, evidence and ecosystem adoption.

The Strategyzer Value Proposition Canvas can help check whether the proposed category corresponds to a defined segment's jobs, pains and gains rather than to a collection of product features. The category name should make the customer value easier to see, not substitute for it.

Calculate the education burden

A new category introduces work. Buyers must learn what it means, whether the problem deserves priority, which budget owns it, how to compare options and why existing solutions are insufficient. Salespeople, partners, analysts and customers also need consistent language.

List the burden explicitly:

  • definitions and examples;
  • comparison with familiar alternatives;
  • evidence of the problem and outcome;
  • sales and partner training;
  • search and editorial content;
  • objection and procurement support;
  • repeated use across product and market touchpoints.

Then ask whether the expected strategic advantage justifies that burden. Category education may be rational when the old frame systematically hides the product's value or forces a false comparison. It is weak when the new phrase mainly sounds more modern.

Design a bridge that works in one reading

A bridge should answer three questions quickly:

  1. 1 What familiar problem or job is this about?
  2. 2 What known alternatives frame the choice?
  3. 3 What useful distinction does the new term introduce?

One structure is: "For [customer] dealing with [recognisable problem], [product] is a [familiar anchor or plain description] that [specific outcome]. We call this [new term] because [meaningful boundary]."

The new term comes after meaning, not before it. If the sentence cannot explain the distinction without circular language, the category may not yet be coherent.

GOV.UK's guidance on writing for user interfaces recommends clear language and using the same terms as users. This does not forbid specialist vocabulary. It makes the team responsible for proving that the vocabulary helps the intended audience.

Illustrative example: teach the term after the job

Illustrative example - the product and observations are hypothetical.

A software company calls its offer "decision continuity management". Research participants never use the phrase. They describe missed context when projects change hands, and they compare shared documents, project tools and extra meetings. When shown the new term, they can understand it, but only after an explanation.

The company does not abandon the concept immediately. It leads with the familiar hand-off problem and existing alternatives, then uses the term as a label for its method. The message test measures whether people can explain the distinction and choose an appropriate next step. If the term adds no recall or decision value over time, it remains internal strategy language rather than the public category.

This preserves a useful idea without making comprehension conditional on prior vocabulary.

Test for accurate action, not mere recall

Create matched message versions using the familiar term, a bridge and the proposed new category. Test with comparable participants or traffic. Measure:

  • accurate description of the product;
  • correct inclusion and exclusion of use cases;
  • appropriate comparison set;
  • memory after a delay;
  • quality of the next question;
  • qualified next action;
  • misunderstanding and false expectation.

A memorable but misleading term is not a win. A familiar term can generate more traffic but weaker fit. A new term can generate fewer responses yet help the correct buyers self-select. Interpret the evidence against the actual commercial decision.

Limitations and trade-offs

Buyer language varies by role, industry, country and maturity. Interview samples reveal meanings but not prevalence, while search platforms show only the behaviour within their data and definitions. Message tests can be affected by audience, channel, creative execution and novelty. Teaching a category can create differentiation but consumes time and budget; using familiar language improves immediate recognition but may preserve an unhelpful comparison. Record the evidence date and avoid declaring category ownership from small or short-term tests.

Further reading

Sources

Your useful next step

Take ten recent sales, research or support records and extract the unprompted words used for the problem, task, alternative and outcome. Place each source on the evidence ladder. Then draft a familiar version, a bridge version and a new-category version, and test whether representative buyers understand the scope and take an appropriate next step without being taught by the interviewer.

Publication note

Publication note: This article provides a positioning and research framework, not a claim about search demand or market size. Search and qualitative evidence have material limits. Protect participant data, document methods and settings, and verify current platform guidance before using behavioural data to make a category investment.