# Brand Positioning Research: How to Find Your Place in the Category

How to research brand positioning with a survey: choose attributes, rate brands, build a perceptual map, find gaps and test a positioning statement.

Published 2026-10-16 · Updated 2026-11-16
Canonical: https://hercules.works/blogs/brand-positioning-research-india

Positioning research asks category buyers to rate your brand and its competitors on the same attributes, then compares the ratings on a map. Pick attributes from interviews, include three to five competitors, ask importance and performance separately and test the positioning statement you are considering. The gaps between what matters and what each brand delivers show where to stand. Hercules Works runs the survey on a verified Indian panel.

## Positioning Is What People Believe, Relative to Alternatives

A brand cannot choose its position alone. Customers place it relative to other brands, on the things they care about. Positioning research measures that placement.

This guide gives a step-by-step design, a question bank and a worked grid. The attribute list and numbers are illustrative.

## Five Steps

1. **Find the attributes.** Interview a few buyers: what do you look for in this category? Turn their words into 8 to 12 attributes.
2. **Pick the brands.** Your brand and three to five competitors.
3. **Ask importance.** How important is each attribute in choosing?
4. **Ask performance.** For each brand the respondent knows, rate it on each attribute.
5. **Analyse.** Compare importance with performance and map the brands.

Ask about brands only if the person has heard of them. Use an aided awareness list first, see [brand awareness](/blogs/brand-awareness-survey-india).

## Reading the Grid

Illustrative averages on a 1 to 5 scale. Importance is how much the attribute matters. The rows show where you lead and where you lag.

| Attribute | Importance | Your brand | Competitor 1 | Competitor 2 |
|---|---|---|---|---|
| Value for money | 4.5 | 3.2 | 4.1 | 3.6 |
| Quality | 4.4 | 4.3 | 4.0 | 3.5 |
| Trust | 4.2 | 3.9 | 4.2 | 3.1 |
| Modern | 3.1 | 4.1 | 3.3 | 2.9 |
| Available everywhere | 3.8 | 2.8 | 4.4 | 3.0 |

Read it across. Value for money matters most and you trail Competitor 1 by 0.9. "Modern" is where you lead, but it matters least. Positioning on "modern" alone would stress a low-importance attribute. The numbers are made up to show the reading.

## Perceptual Maps

A perceptual map plots brands on two dimensions made from the attributes. Build it in your analysis tool from the performance ratings. Choose the two dimensions that explain most of the variation, and label them with plain words. Treat a map as a picture, and test any move it suggests with a [claim test](/blogs/claim-testing-survey-india) or [concept test](/blogs/product-concept-testing-platform-india).

## Testing a Positioning Statement

Write two or three versions of the position, such as "the most trusted choice" or "the modern alternative". Show each to a separate matched cell and ask comprehension, believability, relevance and preference. Compare cells with a significance test. Keep the one that is both believed and relevant.

## From Grid to Priorities: the Weighted Gap

Take the illustrative grid above. For each attribute, subtract the best rival's score from yours, then multiply by importance. This sorts the rows so the biggest shortfalls come first.

| Attribute | Importance | Your brand | Best rival | Gap | Gap x importance |
|---|---|---|---|---|---|
| Value for money | 4.5 | 3.2 | 4.1 | -0.9 | -4.05 |
| Quality | 4.4 | 4.3 | 4.0 | +0.3 | +1.32 |
| Trust | 4.2 | 3.9 | 4.2 | -0.3 | -1.26 |
| Modern | 3.1 | 4.1 | 3.3 | +0.8 | +2.48 |
| Available everywhere | 3.8 | 2.8 | 4.4 | -1.6 | -6.08 |

Availability has the largest shortfall (-6.08), then value for money (-4.05). Availability is a distribution problem, so a message cannot fix it. Value for money is a position you could contest, but only if you can back it. Quality is a lead on a high-importance attribute, but the gap to the best rival is 0.3. Check that it is bigger than the noise before you build on it.

The weighted gap is a sorting aid, not a measured quantity. Report the plain averages and the base size with it.

## Question Bank for a Positioning Survey

| # | Question | Type |
|---|---|---|
| 1 | When you choose a [category] brand, what matters most to you? | Open text, asked first |
| 2 | How important is each of these when choosing a [category] brand? | Grid, 5-point scale, rotated rows |
| 3 | Which of these brands have you heard of? | List with decoys |
| 4 | How well does [brand] do on each of these? | Grid, only for brands known |
| 5 | Which brands would you describe as [attribute]? Select all that apply | Pick-any grid |
| 6 | In a few words, how would you describe [brand] to a friend? | Open text |
| 7 | Which of these brands would you consider buying next time? | Multi-select |
| 8 | Imagine the ideal [category] brand. What would it be like? | Open text |

Ask question 1 before the attribute grid. Once people see your list, they stop naming their own words.

## A Template and Scorecard for Positioning Statements

A common way to write a position is: "For [target buyer] who [need], [brand] is the [category frame] that [main benefit], because [reason to believe]." Write each version in that form, then show only the customer-facing sentence to respondents.

| Version | Understood as intended | Believable (top 2 box) | Relevant (top 2 box) | Preferred | Base (n) |
|---|---|---|---|---|---|
| A: the most trusted choice | | | | | |
| B: the modern alternative | | | | | |

A version that wins preference but is not believed is a risk. Fix the reason to believe or drop it. Test the pair with a [significance test](/blogs/statistical-significance-testing-india) and check that each cell is large enough for the gap you care about.

## Common Mistakes and Why They Bias the Result

- **Attributes written by the brand team.** Words from your own deck may not be words buyers use. Draw them from interviews.

- **A halo effect.** People who like a brand rate it high on everything. Compare gaps between attributes, not raw scores, and use the pick-any question as a cross-check.

- **Too many attributes.** A grid of 30 rows invites straight-lining. Keep to 8 to 12.

- **Rating unknown brands.** Someone who has only heard a name will guess. Rate only brands the person knows.

- **Reporting only the map.** Two dimensions hide the rest of the variation. Show the grid beside it.

- **Mixing positioning and communication.** The survey shows what people believe now. Whether an ad can change that is a separate [ad test](/blogs/ad-testing-platform-india).

## Running This Study on Hercules

If your job is a positioning study among category buyers in India, this is the pilot and main study at list price. Use matched cells for the positioning statements.

Pricing is per response, and the free plan covers a 30-response pilot: see [pricing](/pricing). How respondents are verified and sampled: [panel methodology](/panel-methodology).

The limit: the app reaches smartphone users only, so offline households need face-to-face fieldwork.

## Frequently asked questions

### What is brand positioning research?

A survey that asks category buyers to rate your brand and competitors on the same attributes, so you can see where each stands and where there are gaps.

### How many competitors should I include?

Three to five is a common choice, so respondents are not overloaded. Choose the brands your buyers actually consider.

### How do I choose attributes?

Interview a few buyers about what they look for in the category, and turn their words into 8 to 12 attributes.

### What is a perceptual map?

A chart that places brands on two dimensions built from attribute ratings, showing which brands are seen as similar and where gaps exist.

### How do I test a positioning statement?

Show each version to its own matched cell and compare comprehension, believability, relevance and preference.

### How is positioning research different from brand tracking?

Positioning compares how brands are seen on chosen attributes at one time. Tracking repeats a fixed set of measures to show change.

### Should I ask for importance and performance separately?

Yes. People often say everything matters when asked once. Rating importance on its own lets you weigh the gaps.

### What is a pick-any question?

A question that asks which brands fit an attribute, with no limit on how many. It is quick and gives a second view beside the rating grid.

### Can I test positioning with a small sample?

For a rough read, yes. For a gap between brands, check the margin for each cell, because panel samples are not simple random samples.
