Perceptual mapping in marketing: how to build one, the two axes to choose, and how to read the empty space

A perceptual map is a chart that plots how buyers perceive the brands in a category, using two attributes that influence their choice as the axes. Each brand sits where its buyers place it, not where its marketing team places it. The output is a picture of the competitive set as the market sees it, which is why it is the standard tool for finding an unoccupied position and for testing whether the position you claim is the one you hold.

The word perceptual is the point. The data comes from buyers rating brands, so the map shows reputation, not specification. A brand can be objectively the fastest in its category and sit in the middle of the speed axis because nobody knows.

Choosing the two axes, which is the whole job

A bad pair of axes produces a tidy chart that changes nothing. Three rules keep a pair useful.

• Each axis must drive choice. Use attributes buyers actually trade off: price against durability, speed against thoroughness, breadth against specialism. Attributes everybody expects, like safety in a hospital or hygiene in a restaurant, are entry requirements and compress every brand into one corner.
• The two axes must be independent. Quality against price often collapses into one line because buyers read them as the same thing. If your scatter forms a diagonal, swap one axis.
• Both must be stated in buyer language. Not modularity but how easy it is to change later. Not omnichannel fulfilment but whether I can collect it today.

Common pairs that work: price level against perceived expertise, breadth of range against depth of advice, speed of delivery against customisation, premium feel against everyday usability, local presence against scale.

How to gather the data, three ways

Method	What you do	Strength	Cost in effort
Attribute rating survey	Show 6 to 10 named brands, ask buyers to rate each on both attributes on a 1 to 7 scale, then average	Direct, easy to repeat each year, simple to chart	Moderate, needs a usable sample per brand
Similarity sorting	Ask buyers to group brands by how alike they are and explain the groups, then derive the axes from their reasons	Finds the attributes buyers really use, including ones you would not have listed	Higher, needs interviewing and coding
Evidence mapping	Score each brand from what is publicly said about it: reviews, roundups, pricing pages, analyst notes	Fast, repeatable, works when you cannot field a survey	Low, and it measures what the market is told rather than what it believes

A small sample used honestly beats a large sample used to confirm a decision already taken. Twenty five buyers rating eight brands will tell you something real. The same twenty five asked leading questions will tell you what you wanted to hear.

Drawing it, step by step

• List the brands a buyer genuinely considers, including the do nothing option where that is a real alternative.
• Pick your two attributes and define each end of each scale in one sentence, so every rater reads them the same way.
• Collect ratings, average per brand per axis, and plot the points.
• Size each point by market share or by the share of your pipeline it appears in, so the chart carries weight as well as position.
• Add your own target position as a hollow marker, so the gap between where you sit and where you intend to sit is visible.
• Write one sentence under the chart saying what decision it supports. A map with no decision attached is decoration.

Reading the map, including the empty space

The instinct is to aim at whitespace. Most whitespace is empty for a reason, so test it with three questions before you move.

• Does anybody want this combination? Cheap and highly bespoke is empty because the economics do not hold. Demand has to exist before position matters.
• Can you deliver it and prove it? Moving on a perceptual axis means changing the evidence buyers see: pricing, packaging, guarantee wording, case studies, the people you put in front of them.
• Can you hold it? A position a larger competitor can copy next quarter is a campaign, not a position.

Clusters are as informative as gaps. Three brands on top of each other means buyers cannot tell them apart, so in that cluster the purchase is decided by price, availability or whoever answers the phone first. If you are in the cluster, the useful move is often a sharper offer on one axis rather than a slow drift toward the middle.

What a map will not do, and what to use instead

A perceptual map compares brands on two attributes at one moment. It does not tell you why buyers rate you as they do, and it does not describe the segments inside the market. Pair it with two other things: short interviews for the why, and a written positioning statement for the decision. The map shows the terrain, the statement commits you to a hill. The four parts of a positioning statement and the tests for one are set out in positioning in marketing explained, and the segment side is worked through in segmentation, targeting and positioning.

Redraw the map annually, or after any launch, repricing or competitor entry that changes the set. Keep every version, because the movement between two maps is more persuasive to a board than either map on its own.

Perceptual mapping questions that come up after the workshop

What is the difference between a perceptual map and a positioning map?

In practice the terms are used for the same chart. If anybody distinguishes them, a perceptual map is built from buyer ratings and a positioning map is built from the position a company intends to occupy. Plot both on one chart and the distance between them is your communication gap.

How many brands should go on a map?

Six to ten. Fewer and the competitive structure is invisible, more and the labels overlap and nobody reads it. Include the two substitutes buyers mention that you would not call competitors.

Can I build one without research budget?

Yes. Score each brand from public evidence: review site themes, pricing pages, the language in their top three pages, and what comparison articles say. Label it as an evidence based map so nobody mistakes it for survey data, and treat it as a hypothesis to test in your next ten sales calls.

Should price be one of the axes?

Only if price is genuinely a differentiator in your category and you can plot perceived price rather than list price. The more useful version is perceived value for money, because two brands at the same price often sit far apart on it.

Can a map have more than two axes?

Statistical techniques place many attributes in two dimensions at once, and the output is harder to act on. For a working team, two clear axes and a second chart with a different pair beats one clever chart nobody can interpret.

How do I use it with a sales team?

Give them the map with one sentence per competitor: what that brand is believed to be good at, and the question to ask that makes the difference visible. That is the practical output, more than the chart itself.

The map a machine is already drawing of your category

Ask an AI assistant to compare the options in your category and it will produce a perceptual map in prose. It names a handful of brands, says which is the premium choice, which is the budget one, which suits a small team and which suits an enterprise, and it assembled all of that from reviews, comparison articles and roundups rather than from a survey. That answer is a live reading of category perception, available in minutes, and it is also the version a buyer sees before anybody shows them a chart. Worth comparing against your own map, because where the two disagree you have found either a research gap or a reputation problem. AI Knows Us captures those category answers over time, so a brand can see where it is being placed, what it is being placed against, and which third party pages put it there.