Brand perception map: choose the positioning test it suggests

Use a brand perception map to choose a specific positioning test. Consider Lumen, an example tea brand seeking a stronger premium impression without losing convenience, alongside tea alternatives Harbor and Plainfield. The illustrative ratings below suggest testing a credible premium cue. Choose buyer-relevant attributes, retain knowledge states and valid bases, and plot the summaries on fixed axes; empty space alone does not show demand.

Make the brand points comparable

Ratings and plot coordinates

These means provide a transparent two-attribute map. The scale is 1, does not fit at all, to 5, fits extremely well. Not enough knowledge is excluded from the rating mean and reported separately in a real study. Each dimension has 100 ratings per brand in this example; the summaries alone do not establish precision.

Use the same question, endpoints and offer version for each brand. Know who can rate it. A brand evaluated mainly by loyal users can appear stronger than one familiar to a broad mixed audience. The map should preserve that coverage question rather than imply all dots share identical evidence.

Use the copyable CSV and original SVG source below to reconstruct the result. The table and drawing use these original means directly without adding measurements.

Translate the contrast into a positioning question

Lumen has the highest convenience mean, 4.2, and a premium mean of 3.0. Harbor is lower on convenience at 3.1 and higher on premium at 4.4. Plainfield supplies a further category reference at 3.6 and 2.8. Those contrasts help choose what to investigate, while their causes remain unknown.

A specific question follows: can a truthful packaging cue strengthen premium meaning among the intended Lumen buyers without sacrificing their valued convenience? This preserves an existing association as a guardrail instead of treating the map as permission to move toward any empty area.

Choose a cue the product can deliver. An ingredient detail, feasible presentation or verified experience may be relevant. An attractive premium claim without a physical basis can raise an association score while creating a promise the offer fails to fulfill.

Empty space does not establish unmet demand. It may reflect low need, infeasible delivery or a missing important dimension. A brand moving a dot also need not gain customers. Associations are one part of the buying situation.

Define the associations before optimizing them

Convenience can mean availability, portability, preparation or easy disposal. Premium can mean ingredients, presentation, occasion or price. Ask the audience what the terms evoke before deciding which mechanism the next cue should change.

If the team wants portable tea while respondents interpreted convenience as grocery availability, a packaging change may not address the rated meaning. Use the interpretation to revise the measure or the hypothesis. A precise axis cannot compensate for an ambiguous construct.

This is a map of two declared attributes, rather than a multidimensional scaling model or discovered latent position. That makes its construction easy to inspect. Choose a different analytical approach when the job is to discover unknown dimensions of competitive similarity.

Keep axes fixed across waves. Changing from a 1–5 scale to brand-relative minimum/maximum endpoints makes the visual movement partly an effect of the plotting convention. The same numbers can then appear farther apart or closer together.

Read the bases and distributions behind the dots

A mean can summarize broad moderate agreement or polarized responses. Similar means need not describe similar audiences. Keep the full category distribution and meaningful segments available, especially when a cue targets one group.

Compare familiarity and knowledge exclusions. Not enough knowledge is not a neutral association. If one brand has many unanswerable ratings, that is a different evidence base and possibly an awareness question, rather than a reason to put its missing records at the center.

The example records contain means and valid bases, not individual distributions. They cannot produce uncertainty circles or prove a segment difference. Draw those only from appropriate actual records and analysis.

Preserve wording, sampling, weights and exclusions with the map. Those details determine whether two dots or two waves can be meaningfully compared. A visually clean chart can otherwise conceal a change in who supplied the rating.

Reconstruct the normalized map

The original SVG normalizes each dimension with (mean−1)/4. Lumen becomes (0.8, 0.5); Harbor (0.525, 0.85); Plainfield (0.65, 0.45). The direct 1–5 plot shows the same ordering and source values.

Fixed endpoint normalization is a transparent scale transformation. It adds no latent dimension, no respondents and no validated measure of geometric competitive distance. Keep the original ratings beside the coordinates so a reviewer can reconstruct every point.

Report any later movement in the original 1–5 units, alongside the meaning buyers attach to it. A higher premium rating could describe a valued taste, gift suitability or simply a higher expected price. The follow-up must identify which meaning the cue created before the team adopts the position.

Copy the source records and drawing

The CSV retains the original field names and all three records. Its evidence_type field is fictional_teaching_data for every row.

Complete brand-map CSV

brand,convenience_mean,premium_mean,valid_convenience_n,valid_premium_n,scale_min,scale_max,normalized_convenience,normalized_premium,evidence_type
Lumen,4.2,3.0,100,100,1,5,0.8,0.5,fictional_teaching_data
Harbor,3.1,4.4,100,100,1,5,0.525,0.85,fictional_teaching_data
Plainfield,3.6,2.8,100,100,1,5,0.65,0.45,fictional_teaching_data

Copy the SVG into a plain-text document with an .svg extension to open or edit the drawing. It contains the same fixed 0–1 axes and the three coordinates shown above.

Complete SVG drawing source

<svg xmlns="http://www.w3.org/2000/svg" width="640" height="480" viewBox="0 0 640 480" role="img" aria-labelledby="title desc">
  <title id="title">Fictional tea-brand perception map</title>
  <desc id="desc">Convenience and premium impression normalized to fixed zero-to-one scales. Fictional Lumen is 0.8 and 0.5, Harbor 0.525 and 0.85, Plainfield 0.65 and 0.45. This is a plot of declared invented means, not an estimated market model.</desc>
  <rect width="640" height="480" fill="#fff"/>
  <g font-family="Arial, sans-serif" fill="#172d39">
    <text x="70" y="26" font-size="19" font-weight="bold">Fictional brand perception map</text>
    <g stroke="#e0e7eb" stroke-width="1">
      <path d="M70 50H530M70 130H530M70 210H530M70 290H530M70 370H530"/>
      <path d="M70 50V370M185 50V370M300 50V370M415 50V370M530 50V370"/>
    </g>
    <path d="M70 50V370H530" fill="none" stroke="#172d39" stroke-width="2"/>
    <g font-size="13" text-anchor="middle">
      <text x="70" y="391">0</text><text x="185" y="391">0.25</text><text x="300" y="391">0.5</text><text x="415" y="391">0.75</text><text x="530" y="391">1</text>
      <text x="300" y="419" font-size="16">Convenience fit</text>
    </g>
    <g font-size="13" text-anchor="end">
      <text x="58" y="375">0</text><text x="58" y="295">0.25</text><text x="58" y="215">0.5</text><text x="58" y="135">0.75</text><text x="58" y="55">1</text>
    </g>
    <text x="23" y="220" font-size="16" text-anchor="middle" transform="rotate(-90 23 220)">Premium impression fit</text>
    <circle cx="438" cy="210" r="7" fill="#13766b"/>
    <text x="450" y="200" font-size="14">Lumen (0.8, 0.5)</text>
    <circle cx="311.5" cy="98" r="7" fill="#99601c"/>
    <text x="324" y="88" font-size="14">Harbor (0.525, 0.85)</text>
    <circle cx="369" cy="226" r="7" fill="#6b50a5"/>
    <text x="274" y="253" font-size="14">Plainfield (0.65, 0.45)</text>
    <text x="70" y="448" font-size="13">Invented brands and means; no human study or Sapien prediction.</text>
    <text x="70" y="467" font-size="12">Coordinates = (raw mean - 1) / 4; fixed endpoints 0 and 1.</text>
  </g>
</svg>

Design the matched cue test

Compare current Lumen presentation with one feasible revised cue. Hold product facts, price, exposure and other visual treatment constant unless they are deliberately part of the change. Collect unaided interpretation before the association ratings so a score can be connected to the product people imagined.

Read the premium and convenience distributions with their bases. A higher premium mean is incomplete if important buyers now imagine difficult preparation or a less portable pack. Add consideration or an appropriate choice when the progression decision requires more than associations.

Define the guardrail in words and evidence before collection: which convenience meaning must survive, for whom and under what use occasion? A numerical condition can follow from the actual business requirement and design, rather than a universal percentage selected after seeing the result.

Progress a credible cue, revise a false inference

Progress when the cue strengthens the intended meaning, preserves the valued convenience and is deliverable. Revise when premium depends on a false product inference. Reconsider the territory when the associations conflict for the audience the brand can reach.

The map supplies a hypothesis and a matched test, not a full repositioning launch decision. Keep product experience, audience evidence and commercial feasibility with the recommendation. Modeled ratings retain their grounding and validation requirements just as collected ratings retain their sampling limits.

Further reading

• our brand perception guide
• our positioning territories guide

Bring the decision, audience and evidence needed to a scoped Sapien study discussion.