Auditing Brand Distinctiveness

• Source: Report research library
• Frameworks applied: Ehrenberg-Bass Distinctive Brand Assets (Romaniuk & Sharp), Distinctive Asset Grid, category entry points, mental availability benchmarks (Dawes)

Refuses to score assets from internal opinion and converts an unmeasurable request into a test specification, with an explicit Freedom calibration section that is rare across the library.

Known weakness, published as found: Split the unbranded test-design spec and the Ehrenberg-Bass background into references/test-design.md linked from a one-line router, since it applies only when the user has no research data; add 2-3 committed eval prompts covering the 'no data exists' path.

1. 1.Copy the SKILL.md text below, or download the raw file.
2. 2.Create a folder named exactly auditing-brand-distinctiveness in your agent's skills directory.
3. 3.Save the file inside that folder as SKILL.md.
4. 4.Ask the agent one of the trigger requests below.
5. 5.Check the output against what you already know before it leaves your desk.

• Run a distinctive brand asset audit — which of our assets should we keep, sharpen or retire?
• We're planning a logo refresh. How do we test whether people actually link the current mark to us?
• Our brand colour is basically the same as two competitors'. Plot our assets on the fame and uniqueness grid.

• Who are our best-fit customers and what value themes should we lead with?
• Plan our Q3 demand generation budget across channels

---
name: auditing-brand-distinctiveness
description: >-
  Audits distinctive brand assets using the Ehrenberg-Bass method: inventories
  candidate assets, specifies unbranded recognition tests, scores each asset on
  Fame and Uniqueness, plots the Distinctive Asset Grid, and issues a
  protect/sharpen/build/retire decision per asset with a usage rule. Use when
  the user says "brand audit", "distinctive assets", "brand asset grid", "do
  people recognise our logo", "should we keep the mascot", "rebrand", "logo
  refresh", "our colour is the same as everyone else's", "brand recognition
  research", "sonic branding", or "which assets should we drop". Do NOT use for
  writing the product positioning canvas (see positioning-a-b2b-product), for
  category narrative and lightning strikes (see designing-category-narrative),
  or for allocating channel budget across brand and activation (see
  planning-demand-generation).
metadata:
  version: "1.0"
---

# Auditing brand distinctiveness

Produce one artifact: a Distinctive Asset Grid with a build, protect, sharpen or retire decision and a usage rule for every asset audited. Out of scope: designing new assets, visual identity execution, positioning, and media planning.

Score assets only on measured Fame and Uniqueness. If no test data exists, this skill's job becomes specifying the test — do not score assets from internal opinion, because internal familiarity is the single largest source of inflated Fame estimates.

## Workflow

Copy this checklist into your reply and tick items as you complete them:

```
- [ ] 1. Set objectives and success metrics
- [ ] 2. Inventory all current and candidate assets, including competitors'
- [ ] 3. Specify the unbranded test design
- [ ] 4. Ingest or collect Fame and Uniqueness scores
- [ ] 5. Plot the grid and assign quadrants
- [ ] 6. Issue a decision and usage rule per asset
- [ ] 7. Select 3-5 core assets and assign roles
- [ ] 8. Run the validation loop until it passes
- [ ] 9. Emit the audit
```

**1. Set objectives.** Clarify scope and align success metrics before testing anything: recognition lift, ad recall, shelf conversion, search conversion, or price realisation ([Umbrex framework reference](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)). State which metric the audit is meant to move, because the metric decides which contexts the assets must survive — an app-icon audit and a conference-booth audit reach different conclusions about the same logo.

**2. Inventory assets.** Distinctive assets are non-verbal or symbolic cues that make a brand recognisable at speed, at distance and across contexts: colours, shapes, packaging forms, type styles, characters and mascots, taglines, sonic cues, motion patterns, and sensorial cues such as scent, texture and haptics ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).

Include competitors' assets in the inventory, and document where and how each asset appears ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)). Without the competitor set you cannot interpret a Uniqueness score, because Uniqueness is a share of attributions and needs the other claimants named.

Treat a tagline as an asset **only if it has been tested** ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)). Untested taglines enter the inventory as candidates, never as scored assets.

**3. Specify the unbranded test.** These parameters are non-negotiable, because branded stimuli and biased prompts produce misleading results ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)):

- Show **unbranded** stimuli that isolate a **single** asset — a colour block, a pack silhouette, a sonic sting. One asset per stimulus, because a logo shown on branded packaging measures the packaging.
- Strip all brand names from the stimulus and from the question wording.
- Use a **representative category buyer sample**, not customers only. A customer-only sample measures loyalty, not distinctiveness.
- Ask which brand it belongs to, with no prompted list.
- Segment results by buyer group, usage occasion and market ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).

If the user cannot run this test, output the test specification as the deliverable and mark all scores as unmeasured. Do not substitute internal ratings, social listening or brand-tracker awareness questions; none of them measure asset-to-brand linkage.

**4. Compute the two scores.** Use these definitions exactly ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)):

- **Fame** = how many category buyers correctly link the asset to your brand when the brand name is not shown, expressed as the percentage of respondents who correctly link the cue to the brand.
- **Uniqueness** = how exclusively the asset is linked to your brand versus competitors — operationally, of all brand attributions for that cue, the percentage that go to your brand.

The distinction is load-bearing: an asset can have high Fame and low Uniqueness when many respondents name your brand but many also name a competitor, and that asset is actively helping the competitor every time you run it alone.

**5. Plot the grid.** Assign each asset to a quadrant ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)):

| Fame | Uniqueness | Quadrant | Action |
|---|---|---|---|
| High | High | Core assets | Protect and scale; non-negotiable in high-reach activity and at choice points such as shelf, search and app icon |
| High | Low | Popular but shared category cues | Sharpen, or pair with unique elements so they do not aid competitors |
| Low | High | Promising emergers | Invest to build familiarity; place prominently alongside established assets until Fame grows |
| Low | Low | Non-assets | Avoid clutter — redesign or retire |

Set the high/low thresholds explicitly before plotting and state them in the output, because an unstated threshold is where the audit gets argued away. Anchor the Fame threshold against the benchmark that many well-established brands achieve no higher than 20-30% of respondents linking them to a category entry point, and even market leaders often reach only 50% ([John Dawes](https://johndawes.info/the-955-rule/)) — so a 35% Fame score is not a failure in absolute terms, and a threshold set at 80% will retire assets that are working.

**6. Decide per asset.** Every asset gets one decision and one usage rule. The usage rule is the operational half; a decision with no usage rule changes nothing downstream. Write rules as enforceable constraints — "appears in the first 3 seconds of every video execution", "never used without the wordmark", "removed from all packaging by the next print run".

For high-Fame low-Uniqueness assets, prefer sharpening or pairing over retiring. Retiring a widely recognised cue destroys accumulated Fame that took years to build, and the shared-cue problem is usually fixable by pairing it with a Core asset in the same frame ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).

**7. Select core assets and roles.** Define a prioritised set of **typically 3-5 core assets** with roles: primary assets must appear prominently in all high-reach executions and at the moment of choice; secondary assets support ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)). Resist naming more than five — the point of the set is consistent repetition, and a set of nine is a set of none.

Then pair each core asset with the **category entry points** it should cue. Category entry points are the situations in which the category could be bought or used ([Dawes](https://johndawes.info/the-955-rule/)), and assets perform best when they consistently cue specific buying contexts, strengthening the occasion-to-brand memory link ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).

**8. Validate, fix, re-validate.** Run every check. If any fails, fix and re-run the full list. Only emit the audit when all checks pass.

```
- [ ] Objective metric named in step 1 and referenced in every decision
- [ ] Competitor assets included in the inventory
- [ ] Every scored asset has an unbranded, single-asset stimulus on record
- [ ] Sample is category buyers, not customers only
- [ ] No asset scored from internal opinion; unmeasured assets are labelled so
- [ ] Fame and Uniqueness reported as separate numbers, never averaged
- [ ] High/low thresholds stated numerically with a rationale
- [ ] Every asset has exactly one quadrant and one decision
- [ ] Every decision has an enforceable usage rule
- [ ] Core set contains 3-5 assets, each with primary or secondary role
- [ ] Every core asset is paired with >=1 category entry point
- [ ] Taglines appear as scored assets only if tested
```

## Output format

Use this exact section order and headings; the asset table is read row by row in design and media reviews, and the threshold statement must sit above it. Wording inside sections is yours to adapt.

```markdown
# Distinctive asset audit — <brand>

**Objective metric:** <recognition lift | ad recall | shelf conversion | search conversion | price realisation>
**Test basis:** <n> respondents, category buyers, unbranded single-asset stimuli, fielded <when>. Unmeasured assets are flagged.
**Thresholds:** Fame high = >=<n>%; Uniqueness high = >=<n>%. Rationale: <why>.

## Asset grid
| Asset | Type | Fame % | Uniqueness % | Quadrant | Decision | Usage rule |
|---|---|---|---|---|---|---|

## Core asset set (3-5)
| Asset | Role (primary/secondary) | Must appear in | Category entry points cued |
|---|---|---|---|

## Shared cues to sharpen
| Asset | Also attributed to | Pairing fix |
|---|---|---|

## Retire list
| Asset | Fame % | Uniqueness % | Removal deadline | Owner |
|---|---|---|---|---|

## Unmeasured assets
| Asset | Why unmeasured | Test to run |
|---|---|---|

## Re-test schedule
<Cadence and trigger conditions.>
```

Strictness: keep the table columns exactly as listed. If a score is unmeasured, write `n.m.` rather than estimating a number, because an invented Fame percentage will be quoted back as research.

## Freedom calibration

- **Fixed:** the unbranded single-asset test design, the two score definitions, the four-quadrant mapping and its actions, the 3-5 core asset cap, and the refusal to score from internal opinion. These are the mechanics that make the audit evidence rather than preference.
- **Your judgment:** where to set the high/low thresholds, which competitor assets to include, how to phrase usage rules, and which category entry points to pair with each asset. State the reasoning inline.
- **Ask the user:** existing research data, the asset inventory, the competitor set, and print or production deadlines that constrain retirement timing. Do not estimate Fame or Uniqueness from brand-tracker awareness data; those questions measure a different construct.

## Gotchas

- High Fame with low Uniqueness is the most dangerous quadrant and the one most often celebrated internally, because the asset tests as recognisable. It is a shared category cue, so running it alone in a competitive context aids competitors, and it should be sharpened or paired rather than scaled ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).
- Testing a logo on branded packaging measures the packaging, not the logo. Isolate one asset per stimulus and strip every other cue, because a single co-present cue is enough to carry the attribution ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).
- Sampling customers rather than category buyers inflates every Fame score, and the inflation is largest for the assets the brand most wants to keep. Insist on a representative category sample ([Umbrex](https://umbrex.com/resources/frameworks/marketing-frameworks/distinctive-brand-assets-framework-ehrenberg-bass/)).
- Expect low absolute numbers and do not read them as failure. Building double-digit mental availability is a multi-year task, many well-established brands reach no higher than 20-30% linkage to a category entry point, and market leaders often reach only 50% ([Dawes](https://johndawes.info/the-955-rule/)).
- A rebrand that replaces a Core asset discards measured Fame that cannot be rebuilt inside a campaign cycle. When a redesign is proposed, score the incumbent asset first and put the Fame number in front of the decision-maker, because the cost of the change is exactly that number.
- Distinctive assets are the retrieval cues for mental availability, which is why they are the empirical foundation under the argument that most buyers are out-of-market at any moment ([Dawes](https://johndawes.info/the-955-rule/)). An audit that ends without usage rules binding the assets into high-reach executions has not changed the retrieval odds at all.
- Averaging Fame and Uniqueness into a single "asset strength" score destroys the diagnostic. The four actions differ precisely because the two dimensions fail in different ways; report them as two numbers always.

What does the Auditing Brand Distinctiveness skill do?

Scores brand assets on fame and uniqueness and issues build, protect, sharpen or retire decisions per asset

Where does the Auditing Brand Distinctiveness skill come from?

Report research library. It was written by The Revenue AI Report against a 12 criterion quality rubric and graded in an independent scoring pass.

Why was the Auditing Brand Distinctiveness skill chosen for this library?

Refuses to score assets from internal opinion and converts an unmeasurable request into a test specification, with an explicit Freedom calibration section that is rare across the library.

When should the Auditing Brand Distinctiveness skill not be used?

Do not use it for: Who are our best-fit customers and what value themes should we lead with? Or: Plan our Q3 demand generation budget across channels

How do I install the Auditing Brand Distinctiveness SKILL.md file?

Download the file, create a folder named exactly auditing-brand-distinctiveness inside your agent's skills directory, and save the file inside it as SKILL.md. The agent loads it when a request matches the description.