Metrics of slogan effectiveness: how to measure memorability
Unaided recall vs aided recall: a fourfold difference is normal, not a test error
Unaided recall — the question "what slogans of brands in this category do you remember?" without hints. Aided recall — the brand or slogan shown with the question "do you recognize it?". The first shows whether the phrase lives in the consumer's head on its own, the second shows whether he recognizes it when it is shown to him.
| Parameter | Unaided recall | Aided recall |
|---|---|---|
| What it measures | Slogan surfaces without stimulus | Slogan is recognized when shown |
| Average market result | ~15% | ~62% |
| Nike result ("Just Do It") | 65% (top 3 in the sports goods category) | significantly above average |
| What a high result shows | the real depth of positioning | the minimum threshold of recognizability, often maintained by media budget |
A fourfold gap between 15% and 62% is not an anomaly of a specific study, but a typical market condition for all brands except for a few category leaders such as Nike. If your agency is bringing you only aided recall in 70-80%, ask about unaided — there is a high probability that it will show 8-12%, and this will mean that the slogan does not form spontaneous association at the moment of purchase, but simply 'is not forgotten' when the logo is shown.
What really makes a phrase memorable: data on 820 slogans
An analysis of 820 slogans from different categories showed that the linguistic parameters of the phrase explain 23% of the variance in reproduction accuracy — and they work against the copywriter's intuition, who tends to reach for a short and catchy line.
- Length. Long constructions with rare words are remembered 1,88 times better than short clichés — contrary to the rule 'the shorter, the better'.
- Word frequency. A rare, unfamiliar word in a stream of casual speech creates an effect of 'catching' — the brain registers the violation of expectation rather than the smoothness of the phrase.
- Concreteness of imagery. «Melts in your hands, not in your mouth» (M&M's) is better remembered than the abstract «quality you can trust», because the brain has something to visualize.
- Brand name within the phrase. Slogans where the brand name is embedded in the text, rather than appearing as a separate line underneath, provide a notable boost in reproduction accuracy.
- Semantic diversity. Words from different semantic fields within one phrase work better for memory than internal rhyme.
About rhyme separately: constructions like «tasty and precise» or «fast and simple» sound harmonious at a briefing and fail in blind testing. The brain is accustomed to predictable sound patterns and skips them past memory — rhyme works as background noise, not as a hook.
The slogan was remembered, but it wasn't liked — what to do with this contradiction
In the same study, 820+ slogan recall and liking moved in opposite directions: phrases with rare words and concrete imagery were better remembered, but at first contact were liked by the audience less — they sound unfamiliar, sometimes sharply. Smooth 'safe' formulations are liked immediately and fade from memory within a day.
Hence, a specific rule for choosing priority:
- Brand launch, rebranding, entering a new category — priority is recall, even at the cost of audience discomfort at the start.
- Strengthening emotional connection with an already known brand — a smoother phrase with lower recall can be chosen, because brand recognition is ensured by other channels.
- Attempting to achieve maximum results on both scales simultaneously almost always yields a compromised slogan that loses on both parameters at the same time — this is the most common mistake in the brief.
Sample size and test duration: numbers below which results cannot be trusted
A test on 20 employees of the agency in the conference room — this is filtering out completely failed variants, not the final decision. Below are the minimum parameters for which results can be trusted.
- For aided recall and recognition, 150-200 respondents per segment is sufficient — with a response rate of 50-60%, the confidence interval falls within ±5-7%.
- For unaided recall, the sample size needs to be raised to 300+ people: the proportion of spontaneous mentions is usually 10-20%, and with a small sample, the error completely consumes the result.
- A single measurement immediately after showing the slogan is meaningless without a repeat test after 7 and 30 days — otherwise you are measuring short-term memory, not retention.
- A control group that has not seen the slogan is mandatory: without it, you cannot distinguish the effect of the phrase itself from the overall brand recognition that already existed before the test.
Seven criteria instead of two: how to calculate scores and what they mean
Recall and liking — only two coordinates out of seven. The working evaluation model is a 7-point scale for each of the following criteria: memorability, clarity of formulation, originality, expressiveness, alignment with the brand's field of activity, attractiveness, and reflection of brand uniqueness. Maximum — 49 points.
| Total score (out of 49) | Slogan class | What to do |
|---|---|---|
| 35-49 | excellent | launch without revisions |
| 22-28 | good, typical for the market | strengthen specific weak criteria |
| below 22 | weak | rewrite the concept from scratch |
Most tested slogans on the market score 22-25 points — 'good,' but not outstanding. Systematically failed criteria — originality and expressiveness, and this is a pattern, not random variation: the industry has been moving toward a 'human,' simplified brand tone for several years and ultimately produced a mass of safe phrases that are indistinguishable from one another in blind testing.
Working test scheme for a specific budget
For a premium solution — brand launch, federal campaign — the minimal scheme looks like this: collect 5-8 slogan variants, run each through a 7-point evaluation on 150-200 respondents from the target audience, then separately measure unaided and aided recall with repetition after one week and one month on control and test groups.
For local businesses with limited budgets, the scheme can be simplified to a combined-test on 100 respondents — but this is an informed risk of broader inaccuracy, not a free alternative to the full methodology. Before launching the test, one thing should be fixed: what is more important for this task — the phrase that will enter spontaneous market memory, or the phrase that will appeal to today's customers on first sight. Without this decision, any scores from the report will remain just scores, unlinked to business results.
Unaided recall vs aided recall: a fourfold difference is normal, not a test error
Unaided recall — the question "what slogans of brands in this category do you remember?" without hints. Aided recall — the brand or slogan shown with the question "do you recognize it?". The first shows whether the phrase lives in the consumer's head on its own, the second shows whether he recognizes it when it is shown to him.
| Parameter | Unaided recall | Aided recall |
|---|---|---|
| What it measures | Slogan surfaces without stimulus | Slogan is recognized when shown |
| Average market result | ~15% | ~62% |
| Nike result ("Just Do It") | 65% (top 3 in the sports goods category) | significantly above average |
| What a high result shows | the real depth of positioning | the minimum threshold of recognizability, often maintained by media budget |
A fourfold gap between 15% and 62% is not an anomaly of a specific study, but a typical market condition for all brands except for a few category leaders such as Nike. If your agency is bringing you only aided recall in 70-80%, ask about unaided — there is a high probability that it will show 8-12%, and this will mean that the slogan does not form spontaneous association at the moment of purchase, but simply 'is not forgotten' when the logo is shown.
What really makes a phrase memorable: data on 820 slogans
An analysis of 820 slogans from different categories showed that the linguistic parameters of the phrase explain 23% of the variance in reproduction accuracy — and they work against the copywriter's intuition, who tends to reach for a short and catchy line.
- Length. Long constructions with rare words are remembered 1,88 times better than short clichés — contrary to the rule 'the shorter, the better'.
- Word frequency. A rare, unfamiliar word in a stream of casual speech creates an effect of 'catching' — the brain registers the violation of expectation rather than the smoothness of the phrase.
- Concreteness of imagery. «Melts in your hands, not in your mouth» (M&M's) is better remembered than the abstract «quality you can trust», because the brain has something to visualize.
- Brand name within the phrase. Slogans where the brand name is embedded in the text, rather than appearing as a separate line underneath, provide a notable boost in reproduction accuracy.
- Semantic diversity. Words from different semantic fields within one phrase work better for memory than internal rhyme.
About rhyme separately: constructions like «tasty and precise» or «fast and simple» sound harmonious at a briefing and fail in blind testing. The brain is accustomed to predictable sound patterns and skips them past memory — rhyme works as background noise, not as a hook.
The slogan was remembered, but it wasn't liked — what to do with this contradiction
In the same study, 820+ slogan recall and liking moved in opposite directions: phrases with rare words and concrete imagery were better remembered, but at first contact were liked by the audience less — they sound unfamiliar, sometimes sharply. Smooth 'safe' formulations are liked immediately and fade from memory within a day.
Hence, a specific rule for choosing priority:
- Brand launch, rebranding, entering a new category — priority is recall, even at the cost of audience discomfort at the start.
- Strengthening emotional connection with an already known brand — a smoother phrase with lower recall can be chosen, because brand recognition is ensured by other channels.
- Attempting to achieve maximum results on both scales simultaneously almost always yields a compromised slogan that loses on both parameters at the same time — this is the most common mistake in the brief.
Sample size and test duration: numbers below which results cannot be trusted
A test on 20 employees of the agency in the conference room — this is filtering out completely failed variants, not the final decision. Below are the minimum parameters for which results can be trusted.
- For aided recall and recognition, 150-200 respondents per segment is sufficient — with a response rate of 50-60%, the confidence interval falls within ±5-7%.
- For unaided recall, the sample size needs to be raised to 300+ people: the proportion of spontaneous mentions is usually 10-20%, and with a small sample, the error completely consumes the result.
- A single measurement immediately after showing the slogan is meaningless without a repeat test after 7 and 30 days — otherwise you are measuring short-term memory, not retention.
- A control group that has not seen the slogan is mandatory: without it, you cannot distinguish the effect of the phrase itself from the overall brand recognition that already existed before the test.
Seven criteria instead of two: how to calculate scores and what they mean
Recall and liking — only two coordinates out of seven. The working evaluation model is a 7-point scale for each of the following criteria: memorability, clarity of formulation, originality, expressiveness, alignment with the brand's field of activity, attractiveness, and reflection of brand uniqueness. Maximum — 49 points.
| Total score (out of 49) | Slogan class | What to do |
|---|---|---|
| 35-49 | excellent | launch without revisions |
| 22-28 | good, typical for the market | strengthen specific weak criteria |
| below 22 | weak | rewrite the concept from scratch |
Most tested slogans on the market score 22-25 points — 'good,' but not outstanding. Systematically failed criteria — originality and expressiveness, and this is a pattern, not random variation: the industry has been moving toward a 'human,' simplified brand tone for several years and ultimately produced a mass of safe phrases that are indistinguishable from one another in blind testing.
Working test scheme for a specific budget
For a premium solution — brand launch, federal campaign — the minimal scheme looks like this: collect 5-8 slogan variants, run each through a 7-point evaluation on 150-200 respondents from the target audience, then separately measure unaided and aided recall with repetition after one week and one month on control and test groups.
For local businesses with limited budgets, the scheme can be simplified to a combined-test on 100 respondents — but this is an informed risk of broader inaccuracy, not a free alternative to the full methodology. Before launching the test, one thing should be fixed: what is more important for this task — the phrase that will enter spontaneous market memory, or the phrase that will appeal to today's customers on first sight. Without this decision, any scores from the report will remain just scores, unlinked to business results.
Recent articles
MegaFon, BMW, VTB and more 240+ companies. See how it looks in the showcase reel:
Our branding agency is approached to order:
Branding agency services
