Character rebranding: when and how to change appearance
How to understand that a character really needs rebranding
The desire to 'refresh' a mascot often comes not from the market, but from the team's internal fatigue — designers and marketers see it every day and want changes before the consumer notices. To distinguish between a real need and a whim, it's worth evaluating the character based on several objective signals.
- The brand positioning has changed, but the character remains the same in visual appearance and personality — for example, the product has moved to the premium segment, while the mascot looks like a cheap clipart.
- The demographic profile of the audience has shifted by at least one generation, and the new audience associates the drawing style with the 'parent brand'.
- The character visually outdated technically: it does not scale well on digital media, does not work in animation, and gets lost in social media grids.
- Perception tests show a decline in trust or sympathy toward the image while maintaining loyalty to the product — that is, the problem is specifically with the face, not the item.
- Competitors have updated their characters, and against the backdrop of the market, your mascot visually looks outdated by 10-15 years.
If none of these points are confirmed and the request sounds like 'wanting something new,' this is a reason not to touch the character. Cosmetic boredom of the team does not fall under valid reasons for rebranding.
How much can be changed: rule 7–15%
Following the recommendation of McCann London, the safe volume of character appearance changes during a single rebranding is 7–15% of visual characteristics. This is not a strict mathematical formula but a guideline: exceeding this threshold significantly increases the risk that part of the audience will stop recognizing the character on the shelf, in the app, or in advertising. This is exactly what happened with Tropicana in 2009 year — by removing the orange juice glass, the brand symbol for 55 years, the company changed the package silhouette so much that customers could no longer physically find the familiar product on the shelf. In six weeks, the brand lost 30 million dollars, and the packaging was returned to its original appearance.
A similar lesson was learned by the hard cider manufacturer Woodchuck: attempting to replace the mascot Chuck with an abstract apple led to a 7% drop in sales and a loss of position to the competitor Angry Orchard, which maintained its recognizable character. The mascot was brought back.
What cannot be changed, and what can be safely updated
Practice shows that character recognition is based on three layers, and they are fundamentally unequal in significance. The first and most critical one is the silhouette: the shape of the head, pose, and key accessories. The brain recognizes an object by contour in fractions of a second, before processing details. The second layer is expression: emotion, facial features, and the overall character of the image. The third, most flexible layer, is the drawing detail: line style, palette, textures, and small decorative elements.
| Character element | Impact on recognizability | Can it be changed? |
|---|---|---|
| Silhouette (head shape, pose, cult accessories) | Critical | No — or at most 1-2% per cycle |
| Facial expression and overall character | High | Carefully, within the framework of 5-7% |
| Color palette | Average | Can be adapted for media formats |
| Style of lines and drawing (2D/3D, detail level) | Low-average | You can freely change according to trends |
| Clothing, small decorative details | Low | Flexible element, changes according to season and campaigns |
Colonel Sanders changed his suit, pose, and image style dozens of times throughout the history of the KFC brand, but his white suit, black bow tie, and glasses have remained unchanged — this is that same untouchable silhouette. Chester Chillis changed the degree of 'boldness' in his facial expression and the detail of his fur, but the shape of his head and his aviator glasses have remained constant for several decades.
How to rebrand a character without losing the audience
The main mistake is to switch the character instantly, without a transition period. This is exactly how the Twitter rebranding in X suffered: a complete rejection of the previous visual identity, without any elements of continuity, caused alienation among part of the audience and triggered a wave of criticism that lasted for months. Foreign practice unambiguously advises gradual rollout — the gradual implementation of changes.
- Fix the immutable core of the character — the silhouette and key attributes — and mark them as 'non-negotiable' before the project begins.
- Determine the specific volume of changes within the 7-15% range and distribute it between the drawing expression, palette, and drawing style.
- Launch changes through intermediate versions: first, update digital channels, then — packaging, and finally — outdoor advertising and large media.
- Give the audience a 'bridge' — combine the old and new image in a single campaign during the transitional period from 2 to 3 months.
- Collect feedback at intermediate stages through recognition tests, not only through satisfaction surveys.
- Finalize the final version only after confirming that the recognition level has not dropped below the baseline indicator.
How often to update the character
The optimal cycle for updating visual branding, including the brand character, is once every 5-10 years. But this figure is not a schedule, but rather the upper limit of the market's patience: if the positioning or audience has shifted earlier than expected, there is no need to wait for the formal date. And vice versa, if the character continues to work and even after 12 years — there is no point in touching it merely because 'a lot of time has passed.' The decision is always made based on signals of mismatch, not the calendar.
What ultimately distinguishes a successful rebranding from a failed one
Sleepy from Reddit and Chester from Cheetos — examples of characters that survived dozens of visual iterations precisely because each change was targeted and preserved the silhouette. The failures of Tropicana, Woodchuck, and Twitter/X are united by one thing: abandoning the recognizable contour for the sake of a 'fresh' look. Rebranding a character is managing risk, not a creative experiment, and the main question to ask before starting a project is not 'how to make it prettier,' but 'what will be recognizable after a second even without a logo next to it.'
- The brand positioning has changed, but the character remains the same in visual appearance and personality — for example, the product has moved to the premium segment, while the mascot looks like a cheap clipart.
- The demographic profile of the audience has shifted by at least one generation, and the new audience associates the drawing style with the 'parent brand'.
- The character visually outdated technically: it does not scale well on digital media, does not work in animation, and gets lost in social media grids.
- Perception tests show a decline in trust or sympathy toward the image while maintaining loyalty to the product — that is, the problem is specifically with the face, not the item.
- Competitors have updated their characters, and against the backdrop of the market, your mascot visually looks outdated by 10-15 years.
If none of these points are confirmed and the request sounds like 'wanting something new,' this is a reason not to touch the character. Cosmetic boredom of the team does not fall under valid reasons for rebranding.
How much can be changed: rule 7–15%
Following the recommendation of McCann London, the safe volume of character appearance changes during a single rebranding is 7–15% of visual characteristics. This is not a strict mathematical formula but a guideline: exceeding this threshold significantly increases the risk that part of the audience will stop recognizing the character on the shelf, in the app, or in advertising. This is exactly what happened with Tropicana in 2009 year — by removing the orange juice glass, the brand symbol for 55 years, the company changed the package silhouette so much that customers could no longer physically find the familiar product on the shelf. In six weeks, the brand lost 30 million dollars, and the packaging was returned to its original appearance.
A similar lesson was learned by the hard cider manufacturer Woodchuck: attempting to replace the mascot Chuck with an abstract apple led to a 7% drop in sales and a loss of position to the competitor Angry Orchard, which maintained its recognizable character. The mascot was brought back.
What cannot be changed, and what can be safely updated
Practice shows that character recognition is based on three layers, and they are fundamentally unequal in significance. The first and most critical one is the silhouette: the shape of the head, pose, and key accessories. The brain recognizes an object by contour in fractions of a second, before processing details. The second layer is expression: emotion, facial features, and the overall character of the image. The third, most flexible layer, is the drawing detail: line style, palette, textures, and small decorative elements.
| Character element | Impact on recognizability | Can it be changed? |
|---|---|---|
| Silhouette (head shape, pose, cult accessories) | Critical | No — or at most 1-2% per cycle |
| Facial expression and overall character | High | Carefully, within the framework of 5-7% |
| Color palette | Average | Can be adapted for media formats |
| Style of lines and drawing (2D/3D, detail level) | Low-average | You can freely change according to trends |
| Clothing, small decorative details | Low | Flexible element, changes according to season and campaigns |
Colonel Sanders changed his suit, pose, and image style dozens of times throughout the history of the KFC brand, but his white suit, black bow tie, and glasses have remained unchanged — this is that same untouchable silhouette. Chester Chillis changed the degree of 'boldness' in his facial expression and the detail of his fur, but the shape of his head and his aviator glasses have remained constant for several decades.
How to rebrand a character without losing the audience
The main mistake is to switch the character instantly, without a transition period. This is exactly how the Twitter rebranding in X suffered: a complete rejection of the previous visual identity, without any elements of continuity, caused alienation among part of the audience and triggered a wave of criticism that lasted for months. Foreign practice unambiguously advises gradual rollout — the gradual implementation of changes.
- Fix the immutable core of the character — the silhouette and key attributes — and mark them as 'non-negotiable' before the project begins.
- Determine the specific volume of changes within the 7-15% range and distribute it between the drawing expression, palette, and drawing style.
- Launch changes through intermediate versions: first, update digital channels, then — packaging, and finally — outdoor advertising and large media.
- Give the audience a 'bridge' — combine the old and new image in a single campaign during the transitional period from 2 to 3 months.
- Collect feedback at intermediate stages through recognition tests, not only through satisfaction surveys.
- Finalize the final version only after confirming that the recognition level has not dropped below the baseline indicator.
How often to update the character
The optimal cycle for updating visual branding, including the brand character, is once every 5-10 years. But this figure is not a schedule, but rather the upper limit of the market's patience: if the positioning or audience has shifted earlier than expected, there is no need to wait for the formal date. And vice versa, if the character continues to work and even after 12 years — there is no point in touching it merely because 'a lot of time has passed.' The decision is always made based on signals of mismatch, not the calendar.
What ultimately distinguishes a successful rebranding from a failed one
Sleepy from Reddit and Chester from Cheetos — examples of characters that survived dozens of visual iterations precisely because each change was targeted and preserved the silhouette. The failures of Tropicana, Woodchuck, and Twitter/X are united by one thing: abandoning the recognizable contour for the sake of a 'fresh' look. Rebranding a character is managing risk, not a creative experiment, and the main question to ask before starting a project is not 'how to make it prettier,' but 'what will be recognizable after a second even without a logo next to it.'
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
