A new logo does not give you a new brand. It gives you an old brand that customers can no longer find in their heads. The question is never whether the new identity looks better. It is whether it is worth the recognition you are about to delete.
Summary
Brands are stored in memory as a set of recognisable assets: colours, shapes, characters, sounds, typefaces and phrases. Those assets are how buyers retrieve a brand at the moment of purchase. A rebrand replaces them, which resets the memory link and forces the brand to rebuild recognition it already owned. The correct test is not whether the new design is better. It is whether the asset being replaced was weak enough to be worth losing.
Internet Culture · Playbook · Brand 001
Helps you decide: whether to rebrand, what to keep, and what the change will actually cost.
Why this is worth ten minutes
- Almost every rebrand business case is written in the language of design and almost none of it is written in the language of memory, which is where the asset actually lives.
- Distinctive assets are not aesthetic choices. They are retrieval shortcuts, and their value is measured in two numbers most brands have never collected.
- There is a legitimate case for rebranding. It is narrower than agencies suggest and wider than purists admit.
The room
A founder has lived with the same logo for six years. She has looked at it ten thousand times. It now bores her. Someone shows her a competitor with cleaner type and better spacing. A design partner presents a system with an elegant rationale about evolution, clarity and modernity.
Every person in that room has seen the current identity several thousand times more than the customer has.
That is the whole problem, stated in one sentence. The people commissioning a rebrand suffer from a level of exposure that no customer will ever experience. What reads as tired inside the building reads as familiar outside it. And familiarity is not a design outcome. It is an asset with a replacement cost.
2 What is a distinctive brand asset, and why does replacing one cost money?
Jenni Romaniuk of the Ehrenberg-Bass Institute, in Building Distinctive Brand Assets (2018), defines them as non-brand-name elements that trigger the brand in memory. Colour. Logo shape. Character. Sonic cue. Typeface. Tagline. Packaging silhouette. Celebrity association.
Their function is speed. A shopper walking past a shelf does not read. They pattern match. An asset that triggers instant recognition saves the buyer cognitive effort, and saved effort is the closest thing to loyalty that most low-involvement categories can produce.
Romaniuk measures each asset on two axes:
Fame. What share of category buyers link this asset to your brand when shown it without your name?
Uniqueness. What share of those buyers link it to a competitor instead?
Four outcomes fall out of that grid:
| Low uniqueness | High uniqueness | |
|---|---|---|
| High fame | Avoid. Widely recognised, widely misattributed. You are advertising the category. | Protect. This is a real asset. Do not touch it. |
| Low fame | Ignore. Costs you nothing, earns you nothing. | Invest or abandon. Potentially ownable, currently invisible. |
Most rebrands begin without this grid existing. Which means the decision to delete an asset is made without anyone knowing whether it was an asset.
3 Why do so many rebrands go badly in public?
Because the failure mode is predictable and nobody plans for it.
Gap replaced its wordmark in 2010 and reverted within about a week. Tropicana replaced a distinctive package in 2009 and reverted after a reported sales decline. Airbnb’s 2014 identity was mocked for months and then quietly became the most recognisable mark in travel. Twitter’s move to X in 2023 discarded one of the most valuable distinctive assets ever built, including a bird, a colour and a verb that had entered several languages.
Jaguar’s 2024 identity is the most instructive recent case, because both readings are defensible. The company deleted a leaping cat, a heritage wordmark and a set of cues built over decades, and it did so deliberately, while repositioning to a much higher price point and a different buyer. Critics saw memory arson. Defenders pointed out that Jaguar’s existing salience was attached to a customer base it was choosing to exit.
That is the actual test, and it is uncomfortable:
A rebrand is justified when the memories you own are pulling you toward a market you no longer want.
Not when the logo is dated. Not when leadership changed. Not when a competitor looks sharper.
THE RECOGNITION LEDGER
Before you change anything, list every asset you own. Next to each, write two numbers: how many category buyers link it to you, and how many link it to a competitor. That is your opening balance.
A rebrand is a withdrawal from this ledger. Every high-fame, high-uniqueness asset you remove is recognition you have to buy back with media, at current rates, over years.
The one-line version: if you cannot state what you are deleting, in numbers, you are not rebranding. You are redecorating and calling it strategy.
The uncomfortable corollary: the ugliest asset in your system is often the most valuable one, because ugly is memorable and memorable is the entire job.
4 The mechanism stack
01. Recognition is built by repetition, not by quality. Mere exposure, first documented by Robert Zajonc in 1968, shows that repeated exposure alone increases preference. This is why an inelegant mark seen four hundred times outperforms an elegant one seen twice. Design quality and memory strength are only loosely related, which is deeply annoying to designers and consistently true.
02. Consistency is a compounding rate, and inconsistency is a reset. Every year an asset survives, it gets cheaper to recall. Every change restarts the clock. Brands that look boringly consistent are not lazy. They are collecting interest.
03. The internal audience is the wrong audience. Your team, your board and your agency are the only people over-exposed to your identity. Boredom inside the building is not a market signal, and it is the most common trigger for a rebrand.
04. Colour and shape are processed before language. Recognition happens pre-attentively, at distance, in peripheral vision, on a scrolling feed. Which means the assets that survive compression, cropping and thumbnail size are worth more than the ones that need a brand guideline to explain.
05. Evolution beats replacement in almost every case. Keeping the colour and changing the type is a different act to changing both. Most successful long-run identities have been refined many times and replaced almost never.
5 The counter-weight
Sometimes the memory really is the liability. A brand with strong recognition and negative or obsolete associations is carrying a debt, not an asset. Renaming after a scandal, after a category exit or after a pivot that makes the old name actively misleading is a legitimate use of the tool.
Recognition is not everything. A powerfully recognised brand attached to a bad product, a broken business model or a market that is shrinking will die recognisably. Distinctive assets amplify. They do not rescue.
The Ehrenberg-Bass evidence skews to packaged goods. Categories with high purchase frequency, low involvement and shelf-based decisions reward recognition disproportionately. A B2B software company with a nine-month sales cycle and a nine-person buying committee has a different retrieval problem, and the argument for asset preservation is weaker there than the literature implies.
Public mockery is not a business outcome. Airbnb’s identity was ridiculed and then won. The internet’s opinion of a logo in week one is entertainment, not data. Measure recognition at month twelve, not sentiment at hour six.
Internet Culture verdict desk
Your logo is not an expression of your brand. It is the filing code your customer uses to find you. Change the code and you have not refreshed the file. You have hidden it.
6 Steal this
01. Build the ledger before the brief. Test every asset for fame and uniqueness with a sample of category buyers, not existing customers. Existing customers will recognise anything.
02. Separate the two decisions. Repositioning is a strategy decision about who you serve. Re-identification is a design decision about how they find you. Teams routinely do the second while claiming the first, because design is easier to approve than strategy.
03. Protect the top two assets unconditionally. Whatever else changes, the two highest-scoring assets stay. Most successful evolutions keep colour and silhouette and change everything else.
04. Price the rebuild. Estimate the media required to restore current recognition levels on the new system. Put that number in the business case. Watch the conversation change.
05. If you must change, over-bridge. Run the old and new together for longer than feels necessary. The instinct to remove the old asset quickly, so the new one looks confident, is the single most expensive vanity in the process.
7 FAQ
When should a company actually rebrand?
A rebrand is justified when existing brand associations actively conflict with the market you are moving toward, after a merger or a genuine change of business, or when legal or reputational factors make the current name unusable. Being bored of the current design, or a change in leadership, is not a sufficient reason.
What are distinctive brand assets?
Distinctive brand assets are elements other than the brand name that trigger recognition, such as colours, logos, characters, typefaces, sounds and packaging shapes. Their value is measured by fame, meaning how many category buyers link the asset to the brand, and uniqueness, meaning how few link it to a competitor.
Why do rebrands often fail?
Most rebrands fail because they delete recognition the brand already owned, forcing it to rebuild familiarity that took years and considerable media spend to create. The decision is usually made by people over-exposed to the current identity, without measuring how strongly customers currently recognise it.
Is changing a logo worth the risk?
It depends entirely on the current strength of the asset. Replacing a logo with low fame and low uniqueness costs almost nothing. Replacing one with high fame and high uniqueness is a large, quantifiable loss that must be justified by a strategic gain such as entering a different market.
