Refresh or rebrand? A decision tree for tired identities

Every tired brand arrives at the studio the same way: a founder or CMO slides a laptop across the table, shows us the current site, and says some version of "it just doesn't feel like us anymore." Sometimes they're right and the answer is a rebrand. Sometimes the brand is fine and the product experience is what's tired — the identity is taking the blame for a checkout flow. And sometimes the brand is genuinely strong, fondly recognised, commercially load-bearing, and the itch to change it is boredom mistaken for strategy. Those three situations look identical from the inside. They require opposite responses and wildly different budgets.

This is the decision tree we walk before anyone sketches a letterform. It routes a tired identity to one of three paths — refresh, evolution, or rebrand — and it's saved our clients from at least two of the most expensive mistakes a company can make: rebuilding equity it already owned, and painting over a problem that lives in the product.

First, define the three paths

People use "refresh" and "rebrand" interchangeably, which is how scopes explode. Our working definitions:

• Refresh keeps every strategic load-bearing element — name, mark's core idea, palette's signature colour — and retunes the craft: redrawn wordmark, expanded type system, coherent art direction, modern motion. The audience shouldn't notice a change so much as a sharpening. Typical engagement: six to ten weeks.
• Evolution keeps the name and the strategy but replaces most visual components: new mark built on the old one's logic, new palette grown from the old hero colour, new voice principles. The audience reads it as "same company, next chapter." Three to five months if it's done properly, including the rollout's unglamorous middle.
• Rebrand changes the strategic platform itself: positioning, possibly name, certainly identity system. The audience is being told "we are a different thing now." Six months-plus, and the identity work is the cheap part; the expensive part is everything else the company has to change to make the new story true.

The cost curve between these is not linear, it's closer to 1×, 3×, 10× — in money, but more importantly in organisational attention and risk. Nicking the wrong branch is not a small error.

The diagnostic: five questions, answered with evidence

We score these in the first workshop, and where the answers are vibes we run the 90-minute brand audit to get evidence instead. The five questions:

1. Does the market still hold what you own? Run the equity checks: aided and unaided recognition of the mark and signature colour in your category, brand-term search volume trend, and what sales calls actually say. If recognition is strong and warm, a rebrand spends down an asset to buy a feeling. We've twice talked clients out of rebrands by showing them their own search data — the identity was working harder than anyone in the building believed. The discipline of measuring brand without voodoo matters most at exactly this moment, because it converts "we're bored of it" into numbers.

2. Is the problem the identity or the product? The sharpest test we know: take the brand off the artefact. Screenshot the product experience — the flow, the pricing page, the support reply — with every logo and colour removed, and test it against competitors' equivalents with ten target customers. If the unbranded experience still loses, no amount of identity work fixes it; rebrand money spent there is confetti on a leak. Identity problems are identifiable: assets look dated next to category peers, the system breaks at new touchpoints, the mark is illegible at app-icon size. Product problems are experiential. Different doctors.

3. Has the strategy changed, or just the taste? A rebrand is justified by strategy change — new category, new buyer, new business model, a merger, a reputation that needs severing. If the answer is "leadership is tired of looking at it," that's a taste problem, and taste problems are refresh problems. Companies outgrow identities every few years the way you outgrow a haircut; they outgrow strategies rarely.

4. What is the equity actually attached to? Equity audit, concrete version: which specific elements carry the recognition? Usually it's one or two things — a name, a colour, a distinctive mark shape — and everything else is ballast. If equity lives in the wordmark and the colour, an evolution that keeps both and rebuilds everything else is nearly free. If equity lives in elements that are also the problem (a name that caps your category ambition), that's the rare case where the rebrand premium is worth paying. Heritage brands have a sharpened version of this math — see modernising a heritage brand without erasing its soul.

5. Can the organisation absorb the change? The underestimated question. A rebrand touches every surface simultaneously — legal, product, signage, social, email footers, the sales deck, the van livery — and requires an organisation with the executive attention to land it. If the exec team is mid-pivot, mid-hiring-freeze or mid-anything, the honest answer is refresh now, rebrand when there's oxygen. A strategy nobody can resource is a PDF.

The risk curves, stated plainly

Each path fails differently:

• Refresh risk is invisibility. You spend ten weeks sharpening and the market doesn't notice. That's not failure — that's the design working — but it must be sold internally as craft maintenance, or twelve months later someone says "we paid for a rebrand and nothing changed."
• Evolution risk is uncanny valley. Enough changed to alarm loyalists, not enough to excite anyone new. The mitigation is a strong through-line story: what survived, why, and what it makes possible.
• Rebrand risk is equity destruction plus operational chaos, in parallel. The graveyard of rebrands is mostly category two — launches where the website flipped but the product, emails and invoices kept telling the old story for a year. Our piece on rebrands that worked versus the ones that were new paint is largely a taxonomy of this failure.

Announcing the small change well

One last craft point, because refresh work is chronically under-announced. The temptation is to ship a refresh silently since "nothing strategic changed." Wrong instinct. A refresh announced well is a chance to restate the strategy using the identity as evidence: "same promise, sharper tools" is a story, and it gives customers, press and your own team a reason to look again. Write the announcement from the strategy outward — what you believe, then what you changed to express it — and keep the craft details for the design community, where announcement-day theatre earns its keep.

And if the diagnostic genuinely routes you to evolution or rebrand: do the strategy work first, on one page or not at all, before a single moodboard. The decision tree protects you from the wrong project; the strategy page protects the right one.

Key takeaways

• Refresh, evolution and rebrand are three different projects with roughly 1×/3×/10× cost and risk curves — define which one you're buying before scoping anything.
• Test whether equity is actually held (recognition, brand search volume) before spending it down for novelty.
• Remove the brand and test the naked product experience; if it still loses, the problem isn't the logo.
• Strategy change justifies rebrand; taste change justifies refresh. Boredom is not a strategy input.
• Match the path to the organisation's absorption capacity — an unresourced rebrand is a PDF, not a platform.
• Announce even a refresh: "same promise, sharper tools" is a story that earns the second look.

Frequently asked questions

How often should a brand refresh? Craft refreshes run on roughly a three-to-five-year cadence for digital-first brands — driven by touchpoint evolution (new formats, dark mode, motion) more than by fashion. If you're refreshing more often than every two years, the system isn't the problem; the governance of it is.

Can an evolution change the name? No — a name change makes it a rebrand by definition, because name is the single highest-equity element for almost every company. Names change when the strategy has genuinely shifted categories, and rarely otherwise.

What if the founder just hates the logo? Hatred is data but weak data. Run the equity checks first: if the market holds the mark warmly, the founder's taste is the cheapest thing to change — usually by showing the system working hard across real touchpoints rather than defending the artefact.

Is a refresh worth doing before a fundraise or launch? Yes, if it's quick and the current system visibly breaks at the new stakes (investor deck, press kit, app icon grid). A refresh is the highest-leverage six weeks in brand work; a rebrand before a deadline is how you get both a delayed launch and a compromised identity.

Who should be in the diagnostic workshop? Five to seven people: whoever owns strategy, someone from sales or support who hears the market weekly, one product voice, and — critically — nobody who needs the meeting to reach a particular answer. Diagnostics done by people with a preferred outcome produce the preferred outcome.