Redesign your logo when it can no longer do its job: the business it names has changed, the mark fails where you now use it, or you have a name or trademark problem. Those three justify the cost. Almost everything else that starts the conversation — a new marketing lead, a competitor's refresh, or the fact that you have looked at it for years — does not.
There is also a cheaper answer hiding inside most of these conversations. A large share of "we need a new logo" turns out to be a file problem: nobody can find a vector master, there is no one-colour version, and everyone is cropping a screenshot from the website. That gets fixed by rebuilding what you have, not replacing it. Work out which job you need before you brief anyone.
The signals that mean it is time
Each of these is structural — tidying the existing artwork does not resolve them.
- The name changed. A wordmark that says the wrong name is not fixable. The least arguable trigger there is.
- A trademark or similarity conflict surfaced. Someone else is using something close in your category. Continuing to invest in the mark only makes the eventual change more expensive.
- The business is no longer the business in the mark. A logo built around a single product, service, or region becomes a liability once you sell something else. If the icon depicts work you no longer do, it is arguing against your own sales page.
- It fails at the sizes you actually use. A detailed illustration that reads beautifully on a letterhead can turn to mud as a favicon, an app icon, or a social avatar. A logo that is unreadable where customers see it most is not a taste question.
- It cannot be produced in the media you need. Single-colour on a vehicle, embroidered on a shirt, etched, cut, or reversed out on a dark interface. Gradients, thin strokes, and fine detail fail those tests in predictable ways, and the failure is permanent.
- There is no vector master and it must be redrawn anyway. If the only files that exist are raster exports, someone is redrawing the mark regardless — see why your logo looks blurry for what that involves. That is the one honest moment when reconsidering the design costs almost nothing extra, because the labour is already committed.
The signals that mean wait
- You are bored of it. You see your logo hundreds of times a week; a customer sees it for a couple of seconds, occasionally. Internal fatigue is the most common reason logos get replaced, and it is not a customer problem.
- A competitor rebranded. Their reasons are theirs. Reacting to a competitor's identity is how two companies end up looking more alike, not less.
- Someone new arrived and wants to make a mark. If the case survives being written down as a list of things the current mark cannot do, it is real. If it does not, it is a preference.
- Sales are down. The logo is almost never the cause and almost never the fix. The budget does more at the actual weak point — the site, the offer, the follow-up.
- One stakeholder dislikes one thing. A colour, a letterform, the spacing. That is a refresh conversation at most.
Be strict here, because recognition takes years to build and is discarded in an afternoon. Every sighting made your mark fractionally easier to spot; replacing it resets that meter for a benefit you must be able to name.
Redesign, redraw, or refresh: three different jobs
Most of the disagreement here comes from one word doing three jobs.
| Job | What changes | What it touches | When it is right |
|---|---|---|---|
| Redraw | Nothing visible | Files only | The artwork is fine but the files are missing, messy, or raster-only |
| Refresh | Details: colour, typeface, simplification | Digital assets, new print runs | The idea works but reproduces badly or looks dated in its details |
| Redesign | The mark itself | Everything, everywhere | The business, the name, or the mark's basic viability has changed |
A redraw is the cheapest and the most under-used: a clean vector master, a one-colour version, a dark-background version, defined clear space, a stated minimum size, and correctly named exports. Nothing about the design changes; everything about using it gets easier. If your team is cropping the logo out of PDFs, this is the job you need.
A refresh keeps the silhouette recognisable and fixes what is failing — thickening hairlines, removing detail that disappears when small, correcting a colour that dies on dark interfaces, replacing a typeface you no longer have a licence for. People who know you still recognise you. A redesign replaces the mark: the only option that resets recognition, and the only one that touches every physical object you own.
What a redesign actually costs beyond the design fee
The design invoice is usually the smallest line. The rest is rollout, and rollout is where these projects stall:
- Anything printed — packaging, stationery, brochures, labels, forms — becomes obsolete stock the day you switch.
- Anything physical with a lead time — signage, vehicle livery, uniforms, branded product, trade-show stands. These carry installation schedules, not just costs.
- Every template and document — decks, proposals, invoices, contracts, email signatures, the file everyone made their own copy of.
- Every digital placement — favicon, app icon and store listings, social avatars, email templates, partner directories, marketplace profiles.
- The straddle period, during which two versions are in circulation and both look like mistakes.
None of this argues against redesigning — only for doing it once, deliberately, with the list in front of you.
What it costs to leave it too long
The other direction has a quieter bill:
- Failures accumulate in the channels you sell in. An illegible app icon or an unreadable sign costs you a small amount continuously.
- The mark drifts. Without a master file and a rule, people improvise: someone re-types the wordmark in a similar font, someone stretches it, someone drops it on a white box over a photo. You end up with several logos and no logo.
- Every new asset multiplies the eventual switch. Each brochure, vehicle and template made under the old identity is one more thing to redo. Waiting does not preserve the cost; it grows it.
- You can end up locked out of your own artwork. No vector, no editable source, and no way to reach whoever made it — a redraw at minimum, on someone else's timetable.
How to time it so you only do the work once
- Audit before you brief. List every place the mark appears, physical and digital, with quantities and lead times. That list either shrinks the ambition or justifies it, and it doubles as your rollout plan.
- Pair the change with an event that already forces asset turnover — new premises and signage, a packaging reprint, a name change, a website rebuild. Half the rollout cost then belongs to a project you were doing anyway.
- Stop producing old-identity stock the moment the decision is real, even before the new mark exists. Order short runs and let inventory run down.
- Finish the assets before you launch anything. Master vector, full export set, dark and one-colour variants, minimum sizes, clear space, and the rules written down — the brand style guide is what stops the new mark drifting the way the old one did.
- Sequence the switch by cost, not by visibility. Digital placements change in a day, so do them together; printed material follows as stock depletes; signage and vehicles run on their own schedule. Accept the straddle where it is unavoidable — just never let two versions appear side by side in one photograph or one shop window.
If the audit and the signals together say the mark itself has to go, getting a good replacement works the same way it did the first time: how to design a logo covers the brief, the concepts, and the tests a mark passes before you commit.
FAQ
How often should a logo be redesigned? There is no interval. Logos are replaced on triggers — a name change, a business change, a conflict, a reproduction failure — not on a schedule. Plenty of long-lived marks have had nothing but a redraw and a small refresh in decades.
Can we change just the colours and keep the mark? Yes, and it is often the highest-value refresh available. Check the new palette in one colour, reversed out on dark, and against text for contrast before you commit — a palette that only works in its original context creates the next problem.
Do we need a new logo for dark mode? Almost never. You need a variant set: one version for light backgrounds, one for dark, one in a single colour. A mark that cannot produce those variants has a design problem, not an export problem.
We have lost the original files. Is that a redesign? No — that is a redraw, and it should be priced as one. Rebuild the vector master first; once you hold it, you can decide about the design calmly rather than under pressure.
Will a redesign hurt how easily customers recognise us? Temporarily, yes — that is the real cost. You reduce it by keeping continuity in something: the silhouette, the dominant colour, the wordmark, or the name's typographic treatment. Changing every one of those at once is what makes people ask whether you are the same company.
Decide the job before the budget. A redraw fixes files, a refresh fixes details, a redesign replaces the mark — and only the third touches every printed, painted and stitched thing you own. For more on brand assets that survive the change, see the design guides at MyDesign Tool.