Branding or web design: which do you actually need?
Usually a question of sequence, not suppliers: 58 of the 79 branding agencies we track also do web design. Ask five colleagues to describe what you do. If the answers agree, buy the website. If they disagree, fix the brand first, which published prices show is normally the cheaper of the two.
By Rajat Kapoor
Updated October 2026
Key takeaways
This is rarely a choice between agencies. Of 79 branding firms we hold records for, 58 also do web design, so the real decision is what goes in the brief.
Ask five colleagues to describe the company in one sentence. Agreement means you have a presentation problem; disagreement means no website will fix it.
At every firm that publishes prices for both, the brand identity costs less than the website, because a website is a decision plus a construction project.
Buying a website on an unresolved brand costs twice: once now, and again when the repositioning arrives and the site no longer fits.
A rebrand with no surface to launch on is invisible. If branding comes first, know what it lands on and when.
If traffic arrives and does not convert, that is usually messaging and page structure, and rebuilding the whole site is the most expensive way to test it.
Brand work handed to a web team has to settle arguments in your absence, which means messaging and rules rather than a mood board.
Someone has told you the website looks dated. Someone else has said the problem is not the website, it is that nobody can explain what the company does. You have a budget that comfortably covers one of these and uncomfortably covers both, and two sets of agencies are happy to tell you theirs is the urgent one.
The good news is that this is rarely a choice between suppliers. It is a question about sequence, and about which of two problems you actually have.
Most of these firms do both
Of the 79 branding agencies we hold records for, 58 also work in web design. Only 21 do branding and nothing else, and 37 do web design without offering branding.
That matters for how you shop. The majority of the market will take either brief, which means the decision you are making is not usually which agency to call. It is what to put in the brief, because the same firm will scope a very different project depending on what you ask for, and it will not always volunteer that you have asked for the wrong one.
The two specialist populations are small but they are the ones with the strongest opinions. A branding-only studio will tell you the website is a symptom. A web-design-only shop will build what you give it and will not litigate your positioning. Both are reasonable; neither is neutral.
The cheaper of the two is usually the brand
This surprises people, because branding is the one that sounds like a luxury.
Among the handful of firms that publish prices for both on their own site, the identity work costs less than the website every time. Lil Big Things states brand identity from $6,500 against website projects starting at $27,000 for five pages and $38,000 for ten. SVZ puts branding at $5,000 to over $250,000 and websites starting at $35,000. Isadora Agency, at the enterprise end, publishes branding and identity at $35,000 to $60,000 against website and product design at $95,000 to $165,000 and up.
The pattern holds across the range, and the reason is structural rather than a judgement about value. A brand identity is a bounded deliverable: marks, type, colour, voice, a set of rules. A website is that plus information architecture, copy for every page, design across templates, build, content migration and integrations. One is a decision, the other is a decision plus a construction project.
The practical consequence is that "we can only afford one" is less often true than it sounds. If the brand is genuinely the problem, fixing it is usually the smaller of the two invoices.
The test that tells you which problem you have
Ask five people in your company, separately and in writing, to describe what you do and who it is for. One sentence each.
If the five sentences broadly agree, you have a presentation problem. The company knows what it is; the website is failing to show it. Buy the website, brief it against the positioning you already have, and do not reopen the identity.
If the five sentences disagree in substance, naming different customers or different products, no website will fix that. A design team will ask what the site should say, the five answers will arrive, and either the agency will quietly pick one or the project will stall while you decide. Both outcomes cost more than doing the brand work first.
This is a cheap test and it takes a day. It is also the single thing most reliably skipped before a six figure website project.
Route by where you actually are
Nobody can agree what you do. Branding first, and the budget probably stretches further than you expect. The output you need is positioning and messaging rather than a logo, so check that the scope includes what the company says, not only how it looks.
Everyone agrees, the site looks ten years old. Web design. Bring the existing brand as a constraint and say so in the brief. A firm that opens by proposing to revisit the identity is selling you a bigger project than you have.
You have outgrown the identity and need a site this quarter. Both, from one firm, in that order, as a single engagement. This is the case the 58 dual-capability agencies exist to serve, and sequencing it inside one contract avoids the handover where a brand book meets a web team that was not in the room when it was written.
You were founded three years ago and the logo was made in an afternoon. Branding, but scope it small. An identity refresh is not a repositioning, and firms publish those at very different prices. Say which one you want.
You are raising, or being acquired, or merging. Branding, and earlier than feels comfortable. The deliverables are needed for a dozen surfaces beyond the website, and a brand decided under deal pressure is one you will redo.
Traffic arrives and does not convert. Neither, yet. That is usually a messaging and page structure problem that a focused piece of work can test in weeks. Rebuilding the whole site to fix a conversion rate is the most expensive way to run an experiment.
The site works and the brand works, but they do not look like the same company. This is the real outcome of buying the two separately from two suppliers, and the fix is smaller than either original project: a design system that reconciles them. Ask for that rather than a rebuild.
The order that goes wrong
Buying a website on an unresolved brand is the common mistake, and it is expensive twice: once for the site, and again eighteen months later when the repositioning lands and the site no longer fits.
The less discussed failure runs the other way. A rebrand delivered as a PDF, with no surface to live on, is invisible. The brand book goes into a folder, the website still shows the old marks, and six months later nobody can point to anything that changed. If you buy branding first, know what it is launching onto and when, even if that is a holding page rather than the full site.
If you are briefing both, the useful instruction is that the brand work has to produce something the web team can build against: not a mood board, but messaging, hierarchy and rules specific enough to settle an argument about a page in your absence.
Frequently asked questions
Should branding always come before the website?
Not always, but it should come first whenever the company cannot describe itself consistently. The test is cheap: ask five colleagues to write one sentence on what you do and who it is for. If those sentences broadly agree, the positioning is settled and the website is the job. If they disagree in substance, a design team will ask what the site should say, get five answers, and either pick one for you or stall until you decide. The exception is a hard external deadline, where a focused interim site on the existing brand is usually better than rushing identity work you will redo.
Can one agency do both well?
Most of them claim to, and the majority of this market genuinely offers both: of the branding agencies we hold records for, roughly three quarters also work in web design. Doing both under one contract has a real advantage, which is that the people who decided the positioning are in the room when the pages get designed, and no brand book has to survive a handover. The thing to check is which half the firm is actually known for. Ask for two case studies where it did both for the same client, and look at whether the site expresses the identity or merely applies it.
How much should we budget for each?
Published figures from firms that state both suggest brand identity starts around $5,000 to $6,500 at the accessible end and $35,000 and up at the enterprise end, while websites start around $27,000 to $35,000 and run well past $150,000. Branding is consistently the smaller invoice because it is a bounded set of deliverables, whereas a website adds architecture, copy for every page, build, migration and integrations. Very few agencies publish anything at all, so treat these as the shape of the market rather than a quote, and expect a repositioning to cost considerably more than an identity refresh.
What is the difference between a rebrand and a brand refresh?
A refresh updates how the company looks: marks, type, colour, and often the voice, while leaving the positioning intact. A rebrand changes what the company says it is, which may follow a new market, a merger or a product that has outgrown its description. They are priced very differently and scoped very differently, and the commonest briefing error is asking for one and describing the other. If what has changed is your customers or your product rather than your taste, you are asking for a repositioning, and the deliverable you most need is messaging rather than a logo.
We just need the site to convert better. Which is that?
Usually neither, at least not yet. A conversion problem on a site that already ranks and already receives traffic is most often a messaging and page structure problem, and it can be tested on a handful of pages in weeks rather than rebuilt over months. Start by establishing which pages people arrive on and where they leave, then rewrite and restructure those specifically. If the conclusion of that work is that the company cannot explain itself clearly enough to convert anyone, you have found a brand problem by the cheapest available route.
Our brand book exists but nobody uses it. What do we buy?
Neither a rebrand nor a new website, in most cases. A brand book that nobody follows is usually a brand book that cannot answer the questions people actually have: what this page type looks like, which of two headlines is right, how the thing behaves at mobile width. What closes that gap is a design system tied to real components rather than a PDF of principles. That is a smaller purchase than either project on this page, and it is the usual fix when the site and the brand were bought separately and no longer look like the same company.