This comes up in most first conversations where both the brand and the site need attention. It's a fair question, and the honest answer is that it depends on a few specific things, none of which are about design. Here is how we work it out.
Why the order matters at all
A website is largely an expression of a position. It says who you are for, what you do, why you rather than someone else. If that position is unsettled, the website work stalls, because every page raises a question that nobody has answered yet.
Run the website first with an unresolved brand and you tend to pay twice. The design gets built on assumptions, the assumptions get revisited three months later, and the pages get rebuilt to match. We have inherited plenty of projects where exactly that happened.
Run the brand first with no website in sight and a different problem appears. The identity sits in a guidelines document, nobody uses it consistently, and the momentum quietly leaks away. A brand nobody applies never quite feels real, including to the people who paid for it.
When to start with the brand
Start with the brand when the underlying question is who we are now. That usually shows up after a merger, a change of leadership, a shift into a new market, or the point at which the founder's original idea of the company no longer matches what it has grown into.
It also applies when different people in the business describe what you do differently. If your sales team, your website and your founder each tell a slightly different story, no amount of design will paper over it. That gets resolved in positioning work, not in layout.
One practical marker: if you can't answer “why us rather than them” in a sentence, start with the brand.
When to start with the website
Start with the website when the position is settled and understood, and the problem is that the site is failing to communicate it. That's more common than people expect. Plenty of companies know exactly who they are and simply have a website built for a smaller version of themselves.
It is also the right order when there is commercial urgency. If enquiries have dropped, or a funding round is coming, or a competitor has just launched something better, a website can be improved considerably in ten to sixteen weeks. A brand programme plus a website is a longer commitment.
The identity can then follow, informed by everything the website work surfaced about how customers actually talk about you.
When to do both together
Doing both as one programme is usually the cheapest route when both need work, because the positioning decisions feed directly into structure and copy without a handover in between.
It's also the less frustrating route. When two suppliers own the two halves, you end up in the middle translating between them, and the same questions get answered twice at your expense. That is not a criticism of either supplier; it is just what happens when accountability is split.
The cost is time. A combined programme runs longer than either piece alone, though usually shorter than running them one after the other. If your timeline can't absorb that, sequencing is the sensible compromise.
A practical way to decide
Write down the three questions a prospective customer asks before they buy from you. Now check whether your team would answer them the same way. If they would not, the brand needs work first.
Then check whether your website answers those three questions clearly and quickly. If the answers are correct but buried, that is a website problem and the position is fine.
If both fail, you are looking at a combined programme, and it is worth knowing that early rather than discovering it in month three.
What it costs to get the order wrong
The cost of doing the website first when you needed the brand first is usually rework. Pages get rebuilt, copy gets rewritten, and the second version costs about sixty per cent of the first. It's rarely catastrophic, but it's money you didn't need to spend.
The cost of doing the brand first when the website was urgent is different and harder to see. It's the enquiries you didn't get during the months the identity was in development. That number never appears on an invoice, which is precisely why it gets overlooked.
There's a third failure worth naming. Doing both, but with two suppliers and no single owner. That tends to cost the most, because the disagreements surface late and get resolved in your meeting room rather than theirs.
None of these are disasters. They're just avoidable, and the way you avoid them is a fairly short conversation at the start about what's actually driving the project. If a studio doesn't ask you that in the first meeting, it's reasonable to wonder why not.