Redesigns fail when nobody decided what the site is for.
Most redesign conversations open with how the site should look. Very few open with what it is supposed to do, and the ones that skip that question tend to arrive at the same place: a site that is newer, better looking, and no more useful than the one it replaced.
The decision that matters is not a visual one. It is being able to say, in a sentence, what the site is for — and then being willing to let that sentence throw things out.
"Modern and clean" is a mood, not a brief
Almost every redesign starts here. Modern. Clean. Fresh. Professional. Something that reflects where we are now.
None of that is wrong, and all of it is unfalsifiable. Those words describe how the owner wants to feel about the site, not what the site should achieve. Nobody can tell you in six months whether you got them.
There is a quick test. Could two competent designers read your brief and build completely different sites, and could both of them be right? If so, it is not yet a brief. It is a mood board with sentences.
Most sites have one job
For the majority of small businesses, the site exists to turn a stranger into an enquiry. Not to explain everything the business does. Not to list every service in case one of them is the one someone wants.
Some sites genuinely have a different job, and naming it changes what gets built:
- Get a quote request from someone comparing two or three suppliers
- Take a booking without a phone call happening at all
- Prove the business is legitimate before a call that was always going to happen anyway
- Sell a specific thing to someone who already knows they want it
A decision is only real if it excludes something
Naming the job is a decision, and decisions cost you options. That is what makes them useful.
Once the site is for getting quote requests from homeowners comparing contractors, a great many arguments end on their own. The team page stops competing with the quote form. The full service list becomes one page instead of the shape of the whole site. The question of whether to add a blog has an answer, and the answer is only yes if it feeds the one job.
A brief that excludes nothing has not decided anything. It has just written down everything everyone wanted, which is how a redesign ends up as the old site with new colours.
Who decides is a design decision
Redesigns rarely go over time because the design was hard. They go over because nobody established who gets the final say.
Three people with equal authority and different taste do not produce a better site. They produce rounds. And the thing that survives several rounds of averaging three preferences is not a design — it is the option nobody objected to, which is a different and much worse thing.
One person decides. Others advise, and are genuinely listened to. This is worth settling out loud at the start, when it is an easy conversation, rather than in week five when it is not.
What you are willing to leave out
The clearest evidence that someone made a decision is what is missing.
Every business has more to say than any visitor will read. The instinct is to include all of it, because leaving something out feels like giving something up. In practice the opposite happens: a page that addresses everyone gets read carefully by nobody.
Cutting the third service you barely sell is not losing that work. It is letting the two you actually want more of be read at all.
A measure that exists before the build starts
If you cannot say what would make this a success in six months, you will have no way to tell whether it worked.
It does not need to be sophisticated. Enquiries per month. Calls that came from the site. Quote requests. A number now and the same number later is enough.
Without one, the only available judgement is whether you are still bored of looking at it — which is precisely how businesses end up paying to rebuild the same site every two years, each time for reasons nobody can quite articulate.
What the decision looks like written down
It should be short, specific, and concrete enough that someone could disagree with it.
This site exists to get quote requests from homeowners in Edmonton who are comparing two or three contractors. It should make us look established and straightforward to deal with, and get someone to the quote form in under a minute. Ten qualified requests a month by spring means it worked.
That is not elegant writing. It is not meant to be. But every question that comes up over the following weeks now has somewhere to be resolved, and the answer stops depending on who is in the room that day.
The redesigns that go well are almost never the ones with the best reference images. They are the ones where somebody decided first.