How We Evaluate a Website

We have written before about why most website audits miss the point, which ended on three questions: did anyone ask who your most important audiences are, did anyone map the site against how you actually sell, and did anyone put a number on what a visitor is worth. This is the answer to those questions, which is to say it is how we do it.

There are four passes. They run in this order for a reason, and none of them produces a score.

First: the journeys the business depends on

Before looking at the site, we work out what it needs to do. Not from a checklist, from the business.

A guide service and a care home both sell an appointment, and their customers arrive in completely different states. One is planning something enjoyable months ahead and comparing options on price and dates. The other is frightened, deciding quickly, and comparing on trust. The pages might look similar. The journeys are not remotely alike.

So the first pass writes down the handful of journeys the business actually runs on, in the customer's terms, and then goes looking for each one on the site. The question is never "does this page exist". It is "if I am this person, with this problem, can I get from here to the thing I need, and how many steps does it take".

Most of what we find lives in the gaps. The journey starts fine, the site answers the first two questions, and then the path stops at a contact form that leads to an inbox. Nothing on the page is broken. The journey is.

Second: whether the copy does its job

Separate pass, because it is a separate skill and mixing the two produces mush.

This one is closer to a hit list. Is the value proposition stated plainly, in the place a reader will actually be when they need it, rather than three scrolls down. Is there proof near the claims that need it. Does the page ask for the thing it wants, once, clearly.

Copy problems are usually cheaper to fix than journey problems, which is worth knowing before anyone talks about a rebuild.

Third: what people search for, and the one gap that matters

We look at what the business's customers actually type, and where the site does and does not show up.

The temptation here is to hand over forty keywords. We deliberately do not. Four of them will be the company name and the obvious product, one will be a real gap, and the reader will tune out the whole list before they reach it. One high-signal finding, tied to their business, is worth more than a page of terms they will never act on.

Fourth: what the gap is worth

Every finding gets framed as business value or it does not go in the report.

That means arithmetic, and the arithmetic is deliberately conservative. If a booking path takes five messages to complete, some share of decided customers never finish it. Put a number on that share, multiply by what a customer is worth, and the finding stops being a design opinion. It becomes a line item the owner can weigh against what fixing it costs.

This is also the honesty check on our own work. A finding we cannot put a number to is usually a finding about taste.

How the findings get ordered

By what they cost, not by how much they annoy us.

Critical means the revenue path is broken: a customer who has decided to buy cannot complete the transaction. That is the top of the list every time, however good the site looks. High means a journey is leaking badly at a specific step. Medium is real, measurable, and can wait a quarter.

Nothing on the list is there because it is unfashionable.

What we deliberately leave out

A score. A number out of a hundred compresses everything above into a figure that cannot be acted on, and it invites the wrong conversation.

Everything we noticed. A thorough-looking report is easy to produce and hard to use. If there are three things worth doing, the report says three things. The temptation to pad it is a tell that the audit did not find much.

The obvious. Nobody needs to be told their images are large. Tools do that part for free and they do it better than we would.

Doing it yourself

Most of the first pass is available to anyone willing to be uncomfortable for an afternoon. Write down your three most important customers. For each one, start where they start, usually a search or a recommendation, and walk to the moment money changes hands. Count the steps. Mark every point where progress depends on you or someone on your team noticing something and acting.

That exercise finds the same critical items we would find. What it will not do is price them, and pricing them is what turns a list of annoyances into a decision about where the next few thousand dollars go.

If you would rather we walked it with you, that is where every project starts, before anyone talks about scope. Tell us about your business, or read how we price the work that follows.