Something about it is off, and you cannot say what

The usual situation

Nothing is broken. It still feels wrong.

You look at your own website and something is not right. Every part of it seems fine on its own. The photographs are good, the words are yours, and nobody has ever complained. But it does not feel like the thing you meant to make, and when you try to say why, you cannot.

So most people decide it is just them. That they are being fussy, or that they have looked at it too long. And they leave it.

In my experience that feeling is correct far more often than not, and what is causing it is usually specific, structural, and fixable. It is simply invisible unless you go looking with something other than your eyes.

What I do

Some of it is measurable. The rest takes listening.

Anyone can tell you what they would have done differently. That only swaps their taste for yours, which is why so much advice about websites feels like being overruled.

Some of what makes a page feel wrong is genuinely countable. Contrast is a number. Content widths are a number. Competing things to click can be added up, and whether the page survives a phone can be tested rather than guessed. Here is some of what I count:

01 Whether the page settles, or keeps shifting under you. Your eye returns to the same left edge on every line. When that edge moves between sections, the page never comes to rest, and you feel it without being able to see it. The cause is usually four or five or seven different content widths where there should be two. It takes minutes to count and it is almost never noticed by anyone.
02 Whether people can actually read the first thing you wrote. Pale text over a photograph is the usual version. It looks elegant to whoever made it and is genuinely hard work for everybody else. Contrast is a measurement rather than a matter of taste, and it is very often the opening paragraph that fails.
03 How far someone has to travel to reach the best thing about you. On most sites the most compelling thing is in the lower third, after the explaining. The farther down it sits, the fewer people ever reach it. I find out where it actually is and how long it takes to arrive.
04 How many different things the page asks of a visitor. Two buttons a few inches apart. A price in one place and another price in a different shape further down. Each was sensible on the day it was added. Together they leave someone unsure what they are meant to do, so they do nothing.
05 Whether it holds together on a phone. For most of the people I work with, this is where their visitors actually arrive, and it is where sites quietly come apart. Text running off the edge, one oversized image forcing the whole page wider than the screen so everything else shrinks, buttons too close together to tap. It is rarely the version anyone checks.

Other things cannot be counted, and they are usually the ones that matter most. What is most alive in your work. What somebody needs to understand about you before anything else will land. Whether the shape of the site is actually holding up what you are trying to say.

I look at both, and I tell you which is which. You will always know whether you are reading a measurement or a judgement, and you are free to disagree with the second kind.

You get a written report: every finding, the evidence for it, and an order, because a list of twenty problems is no more useful than none. I will tell you which three actually matter and which you can happily ignore.

One I did recently

I started with one of my own.

The same website before and after: the earlier version has washed out text over a photograph, the later version has the headline on a clean background with the photograph beneath

Before, and after. Same words. Same photographs.

This is the site for my BioGeometry practice. I had known for months that something about it was wrong, and I could not say what. I assumed it was me being fussy about my own work, which is what almost everybody assumes.

Then I measured it. Here is what an hour turned up.

01 Seven different content widths. 600, 660, 700, 736, 800, 900 and 1100 pixels. Nothing on the page shared an edge with anything else.
02 The opening paragraph failed contrast. Grey text sat on a pale photograph under a cream overlay, which lifted the background toward the text instead of away from it. It measured 3.54 to 1 against a 4.5 minimum.
03 Alignment changed from section to section, and in one case three times inside a single section.
04 Two images in twenty thousand pixels of page, and my own portrait was two thirds of the way down, where almost nobody reaches.

Every one of those is countable, and once they were counted it was obvious what to do about them.

The harder part came afterwards, and none of it was arithmetic: which sections were saying the same thing twice, and which piece of writing deserved to be the centre of the page rather than ninth on it. That is the listening half, and it is usually where the largest change comes from.

The rebuild took a day. Not one word of the original writing was thrown away.

I am showing you my own because it is the honest version of what this is. I was the client. The feeling you have about your site is one I have had about mine.

What you get

A written report, in the order that matters.

You send me the address. That is all I need. It does not matter what your site was built with, because I am looking at it the way a visitor does.
I go through it properly, on a laptop and on a phone, measuring rather than glancing.
You get the report within a week, usually sooner. Every finding, the evidence behind it, and which three I would fix first.
Twenty minutes to talk it through, if you want it. Some people would rather just read it, which is fine.

The report is yours. Take it to whoever built your site, or to anyone else, or keep it and do the work yourself. There is nothing in it that only I could act on.

What it costs

Said plainly.

The audit

$333

One website. A written report, ranked, with the evidence for every finding. Within a week. Twenty minutes to talk it through if you want them.

If I find nothing worth fixing, you pay nothing

If you decide afterwards that you want me to do the work, the $333 comes off the price of the build. If you would rather take the report somewhere else, that is a completely fine outcome and I will not chase you.

This is not a redesign, and it is not a plan for a site you have not built yet. It is a diagnosis of the site you already have.

If what you actually need is the thinking rather than the diagnosis, that is a different thing and it is $400. If you are not sure which of the two you want, book the free twenty minutes and I will tell you, including if the answer is neither.

Reasonable questions

The things you might be wondering.

What if there is genuinely nothing wrong with it?

Then I tell you that, and you pay nothing. I would rather say so than manufacture a list to justify the fee. It does happen that a site is doing its job and the owner has simply been staring at it too long, and that is worth knowing too.

My site was built on Squarespace. Does that matter?

No. Squarespace, Wix, WordPress, Shopify, something a friend made in 2019. I am looking first at the page as it arrives in front of a real person, because that is the experience the audit is about. How it was built matters afterwards, when we come to what is realistic to change.

Will you just tell me to rebuild it with you?

Only if that is actually the answer, and often it is not. Most of what I find can be fixed in the tool you already have, by you or by whoever built it. The report is written so that is possible.

Is this going to be a list of things I did wrong?

No. Almost everything I find is structural and was invisible to everyone involved, including whoever built it. The point is that you have been carrying a feeling you could not justify, and often there really is something underneath it. That is usually a relief rather than a telling off.

Do you use AI to do this?

Yes. I use the same practices I teach at Sacred Tech Sessions in all of my own work.

The measuring is automated, because measuring is work for a machine. Counting widths and checking contrast ratios is done more accurately by software than by anyone’s eye, including mine.

The judgement is mine. Noticing that something is wrong in the first place, understanding what your work is trying to be, and deciding which faults matter to you and which are beside the point. That part is judgement rather than measurement, and it is mine.

Can you look at more than one site?

Yes. Ask me and I will quote you properly rather than multiplying by the number of sites, because the second one is always quicker than the first.