Of everything that goes wrong on a small business website, one failure turns up more than any other, and it is not a design problem or a technical one. The first screen does not say what the business sells.
It is the most common single reason a submission does not clear our entry bar. A reviewer opens the site, reads the first screen, and cannot describe in their own words what a person would be buying. When that happens there is nothing to record, so the submission goes no further. The patterns behind those refusals are collected on the refusal file, and this one sits at the top of the list.
The good news is that it is also the cheapest thing on the site to fix.
What this is, and who it is for
This is for anyone who suspects their site is clear and keeps meeting people who prove otherwise. You explain the business at a market stall or on a call and watch it land immediately. The same explanation is nowhere on your homepage.
The fix does not need a redesign, a new theme, a photographer or a budget. It needs one sentence and a decision about where to put it.
The test: read your first screen to someone who doesn't know you
Do not evaluate this yourself. You will pass. You know what the words mean.
Find someone who has never heard of your business. A neighbour, a relative, somebody at the next table. Open the homepage on a phone, hand it over, and give them ten seconds. Then take the phone back and ask one question:
What do they sell?
Do not prompt. Do not say "keep scrolling". Do not explain afterwards until they have answered. Write down exactly what they say, in their words.
There are three outcomes. They name your product or service, in which case this post is not for you and you can go and check the rest of the ten-item checklist. They describe a category vaguely, "something design", "some kind of consultancy", which means the page is close but not landing. Or they describe a mood, an ambition or a set of values, which means the first screen is doing something other than the job.
Run it on three people if you can. One confused reader is a data point about that reader. Three is a data point about the page.
Why this happens to good businesses specifically
This is not a failure of care. It happens most reliably to people who know their trade extremely well, and there is a mechanism behind it.
Once you understand something thoroughly, you lose the ability to remember not understanding it. Reading your own homepage, you supply the missing half of every sentence automatically, from the enormous store of context in your head. The page feels complete because you complete it. Every time you look, you complete it again.
There is a second push in the same direction. The longer you run a business, the more of your attention goes to the parts your competitors get wrong, your philosophy, your process, the corner you refuse to cut. Those are the interesting parts to you. They are the differences between you and the other options. But a stranger has not chosen the category yet, so a difference is useless to them. They are still working out what the category is.
So the expert writes about the interesting part and the stranger needs the boring part, and neither of them is being unreasonable.
The four things people write instead of what they sell
Almost every first screen that fails this test has one of these four things in the place where the plain sentence should be.
A values statement. "We believe good work takes time." "Craft, care and community." True, often deeply felt, and compatible with a bakery, a law firm or a furniture restorer. A stranger cannot narrow anything down from it.
A process description. "A collaborative, discovery-led approach that puts you at the centre." This describes how you work with a client who has already decided to hire you. It answers question four for a reader who is still on question one.
Credentials. "Twenty years' experience. Fully insured. Award-winning team." These are reasons to choose you over someone else in your field, which is exactly the right material for later in the page. On the first screen they assume the reader already knows what field that is.
A tagline. "Making the everyday extraordinary." "Your partner in progress." A tagline is a compression of something the reader already understands. Put one in front of somebody who understands nothing yet and it compresses air.
None of these are bad writing. Three of the four are things a stranger will genuinely want, twenty seconds later. The problem is only ever placement.
The fix: one sentence, and where it goes
Write a sentence containing the thing you make or do, and the person you make or do it for. Plain nouns. No metaphor.
The template, which you should abandon as soon as you have something better:
We [make / sell / repair / design / teach / write / run] [the thing] for [the kind of customer], [where or how].
That is it. It goes at the top of the homepage, above anything that requires scrolling, in the largest text on the page. It goes before the values, before the process, before the credentials, before the photograph.
Three rules that stop it drifting back into fog.
Use the word your customer would use, not the word your industry would use. If people phone up and say "curtains", the sentence says curtains.
Include a noun somebody could buy. If the sentence contains no purchasable object or bookable service, it is not finished.
Say what you do not do, if that is what makes it clear. "Dog grooming only, no boarding" removes more confusion than another adjective would.
Keep the tagline if you love it. Put it underneath.
Worked rewrites of four anonymous first screens
The four examples below are invented. They are composites written for this post to illustrate the pattern. They are not real businesses, and we never publish a teardown of a real site, good or bad.
Invented example 1: a two-person joinery workshop.
Before: "Honest craft, built to last."
After: "We make fitted wardrobes, alcove units and kitchen cabinets to order, in oak and painted birch ply, from a workshop in the north of the county."
The old line survives as a strapline under the new one. Nothing was lost and a stranger can now decide in two seconds whether to keep reading.
Invented example 2: a freelance bookkeeper.
Before: "Taking the stress out of your numbers so you can focus on what you love."
After: "Bookkeeping and quarterly returns for sole traders and small limited companies, working remotely, on fixed monthly fees."
The original sentence is about a feeling. The replacement contains four filters, service, customer type, working method and pricing model, so most of the wrong enquiries stop before they start.
Invented example 3: a small ceramics studio.
Before: "A collaborative, considered approach to making, rooted in place."
After: "Hand-thrown stoneware tableware, sold in small batches, plus evening wheel classes for beginners."
The process language moves down the page, where somebody who has decided they like the pots will actually read it.
Invented example 4: an independent software tool.
Before: "The modern platform for ambitious teams."
After: "A shift-planning tool for cafés and restaurants with five to fifty staff. Web and phone. Free to try without a card."
"Platform for ambitious teams" is compatible with several hundred products. The replacement names the job, the customer and the size band, which is what lets a manager of six people recognise themselves.
What to do when you sell several unrelated things
Plenty of small businesses genuinely do two or three things that have nothing to do with each other. A photographer who also runs a print shop. A translator who teaches. A shop that repairs the thing it sells.
Three ways out, in order of how often they work.
Name them all in one sentence. "Portrait photography, and fine art printing for other artists." Two clauses, no confusion. This handles most cases and people avoid it because they think it looks unfocused. It reads as busy, which is fine.
Split the first screen into named routes. Two or three clear blocks, each labelled with the plain noun, each linking to its own page. The homepage becomes a signpost. This works when the two audiences barely overlap.
Give the smaller thing its own site. Only worth it when the second business has its own customers, its own vocabulary and enough volume to justify maintaining another site. Two half-maintained sites are worse than one busy one, and the seven-item shopfront floor applies to each of them separately.
What does not work is choosing a word abstract enough to cover both. That is how "solutions" happens.
Checking it worked
Same test, different people. Fresh readers only, because anyone who has seen the old version is now contaminated.
Ten seconds. Take the phone back. Ask what they sell. If they can name the product or service in ordinary words, the sentence is doing its job.
Then check two things that tend to break when the top of a page changes. Read the page on a phone to make sure the new sentence has not been pushed below the fold by a header or a cookie banner. And read the rest of the site once through, looking for anything that now contradicts the new sentence, an old services list, a stale menu item, a heading that describes the business you used to be. Contradictions are a separate item on the checklist and they get introduced by exactly this kind of edit.
The same clarity problem shows up in a longer form when a whole site is assessed rather than a first screen, which is the patterns this shows up as in refusals.
In Plain English
You cannot read your own homepage as a stranger, because you already know the answer to the only question that matters. So you write about the interesting parts, and the interesting parts are only interesting to people who already know what you sell.
Put a plain sentence at the top saying what you make or do and who for. Keep the values, the process and the credentials, and move them down. Then hand your phone to somebody who has never heard of you, give them ten seconds, and ask what you sell.
It takes about ten minutes and it is the highest-leverage change most small sites can make. It also has nothing to do with search engines, which this site does not cover.
Questions this raises
Won't a plain sentence make me sound generic?
It makes you sound like the thing you are, which is the opposite of generic. The vague version is what everyone in your field already has, because everyone reaches for the same abstractions. Specific nouns are what nobody else's homepage has.
My designer says the first screen should be an image.
It can be. The sentence goes on the image, or under it, in text a screen reader and a slow connection can both reach. The argument is about the words, not the picture.
Does fixing this improve my chances of being listed?
No. Nothing improves your chances except clearing the entry bar, and we do not track who has read what. We publish this because it is the failure we see most, and describing it is more useful than keeping it to ourselves.
Run the findability checklist to see what else a stranger cannot answer. The free, ungated version lives with the other templates, and the structural companion is shopfront basics.