You have found a small shop, a maker, a one-person software product or a freelancer you had never heard of ten minutes ago. It looks promising. You are about to hand over money to someone whose name you cannot verify against anything you already trust.
For the broader context behind this business guide, read How Business Directories Actually Make Money: The Five Models.
What follows is the check we run on entries here, compressed into something you can do yourself in under five minutes with no tools, no accounts and no special access. It is deliberately simple, because a check that requires effort is a check nobody performs.
Read this before you start: passing every check below guarantees nothing. These checks establish that a business appears to be real, appears to be trading, and appears to be reachable. They do not establish that it is competent, solvent, safe, or that it will do what it says. They cannot detect a determined deception. This is not fraud advice and it is not a fraud screen — we say the same thing about our own register, which cannot do that either. Treat all of this as a way of noticing obvious absences, not as protection.
What this is, and who it is for
This is for the person who wants to buy from small independent businesses and keeps hesitating at the last step because there is nothing familiar to hold onto. The instinct to hesitate is reasonable. The usual response to it — looking for reviews, looking for a badge — mostly does not help, because small businesses do not accumulate reviews and badges are claims about a process that is rarely published.
So instead of looking for reassurance signals, look for the absence of things a real trading business almost always has. Absences are much easier to spot than presences are to verify.
When five minutes is worth it, and when it isn't
Run these checks in proportion to what you stand to lose. Buying something small, from a site that will handle payment through a method you already use, with a delivery timescale of days? Skim two or three of the checks and get on with it.
Commissioning work, paying a deposit, buying something expensive, or dealing with anything where you will need to reach the business again weeks later? Run all seven, and pay particular attention to checks 2, 5 and 6.
If you found the business through a listing on some other site, note that the listing itself is not one of these checks. Why so many listings are for closed businesses explains why a listing tells you very little about whether anything is still there.
Check 1: does the site say what it sells, plainly?
Do this: open the homepage. Without scrolling, without clicking, and without reading anything in a slideshow, can you say what this business sells and to whom?
Time: ten seconds.
What it is a proxy for: whether anyone is minding the shop. A homepage that does not state the offer is usually not a mystery, it is a symptom. Either the site was built for someone who already knows what the business does, or it has not been revisited since the business changed. Both matter to you, because both suggest the front of the shop is not being watched.
How it fails honestly: a very established local business whose customers all arrive by word of mouth may have a homepage that assumes you already know. This is common and is not, by itself, a warning.
Check 2: is there a way to reach a human?
Do this: find the contact route. Not a chat widget, not a form with no address behind it — a specific route to a person: an email address, a phone number, a physical address, a named individual.
Time: thirty seconds.
What it is a proxy for: whether the business expects to be contacted after the sale. A business that intends an ongoing relationship makes itself reachable, because it has to be. A form with no other route means every future conversation happens on the business's terms, at the time of their choosing, with no record you hold.
This is the check we weight most heavily on entries here, and it is the one most worth your time. If something goes wrong, everything depends on this.
How it fails honestly: a sole trader working from home will not publish an address, and should not. That is a privacy decision, not evasion. We never publish a residential address for the same reason. Look for a named person plus at least one direct route instead.
Check 3: is anything on the site dated, and how recently?
Do this: look for anything with a date attached. A blog post, a news item, a stock update, a shipping notice, a copyright year in the footer. Note the most recent one you can find.
Time: thirty seconds.
What it is a proxy for: whether anyone has touched this site recently. A site with nothing dated on it anywhere is not necessarily abandoned, but it gives you no way to tell the difference between a business trading yesterday and one that stopped two years ago.
The footer copyright year is the weakest version of this signal, because it is often generated automatically and updates itself whether or not a human has been near the site. A dated post, a dated product update, or a dated shipping notice is worth far more.
How it fails honestly: plenty of thriving small businesses have not published anything in three years because they were busy working. A stale blog is not evidence of closure. Combined with several other stale signals, it starts to mean something.
Check 4: does the business identify itself?
Do this: look for who is behind this. A name, a trading name, a company identifier, an "about" page that says something specific rather than describing a category of business in general terms.
Time: forty seconds.
What it is a proxy for: whether anyone has attached their name to the thing. Self-identification is not verification — a name on a page proves nothing on its own — but the absence of any identifiable party is informative, because it means nobody has accepted association with the business in public.
Read the "about" page for specificity rather than warmth. "We are passionate about quality and committed to our customers" could sit on any site in any industry. "I make these in a workshop behind a converted garage and each one takes about a day" is text about a specific business.
How it fails honestly: some people trade under a brand and keep their own name off the site for personal-safety reasons. A consistent, specific trading identity with no individual named is different from no identity at all.
Check 5: do the payment and delivery terms exist and make sense?
Do this: find out how you pay, when you pay, what happens next, and how long it takes. Then check whether those things are stated anywhere before checkout.
Time: a minute.
What it is a proxy for: whether the business has actually done this before. Terms are boring to write and nobody writes them speculatively. A business with specific, slightly awkward, obviously experience-derived terms — "orders placed after Thursday ship the following week", "we ask for half up front on commissions over a certain size" — has been through the process enough times to know where it goes wrong.
What you are looking for is presence and coherence, not favourability. Terms that are strict but clear tell you more than terms that are generous and vague.
How it fails honestly: a business that is new, or that sells through conversation rather than checkout, may not have written terms yet. In that case ask, and treat the answer as the check.
Check 6: does anything contradict anything else?
Do this: compare what you have found. Does the contact page match the footer? Does the "about" page describe the same business the homepage sells? If there is a social profile linked, does it describe the same operation, and when did it last post? Does the delivery page contradict the checkout?
Time: a minute.
What it is a proxy for: whether the site is being maintained as one thing or has accumulated in layers. Contradictions are the most informative single signal in this whole list, because they are hard to fake in either direction. A business that has updated its offer but not its "about" page has changed and not noticed. A business whose linked social profile went quiet two years ago has changed how it operates and not updated the link.
Contradictions rarely mean deception. They usually mean neglect. Either way they are the thing to ask about before you buy.
Check 7: does it look maintained, or preserved?
Do this: stand back. Broken images? A newsletter signup that errors? Links to services that no longer exist? A shop section with everything out of stock? A "new for last year" banner?
Time: thirty seconds.
What it is a proxy for: the difference between a site somebody looks after and a site that is simply still hosted. This is the closest a quick pass comes to answering "is anyone there", and it is the composite of everything above. Mistakes that make a real business look abandoned lists the specific signals from the other side of the counter, written for operators trying to fix them.
How it fails honestly: this is the check most likely to be unfair to a genuinely good, genuinely busy small business. Which brings us to the important part.
When a good business fails several of these
This happens constantly, and it is the main limitation of the whole method.
A skilled maker with a two-year waiting list has no reason to maintain a website. A shop that does all its trade through the door does not need an up-to-date stock page. A freelancer who gets all their work by referral has a site that exists as a formality. Every one of them will fail three or four of these checks while being an entirely real, entirely competent business.
So weigh the checks by what they are actually telling you:
- Failing checks 1, 3 and 7 — the site is not maintained. This tells you about the site, not the business. Weight it low unless you needed the site to be current for a specific reason, such as prices or opening hours.
- Failing checks 2 and 5 — you have no reliable way to reach them and no idea what happens after you pay. Weight this high. It is the difference between a small business and an unreachable one.
- Failing check 4 — nobody has attached their name to it. Weight this in proportion to the amount at stake.
- Failing check 6 — something contradicts something. Weight this high regardless of amount, and resolve it by asking before you buy rather than by guessing.
And a general rule that saves a lot of hesitation: if the business is reachable, ask. A business that answers a plain question promptly and specifically has just passed a better check than any of the seven above. One that does not answer at all has told you something too.
What a clean pass does and does not guarantee
It guarantees nothing. This is worth repeating rather than tucking into a footer.
A clean pass on all seven checks means: this business appears to exist, appears to be trading, appears to be maintained, and appears to be reachable. Every one of those is an appearance, established from the outside, on a single day, by a person spending five minutes.
It does not mean the business is good at what it does. It does not mean it is financially stable. It does not mean it is insured, licensed or qualified. It does not mean your order will arrive. And it does not mean the business will still be there next month — a business can pass every check on a Tuesday and stop trading on the Friday.
We hold ourselves to exactly the same limit. A check on this site establishes that on a stated date, a named person found a business and it was recognisably the business we described. That is the entire claim, and our method says so at length. If you want the same checks run at scale rather than by hand, how we run them is the operating procedure in full.
The register is one place where these checks have already been run by someone else, with a date attached so you can see how old the answer is. It is small, it is narrow, and its absence of a business means nothing at all about that business. It is a shortcut for a few things, not a substitute for judgement about any of them.
One last exclusion. Nothing here touches what to do if a purchase goes wrong — payment disputes, refunds, remedies or rights. Those depend entirely on where you are, how you paid and what you bought, and any answer we gave would be wrong for most readers. We do not give it.
In Plain English
Seven checks, five minutes, no tools:
- Does the homepage say what they sell? Ten seconds, no scrolling.
- Can you reach a human? A named person and a direct route, not just a form. This is the one that matters most.
- Is anything dated recently? A dated post beats a footer year.
- Does anyone identify themselves? Specificity beats warmth.
- Do payment and delivery terms exist? Awkward and specific beats generous and vague.
- Does anything contradict anything? Usually neglect, occasionally more, always worth asking about.
- Is it maintained or just still hosted? The composite of everything above.
Good businesses fail several of these regularly, usually because they are busy. Weight checks 2, 5 and 6 heavily and the rest lightly. If you can reach them, ask — a prompt specific answer beats all seven.
And none of it is a guarantee. It is a way of noticing what is missing. The definitions behind the terms used here are in the glossary, and the checklists we give away free are at templates. If you are stuck between two of them, choosing between two of them picks up where this leaves off.
See how we check → /method/