The farewell sale that never started

An Instagram ad caught me on Saturday morning. Waxed canvas holdall on a workshop bench, handwritten card propped against it: farewell sale, up to 80% off, final stock, signed Jack. Underneath, a letter. After 29 years at the same workbench Jack was giving up ready-made bags to go custom-only, and what was left was going at the deepest discounts he'd ever offered.

I like a good bag. I clicked.

The product page had everything you'd expect from a well-run shop. £619.95 crossed out and £154.95 in its place, a countdown to midnight, "Only 7 in stock". Rated 4.9 from 1,247 reviews, trusted by over 5,000 customers, six testimonials with photos from Manchester, Dublin and Cork, two of them mentioning the handwritten note Jack had tucked into the parcel. Free delivery. Thirty-day returns, no questions asked.

Then the About section, and the photo of Jack in his workshop, and something in it was wrong. I couldn't have told you what at that point. The way he was holding the bag, maybe. The tools on the wall were a touch too arranged.

Five minutes on a phone

I wasn't at a desk, so this is what I could check from the sofa.

The domain was registered on 4 September 2026. The ad ran on 13 October. Jack has been trading for 29 years.

The Internet Archive had sixteen captures of the domain between 2018 and 2025 and none of them held anything resembling a bag shop. Whatever lived there before had nothing to do with Jack. A lapsed domain, picked up and rebuilt, and that gets its own post next week.

The About photo is a perfect 1024 by 1024 pixel square, which is the shape generative tools produce unless you tell them otherwise. An image detector put it at 99%. I've no problem with AI imagery as such, I use it for most of the banners on this blog. Illustrating a story is one thing. Manufacturing the person telling it is another.

Back at a laptop that evening, the rest unravelled quickly.

Every product image on the site was uploaded on 4 September at 18:41. Thirty-seven products, live in one sitting, on the day the domain was bought. Shopify's product data sits in the page source if you know where to look, and for the bag I'd been eyeing up it shows stock tracking switched off. "Only 7 in stock" is a line of text somebody typed. Nothing is counting.

The product page promised returns with no questions asked. The returns policy, two clicks away, says you pay the return postage, the bag has to be unused with tags attached, and refunds on sale items come as store credit for the website. Everything on the site is a sale item. Store credit, for a shop that's closing.

No company name anywhere. No registered address or phone number, no VAT number, one email address beginning "team@" on a site where Jack says he packs every parcel himself. The policy page calls it a closing-down sale. The About page says he isn't closing.

I'm not going to name the site. It would hand them traffic, and the domain will be dropped and replaced long before this post stops being read. The pattern is what matters, because it's a template, and I'd put money on there being hundreds of Jacks.

The template is borrowed from real retail

Every device on that page is one we've built for clients, or told them to build. Scarcity messaging and countdown timers. A founder story on the About page. Review counts next to the price, free delivery badges, the mailing-list pop-up with a discount attached. That's the standard conversion kit for a small e-commerce business, and it's standard because it works.

The fakes have lifted the whole kit, put a sob story on top, and are running it with ad budgets no genuine small brand can match, in the eight weeks before Christmas. Which? found seven ads in a single week claiming a well-known high street chain was closing down with a farewell sale. Santander has warned about closing-down scams that appear within days of a real retailer announcing store closures. The ASA has had a programme on "closing down" claims in social ads running since 2024 and the upheld rulings keep coming, several of them this October.

So genuine retailers pay twice. Once when a customer gets burned by a Jack and tightens up on everything that looks similar. Then again, more slowly, as the urgency devices that do work start to read as scam signals. A real sale with a real countdown now sits in the same visual language as the fakes, and the people being trained to spot that language are your customers.

Proof of existence

What does a real shop have that a template built in an afternoon can't copy? Mostly time, and accountability. A drop-and-run operation can afford neither.

Your domain's age is an asset. Ten years of archive captures showing the same business selling the same things is something money can't buy quickly, and rebrands throw it away far more often than they should. More on that next week.

Then the footer. Company number, registered address. UK company law requires them on your website anyway, and the point is that Companies House is public, so a sceptical customer can confirm you exist in about twenty seconds. A phone number somebody answers does the same job for the people who'd rather ring.

The About page should be about people who exist, in your own premises, with names that match what a visitor would find on LinkedIn. One decent photo of a real person in a slightly untidy office beats a flawless portrait of nobody.

Reviews are the one I'd move first. A count typed into your own page is now worth very little, since "1,247 reviews" costs nothing to write. On Trustpilot or Google or Feefo, with dates, and with the odd three-star in among them, they're harder to fake and easier to check.

Social history counts for more than it used to. Instagram shows an account's join date and the country it's run from under "About this account", and Meta's Ad Library shows what any page is currently advertising. Eight years of posts and the replies underneath them are hard to fabricate. Jack's account was a few weeks old.

Returns policy: write it yourself, in your own words, and make sure it says what the product page says. A reasonable customer reads the policy before a £150 purchase, and a template contradicting the sales copy is one of the loudest tells there is.

If you run a sale with urgency attached, give it a real end date and honour it. The ASA rulings I mentioned were against legitimate retailers who let "last chance" run for months. A fake countdown on a real shop puts you in Jack's category, with the regulator's attention as a bonus.

Stock counters deserve their own line. If your platform can show real stock, show it. If it can't, take the counter off. "Only 7 left" with nothing behind it is now a scam signal, whoever's site it's on.

Where this goes

Trust used to be the default online and scepticism the exception. For a first-time visitor arriving from a social ad that has flipped, and it will stay flipped for as long as the Jacks are profitable.

The retailers who'll do well through peak are the ones who make themselves easy to verify. Most of what that takes is already true of a real business. The work is putting it where a sceptical stranger will find it in the first thirty seconds.

If you'd like a second pair of eyes on how your site reads to that stranger, that's a conversation we have often, and it usually starts with a short discovery session.