Why move off WooCommerce in the first place?

The shop's WooCommerce stock counts had drifted away from the till. Two systems with their own ideas about inventory means at least one of them is wrong, and neither side trusts the website. Replacing both with a single cloud POS that powers till and website off one stock ledger removed the drift entirely. The decision wasn't about WooCommerce specifically — it was about not running two unsynchronised systems.

How does the AI rewrite supplier feeds without making things up?

Hallucination is contained by what the model is given to work with. The pipeline only sees the raw supplier line — product code, spec, price — plus a strict template and tone-of-voice rules taken from the shop's existing best listings. It rewrites benefits, fit notes, and use cases from those inputs only, in the shop's voice. The model isn't inventing facts; it's restructuring and rephrasing what the supplier already provided. Items the supplier feed lacks key info on are flagged for human attention rather than auto-listed.

Could a smaller shop expect similar results?

The 100× lift was specific to this shop — strong in-store reputation, weak online execution, and a product range with real online demand. The pattern that's reusable is the systems part: one stock ledger, in-store and online drawing from the same database, AI doing the slow listing work. Those structural changes pay off across most retail with an online channel, even if the absolute numbers differ.

Back to case studies

Independent retail · 2025

Bike Shop Systems Integration.

Online sales from ~£2k a year to £200k+ in year one.

Online sales, year one

£200k+

Starting point

~£2k / year

Time to list new stock

Minutes, not hours

What wasn't working.

The shop ran on an ageing WordPress/WooCommerce store that had drifted away from the till system. Stock counts on the website and at the counter were never the same. Listings were thin — product codes, a line of spec, a photo — and buried in search. Adding new products was a manual slog: copy from the supplier sheet, rewrite, upload, check.

The shop was selling well in-store but online had stalled at about £2,000 a year. Not because the demand wasn't there — because the website couldn't be kept up-to-date fast enough to be worth buying from.

What we built.

One stock ledger, two fronts

We replaced the old stack with a cloud POS that runs both the till and the website off a single product database. What sells in-store updates online instantly; what sells online drops out of the in-store stock. No more reconciling.

AI-powered listing pipeline

Supplier product feeds now pass through a Claude-based pipeline that rewrites every listing into proper SEO-aware copy — benefits, use cases, fit notes — in the shop's own voice. A product that would have taken 20 minutes to list properly now takes less than a minute.

Room to grow

The system handles multi-channel fulfilment, click-and-collect, and seasonal range changes without a rebuild. The shop can add lines, suppliers, or locations and the platform scales with them.

What changed.

Online sales crossed £200,000 in the first year on the new platform — a 100× lift from the old setup. More importantly the daily admin load dropped: the owner is no longer spending evenings updating listings, and the weekly stock reconciliation that used to eat hours is gone.

The shop now uses its website as a real second channel. Range changes and price updates happen once, propagate everywhere, and new stock is live within the day it arrives.

Independent retail · WooCommerce → Cloud POS + AI copy pipeline

Got something like this you want built?

Start a project

Frequently asked questions.

Why move off WooCommerce in the first place?
The shop's WooCommerce stock counts had drifted away from the till. Two systems with their own ideas about inventory means at least one of them is wrong, and neither side trusts the website. Replacing both with a single cloud POS that powers till and website off one stock ledger removed the drift entirely. The decision wasn't about WooCommerce specifically — it was about not running two unsynchronised systems.
How does the AI rewrite supplier feeds without making things up?
Hallucination is contained by what the model is given to work with. The pipeline only sees the raw supplier line — product code, spec, price — plus a strict template and tone-of-voice rules taken from the shop's existing best listings. It rewrites benefits, fit notes, and use cases from those inputs only, in the shop's voice. The model isn't inventing facts; it's restructuring and rephrasing what the supplier already provided. Items the supplier feed lacks key info on are flagged for human attention rather than auto-listed.
Could a smaller shop expect similar results?
The 100× lift was specific to this shop — strong in-store reputation, weak online execution, and a product range with real online demand. The pattern that's reusable is the systems part: one stock ledger, in-store and online drawing from the same database, AI doing the slow listing work. Those structural changes pay off across most retail with an online channel, even if the absolute numbers differ.

Need something like this built.

If this case study sounds like the kind of problem you have, here's where we'd start.

Let's talk.

Tell us what you're trying to do. We'll reply within one working day. If we're not the right team for it, we'll say so.

→Reply in one working day

→First call is free — no pitch

→If we're not the right fit, we'll say so

Prefer email?hello@tryfantech.com

Based in the UK, working with businesses across the country