The website wasn't the problem. Everything behind it was.
RNIB knew their website had problems. What they didn't yet know was how many of those problems had nothing to do with the website at all, and everything to do with what was propping it up.
Where it started
Find the problem before you fix the platform.
RNIB were preparing to replatform their website and needed to understand where the real problems were before they started, so the work could be properly prioritised. That meant stakeholder interviews, a hard look at the data, and a willingness to follow the evidence wherever it led.
It led somewhere uncomfortable. The root cause wasn't the website. Account data was split across three systems with no single view of a customer, and because the site didn't enforce logins, customers were rekeying their own details at the point of order. Rekeying is always error-prone. Orders got cancelled. Customers rang the help centre to re-place them. Expensive, trust-eroding, and entirely preventable.
The output was a prioritised list: fix this now, fix this next, fix this later. A clear brief before a single line of new platform code was written.
The rebuild
No taxonomy export. So we rebuilt it by hand.
When the replatform got underway, the first practical problem was blunt: the existing platform couldn't export the taxonomy. There was no starting point to work from, so the whole thing had to be rebuilt manually, with the full catalogue in front of us.
Once the data was in hand, the taxonomy was restructured properly and a bespoke metadata schema was designed in Airtable, linked directly to the pages. Shopify is excellent for smaller, more contained catalogues. For something the size and complexity of RNIB's range, covering specialist equipment, everyday accessibility products, and everything in between, its native metadata structure doesn't hold. It needed something designed specifically for it.
The D2C site launched on Shopify in early 2025. In the six months following, overall sales volume held steady while call centre volume dropped significantly. Fewer errors, fewer cancelled orders, fewer calls. The work paid for itself.
Then the B2B site
A second catalogue. A different set of needs.
From mid-2025, RNIB came back for the B2B site. Same platform, different audience, different catalogue logic. The work here was a metadata and taxonomy update rather than a full rebuild, reframing the structure to fit a wholesale and trade context rather than direct-to-consumer. The D2C metadata got a refresh at the same time, keeping both sites consistent as the catalogue evolved.
Sound familiar?
Replatforming without fixing the foundations first is very expensive.
A new platform won't fix broken taxonomy, inconsistent data, or a metadata structure that was never designed for your catalogue. It'll just make the problems faster.
The diagnostic finds out what's actually going on before you spend anything on a fix.
Book the diagnostic