Case studies
The same problems keep showing up. In very different places.
After nearly 30 years, four failure modes come up time and again. The clients are different. The sectors are different. The underlying problem is usually the same. Here's what that looks like in practice.
The Google loop
Visitors leave your platform to find what should be on it.
Are your visitors Googling their way back? →Public sector, NHS
During a pandemic, people were searching for COVID symptoms and leaving NHS.uk to find them on Google. Search accuracy went from 30 to 60 percent up to around 95 percent.
Tourism, attractions
People were visiting the website but not buying tickets, and queuing in person instead. £5k budget. £1m extra in the first month.
The noise problem
Too much, structured badly. Visitors can't see the thing they came for.
Does your catalogue bury the thing people want? →Charity, retail
A catalogue too complex for Shopify's native structure, account data split across three systems, and customers rekeying orders that kept getting cancelled. Sales held. Call centre volume dropped significantly.
Education, publishing
24,000 websites, 56,000 domain names, and a brief to reduce it all to one. The answer was more considered than that. The structure was still valid five years later.
The spelling problem
Search can't connect what visitors type with what you actually have.
Does your search speak a different language to your visitors? →The word-retrieval problem
Visitors know what they want. They just can't find the words your platform uses for it.
Do your visitors know what they want but not what to call it? →Public sector, NHS
Clinical forms that couldn't anticipate how clinicians actually completed them. Designing the logic around the humans using them changed everything. Already saving money on paper, printing, and billing errors.
Higher education
Prospective students who couldn't articulate what studying for a degree would mean for them until they saw it reflected back. Time to convert dropped from two years to six months.
Not every problem fits a category
Some work is harder to label.
Not all findability problems start with a search box. Sometimes the problem is in the data, the structure, or the commercial model sitting underneath the platform.
Energy, petrochemical
A mountain of fuel card data with no commercial packaging. Designed a suite of API data products, from enterprise ERP feeds down to the man with a van.
Charity, retail
Three phases over several years, two Shopify sites, and a taxonomy rebuilt entirely by hand. The full story is more complex than any single failure mode.
Next step
Recognise any of this? The diagnostic will tell you which one you've got.
Most clients come in thinking they have one problem and find out they have another. The diagnostic is how we find out what's actually going on, before anything gets fixed.
Book the diagnostic