Case studies
A national law firm needed to rebuild a website that had grown to around 17,000 pages, and to appoint an agency to do it. I ran the project independently and under NDA, in the firm's corner rather than the agency's. A full crawl of all 17,000 pages, a GA4 review, an internal survey and a benchmark against thirteen peer firms produced an evidence-led brief, and the agency was chosen through an ITT, shortlist and pitch day on fit rather than on price.
The website had grown over years, built up by a lot of different hands, until it reached roughly 17,000 pages. At that size nobody holds the full picture any more, and the firm knew it. They wanted a refresh and a new agency to deliver it.
The risk was doing what most firms do at this point: picking an agency and signing a brief before anyone had understood what they actually had, who used it and what was genuinely not working.
On top of that sat the realities of a large firm. A lot of stakeholders with different priorities, a low tolerance for anything sloppy, and a non-disclosure agreement covering the work. The job was to bring order to all of it and make sure the money got spent well.
I came in as the independent project lead, in the firm's corner rather than the agency's, with nothing riding on the build.
The first phase was understanding what was actually there, before a word of a brief got written. That meant a proper analytics review in GA4, an internal survey of the people across the firm who rely on the site, and a full Screaming Frog crawl of all 17,000 pages, which is the only honest way to see the shape of an estate that size. Alongside it we reviewed the technical health, hosting and page speed, the user journeys, the design across devices, and the whole tech stack including how the CMS and CRM talked to each other. We also pinned down the areas that mattered most commercially, from the people and expertise pages to a key private wealth offering and the firm's productised services.
Next we benchmarked the firm against thirteen peers, running Screaming Frog across each and Raven and SEMrush over the strongest, looking at readability, Core Web Vitals and what each did well. Not to copy anyone, but to know exactly where the firm stood.
Then we got the teams in a room. A workshop to set objectives, check and refine the personas, and do the big unglamorous job of consolidating a tangle of sectors and services into a structure that made sense.
Only after that did we build the brief, in order: scope and requirements, personas and user journeys, a content matrix, a simplified site structure, and last of all the right CMS to sit underneath it. We also checked the data protection and cookie position against the ICO's audit framework, because for a law firm that cannot be an afterthought.
With the brief settled, the agency selection ran as a proper process. An invitation to tender built straight from the brief, a defensible shortlist, a full review of the responses against the requirements, and a pitch day where the shortlisted agencies presented to the team and answered for their thinking. Throughout, a clear project management proposal, monthly progress reports and regular check-ins kept it all moving.
The firm went from 17,000 pages and no shared view of them to an evidence-led brief the whole firm recognised, a clean and defensible shortlist, and an agency chosen on fit and evidence rather than on the cheapest quote or the flashiest deck. The decision could be justified to anyone who asked, which for a firm spending significant money is the point.
The hard numbers so far are about the rigour of the work rather than post-launch performance: 17,000 pages analysed, thirteen peer firms benchmarked, a full ITT, shortlist and pitch process, and monthly reporting from start to finish.
The build follows from here, so the traffic and conversion figures will come later. I would rather say that plainly than dress up a number that isn't in yet. The part that decides whether the money is wasted, understanding the site and buying the right partner, was done properly first.
If you are sitting on a large, sprawling website and the instinct is to jump straight to a shiny new one, this is the argument for slowing down at the start. Understand what you actually have, turn it into a brief built on evidence, then run an agency choice you could defend to your board. Do it in that order and even a project this big stays calm and the spend stays justified. And if discretion matters, it can all be done under NDA.
Running a selection from the client side sits under advise. On the order that keeps a legal website project sane, see how to run a law firm website project without wasting money.
Evidence first, technology last. I start by understanding the existing site in real detail, turn that into a clear brief, then run a fair process to choose an agency, and keep the whole thing governed with regular reporting.
Because most large website projects go wrong before the build, when a firm signs a brief for a site nobody really understood. Understand what you have, then rebuild.
Through a proper, defensible process. An invitation to tender built from the brief, a shortlist, a full review of responses and a pitch day. Choose on fit and evidence, not the cheapest quote or the flashiest deck.
Yes, and often do. This project was delivered under NDA, which is normal at this level.
I build it in at the brief stage, checked against the ICO's audit framework, rather than fixing it after launch.
Most clients begin with a conversation. No pitch, no hard sell.
Just a straightforward discussion about where you are and whether I can help.
Book a free 30-minute call