Evidence
Why law-firm websites make potential clients choose a practice area before explaining their problem
We audited 50 law-firm homepages across the UK and Ireland, with criteria fixed before we looked at a single site. Most offer a menu of practice areas and no way to say what happened to you. The full findings, including the ones that cut against us.
Jakov Manojlovski, founder of Predvora · August 30, 2026 · 7 min read

Imagine something has gone wrong. You have been served with papers, or your deposit is gone, or your business partner has stopped returning calls. You search for a law firm in your city and land on one of the most polished legal homepages in London. You arrive wanting to do exactly one thing: tell someone what happened.
Here is what that page offers you instead: audience menus, a sectors list, a services list, and a search box asking how it can help. There is no phone number anywhere on the page. No email address. No form. It is the politest, most expensive way a website can say: work out what we call your problem, then come and find us.
That homepage is real, and we did not go hunting for it. It turned up in an audit of fifty law-firm homepages across the UK and Ireland, sampled mechanically and scored against criteria we committed to in writing before examining a single site. This article reports what we found, including the findings that cut against the argument we came in with. The methodology is at the end.
Choose before you speak
Thirty-six of the fifty homepages (72%) offer the arriving visitor navigation and nothing else: menus, practice-area tiles, generic contact buttons. Nowhere on those pages is there any element that invites you to describe your situation in your own words.
And the navigating is not a small ask. The median homepage presents twenty practice-area labels to choose between. A quarter present more than forty-seven. Six present more than a hundred, and two firms' menus enumerate over two hundred distinct legal-service labels before the visitor has said a single word. Every label is written in the firm's vocabulary, not the client's. A person whose problem is "my landlord will not return my deposit" is left to work out, unaided, whether that makes them a property litigation matter, a landlord and tenant dispute, or civil litigation.
Where a homepage does attend to the visitor, it is usually in one specific way: with a phone number. Fifteen homepages (30%) put a number above the fold on desktop. One national firm greets visitors with a header that asks, in effect, whether they have a question, and answers it with a phone number. The invitation is real. The website itself just cannot receive the question.
At the other end, twenty-one homepages (42%) show no phone number anywhere on the page at all, and on one global firm's homepage the word "contact" does not appear even inside the opened menu. Reaching a human there means opening the menu, choosing locations, and drilling into an office page.
The questions nobody answers
A potential client arrives with roughly three questions. Can you help with my situation? What does it cost to find out? How quickly will someone respond?
On thirty-seven of the fifty homepages (74%), none of the three is answered. Six homepages state what an initial consultation costs, including "free". Five make any promise about responsiveness or availability. Seven say anything at all about how the firm charges. Many footers link to a pricing information page; the homepage itself stays silent.
What cut the other way
An audit that reported only its convenient half would be marketing, so here is the other half, with the same prominence.
"You cannot reach a human from a law-firm homepage" would be false. On thirty of the fifty (60%), a usable contact channel sits on the page with zero clicks needed. Be careful with that number: it counts channels you reach by scrolling, including phone numbers that live in the footer. The stricter measure, a phone visible above the fold on desktop, is the 30% above. On mobile, where a call is one tap away by design, thirteen homepages (26%) put a tappable number above the fold.
"No firm invites your own words" would also be false. Fourteen homepages (28%) carry some kind of free-text intake. Twelve of the fourteen park it at the bottom of a long page, which supports "unattended at the moment you arrive" but not "unattended entirely".
And two firms put genuine listening at the top of the page. One opens its homepage with a callback panel: choose a category, describe your situation in a free-text box, book a time that suits you. Another, a small high-street firm, lists an out-of-hours mobile number above the fold and staffs the only live chat we found in the whole sample, and that chat opens with a blank box and an open question rather than a script. Exactly one firm in fifty publishes concrete prices on its homepage. It is worth saying plainly that the counterexamples we found were consumer-facing and high-street practices rather than corporate firms. We did not score firm type, so treat that as an observation about who the counterexamples happened to be, not as a measured comparison between segments.
This is not a law-firm failing
It would be easy to sneer, and it would be wrong. Every one of these fifty homepages is doing what almost every business homepage does, and that is precisely why the pattern matters beyond law.
The structure is rational. Firms organise their websites around their own taxonomy because that is how firms think and how responsibility is divided inside them. Practice-area pages are how firms win the searches that matter to them. Silence about prices is partly professional caution in a regulated market. Even the near-total absence of chat is a reasonable scar: a decade of scripted chatbots taught firms and visitors alike to close the bubble, an argument we made in the chatbot piece. Each decision is defensible on its own. Their sum is a homepage that cannot be told anything.
The cost is invisible, and we did not measure it
This audit measured what homepages present, not what visitors do. We will not claim to have counted lost clients, because we did not. What we can say is structural: the person most of these homepages are hardest on is the one who does not yet know what their problem is called, and that person is the new client. A visitor who leaves with an unanswered question was not unconvinced, they were unattended, an argument we made at length in the manifesto. Nothing in a firm's analytics rings when someone closes the tab.
How we ran this audit
- Sample. The ten largest UK and Ireland cities by population. For each, a logged-out Google search for "law firm" plus the city name, walking the organic results in order and taking the first five qualifying firms, with every skip logged with its reason. No substitutions after selection; n = 50.
- Criteria before data. The scoring rules were committed in writing before any site was examined: entry orientation (navigation-first versus problem-first, and the kind of any intake), clicks to a human contact, phone visibility on desktop and mobile, practice-area label count, and whether consultation cost, response time, or fee model is stated on the homepage. Ambiguous cases were scored against our own thesis by rule.
- Audit date. 2026-08-30. Websites change; this is a snapshot.
- Definitions worth repeating. "Zero clicks to a human" counts channels reached by scrolling, footers included. "Above the fold" is measured separately, on desktop and on mobile viewports.
- Known biases. Sampling from city search results over-represents large, search-optimised firms, so these numbers describe the firms a searching client actually finds, not the whole profession. Practice-area counts include granular menu topics where firms expose them, and a few sites enumerate nothing on the homepage at all, which pulls the median down.
- Anonymity. Aggregates and paraphrased vignettes only, per our naming policy. No firm is named without written permission.
- Product surface. None. No Predvora feature was used in this audit; it is a reading of public homepages, nothing more.
The manifesto argued that a business website should represent, not just explain. This audit is what that argument looks like on the ground in one profession: fifty polished, professional, rational homepages, thirty-six of which cannot be told anything at all, and only two of which listen at the moment a client arrives carrying the one sentence that matters. Your website explains. It should represent. In law, right now, it mostly navigates.
Written by Jakov Manojlovski
Founder of Predvora

