Six of 20 legal tech homepages we opened on 21 October 2026 scored zero on clarity: their headline and the line under it named no product category, no buyer and nothing concrete the product does. All six sell mainly to larger firms, corporate legal departments or both. The four practice management sites in the sample, which sell mostly to small firms, averaged 2.25 out of 3 on the same tests.
Below are the nine tests, every measurement and what the top scorers do that others could copy. They are the same tests we ran in 24 B2B SaaS Websites Scored on Clarity, Speed and Demo Path (2026), so the two samples compare directly.
The rubric: nine pass or fail tests
Each test passed scores one point, so each dimension scores 0 to 3 and the total is out of 9.
Clarity, three tests. C: the hero names a product category buyers already use. A: it names who the product is for, as a role, team or company type. S: it gives one concrete specific, such as named functions, a standard or a number, beyond the category itself. The basis is Nielsen Norman Group's first homepage guideline, which asks for a tagline that "summarizes what the site or company does", and applies it to well-known companies too. The same group's dwell-time analysis of 205,873 pages found the first 10 seconds decide whether a visitor stays.
Speed, three tests. Time to first byte of 0.8 seconds or less, web.dev's "good" line. HTML transfer of 155 KB or less, and of 78 KB or less: the 90th and 75th percentiles of mobile home page HTML in the Web Almanac's 2025 page weight chapter.
Demo path, three tests. A demo, sales or trial form one click or less from the first screen; five fields or fewer on its first step; and a stated response time or call length.
Clarity method. The hero means the headline, the line under it and any label directly above the headline, read at 1,024 by 768 pixels without scrolling. Counting the label is the one change from the SaaS rubric, and it changed one result: Smokeball's. "Legal AI" on its own did not pass the category test, because it names the technology, not the job. "AI contract review" did. Cookie banners, which our connection triggered on most sites, were ignored.
Speed method. Google's PageSpeed Insights API answered with a 429 "quota exceeded" error again, so we used curl: five requests per homepage between 09:37 and 09:42 UTC from one connection in Serbia, taking the median time to first byte and the median gzip size of the HTML document. Browsers usually receive brotli, which is smaller, so our sizes err high.
Demo path method. We followed the most prominent demo, sales or trial action in the first screen and counted visible text boxes, dropdowns and text areas on the first step, in a rendered browser at 1,280 pixels wide. A phone number with a country picker counts as one field; a lone consent box counts as none. We submitted nothing.
Who is not here. We left out USLege, a legal-tech redesign we worked on, so we are not scoring our own work.
The scorecard: clarity and speed
Letters show which clarity tests a hero passed: C for category, A for audience, S for a concrete specific. Platform comes from our CMS check of 80 legal tech vendors, published as Legal Tech Website Platform and CMS Statistics 2026.
| Site | Product | Platform | Clarity | Passed | TTFB (s) | HTML (KB) | Speed |
|---|---|---|---|---|---|---|---|
| Clio | Practice management | Not identified | 2 | AS | n/a | n/a | n/a |
| MyCase | Practice management | Astro | 2 | CS | 0.12 | 101 | 2 |
| Smokeball | Practice management | Webflow | 2 | CA | 0.04 | 57 | 3 |
| PracticePanther | Practice management | WordPress | 3 | CAS | 0.06 | 54 | 3 |
| Filevine | Case management | Custom | 1 | A | 0.73 | 232 | 1 |
| Litify | Case management | Webflow | 0 | none | 0.04 | 36 | 3 |
| Lawmatics | Intake and CRM | Webflow | 2 | CS | 0.04 | 108 | 2 |
| Harvey | Legal AI | Next.js | 1 | A | 0.15 | 78 | 3 |
| Legora | Legal AI | Framer | 1 | A | 0.10 | 64 | 3 |
| Spellbook | Contract drafting | Webflow | 2 | CS | 0.04 | 56 | 3 |
| EvenUp | Personal injury AI | WordPress | 2 | AS | 0.24 | 55 | 3 |
| Luminance | Contract AI | Framer | 0 | none | 0.10 | 75 | 3 |
| Ironclad | Contract management | WordPress | 1 | C | 0.35 | 89 | 2 |
| Juro | Contract management | Webflow | 1 | C | 0.04 | 38 | 3 |
| SpotDraft | Contract management | Webflow | 3 | CAS | 0.05 | 71 | 3 |
| LinkSquares | Contract management | Next.js | 1 | C | 0.13 | 34 | 3 |
| Everlaw | eDiscovery | Netlify, Contentful | 0 | none | 0.18 | 59 | 3 |
| DISCO | eDiscovery | Webflow | 0 | none | 0.09 | 86 | 2 |
| NetDocuments | Document management | WordPress | 0 | none | 0.07 | 49 | 3 |
| iManage | Document management | Umbraco | 0 | none | 0.87 | 37 | 2 |
Source: Octo Agency, rendered browser read and five curl runs per homepage, 21 October 2026. Clio answered curl with a bot-protection page (HTTP 403), so its speed is n/a; its US page was scored, although our browser was offered the UK site. Harvey's HTML was 77.6 KB.
The scorecard: demo path and totals
| Site | Form | Clicks | Fields | Timing stated | Demo path | Total |
|---|---|---|---|---|---|---|
| Clio | Demo | 1 | 5 | No | 2 | 4 of 6 |
| MyCase | Demo | 1 | 7 | Yes, 30 min | 2 | 6 |
| Smokeball | Demo | 1 | 6 | No | 1 | 6 |
| PracticePanther | Trial, in hero | 0 | 1 | n/a | 2 | 8 |
| Filevine | Demo | 1 | 5 | No | 2 | 4 |
| Litify | Demo | 1 | 6 | Yes, 30 and 60 min | 2 | 5 |
| Lawmatics | Demo | 1 | 7 | No | 1 | 5 |
| Harvey | Sales | 1 | 8 | No | 1 | 5 |
| Legora | Demo | 1 | 7 | No | 1 | 5 |
| Spellbook | Demo pop-up | 1 | 3 | No | 2 | 7 |
| EvenUp | Call | 1 | 6 | No | 1 | 6 |
| Luminance | Demo, on homepage | 1 | 3 | No | 2 | 5 |
| Ironclad | Demo | 2 | 5 | No | 1 | 4 |
| Juro | Demo | 1 | 6 | Yes, 30 min | 2 | 6 |
| SpotDraft | Demo | 1 | 3 | No | 2 | 8 |
| LinkSquares | Demo | 1 | 6 | No | 1 | 5 |
| Everlaw | Demo | 1 | 5 | No | 2 | 5 |
| DISCO | Demo | 1 | 7 | No | 1 | 3 |
| NetDocuments | Demo | 1 | 6 | No | 1 | 4 |
| iManage | Demo | 1 | 10 | No | 1 | 3 |
Source: Octo Agency, rendered browser read of each form, 21 October 2026; no form was submitted. Fields counts the first step only; timing means a stated call or demo length
Clarity: small-firm software says what it is, enterprise software sells a mood
The split follows the buyer. The four practice management vendors, Clio, MyCase, Smokeball and PracticePanther, sell mostly to small and mid-sized firms, and all four passed at least two clarity tests. PracticePanther's headline names legal practice management software, and the line under it lists case management, calendaring, client communication, billing and payments.
The six zero scores all sell mainly to larger firms and corporate legal departments: Litify, Luminance, Everlaw, DISCO, NetDocuments and iManage. NetDocuments' hero reads "Context Lives Here"; iManage's announces that its next-generation platform has arrived. Everlaw's "Do Evidence Justice" sits above "AI built for litigation and investigations". None of the six says eDiscovery, document management or contract management in the hero, although NetDocuments' title tag says "Document and Email Management for Legal Firms and Departments". As in the SaaS sample, these sites know their category; the hero leaves it to the menu and the search snippet.
The category often sits in small type. Smokeball's H1 is "Legal practice management software" set at 14 pixels, above a 200-pixel slogan, "Powering people law"; under the SaaS rubric it would have failed the category test. Lawmatics uses a 12-pixel "Legal software" H1, though its visible headline also names a legal CRM. Ironclad, Spellbook, SpotDraft and Luminance set a category label above the headline; Luminance's "Contract intelligence" is a vendor term, so we did not count it.
AI is the default word, not the problem. Twelve of 20 heroes use it. SpotDraft scores 3 with it, because "AI-native CLM for in-house legal" names the category and the buyer in five words. Filevine's "Legal AI that runs the firm" names a buyer and nothing else.
SpotDraft shows the cost of the second line. Its hero is the clearest in the sample, and it reads "drafts, review, and tracks every contract". A typo in the most-read sentence on the site is cheap to fix and easy to miss.
Legal Tech Websites: How to Explain a Complex Product in Five Seconds covers how to write the headline pair.
Speed: legal tech homepages are light, with one exception
Time to first byte separated almost nobody. Eighteen of 19 measured homepages answered within 0.8 seconds, with a median of 0.10 seconds. iManage was the exception at 0.87 seconds; its five runs ranged from 0.39 to 1.03. web.dev notes that TTFB is not a Core Web Vitals metric in itself, so this sits behind the visitor's experience rather than measuring it.
Document weight was low too. The median homepage sent 59 KB of HTML, against the Web Almanac's 33 KB median and 78 KB 75th percentile for mobile home pages. Fourteen of 19 came in at 78 KB or less.
Filevine is the one heavy page. It sent 232 KB compressed and 930 KB uncompressed. We decoded the document: 76% of it is three animation files for its LOIS product, embedded as base64 text. Loading those files separately, after the first paint, would put Filevine near the sample median.
Platform was not the story. The seven Webflow sites all answered in 0.09 seconds or less, and the WordPress, Framer and Next.js sites answered within 0.35 seconds too. We build marketing sites in Webflow, so weigh that with our bias in mind: a cached page on any modern host answers fast.
Demo path: one click is universal, the form is not
Nineteen of 20 sites put a form one click or less from the first screen. Ironclad was the exception: at 1,024 pixels wide its navigation collapses into a menu and the hero buttons sit below the fold, so a demo takes two clicks. At 1,280 pixels its Request Demo button is visible.
Six fields is the legal tech median. First steps ran from one field to ten. Eight sites asked five or fewer; six asked seven or more, with iManage at ten, including where the visitor found it. Our Legal Tech Demo Request and Lead Form Statistics 2026 post found the same median on a partly overlapping sample of 20 forms three days earlier, and covers phone and firm-size questions in detail.
The best published field-count data is thin and comes from a vendor. Factors.ai, which sells analytics, reports that demo forms with four to six compulsory fields had a 51% success rate against 41% for one to three and 30.60% for seven or more, from what it describes as over 50,000 sessions. Read it as a reason to cut the seventh field, not to add a fourth.
Almost nobody says what happens next. MyCase promises a 30-minute video demonstration; Juro an expert-run, 30-minute demo; Litify a 30-minute introductory call, then a 60-minute demo. Legora says its team will get back "shortly", which does not count. The other 15 forms say nothing about timing. Chili Piper, which sells scheduling software, reports from nearly 4 million form submissions that letting buyers book immediately after the form lifts booking from 30% to 66.7% on average, and names legal and compliance software among its highest-converting categories.
Labels and required fields send odd signals. Everlaw's "Get Started" button opens its demo form, which makes "How did you hear about us?" a required free-text box. PracticePanther puts an email-only trial form in the hero, then shows the same form again in a pop-up on arrival, which covers the headline.
How legal tech compares with B2B SaaS
The SaaS sample was read three days earlier with the same tests. On speed, the SaaS figures were one curl run and these are medians of five.
| Measure | Legal tech, 20 sites | B2B SaaS, 24 sites |
|---|---|---|
| Median total, of 9 | 5 | 5 |
| Hero names a category | 9 of 20 | 14 of 24 |
| Hero names the buyer | 8 of 20 | 7 of 24 |
| Clarity score of zero | 6 of 20 | 3 of 24 |
| Median time to first byte | 0.10 s | 0.34 s |
| HTML over 155 KB | 1 of 19 | 9 of 23 |
| Median first-step fields | 6 | 5 |
| Call length or response time stated | 3 of 19 | 3 of 23 |
Source: Octo Agency scorecards, 18 and 21 October 2026
Legal tech is lighter and faster at the server, asks one more field, and is less likely to say what it sells. It is slightly more likely to name its buyer, probably because many vendors sell to one kind of firm.
What this means for a legal tech website
Run the tests on your homepage and two competitors; each site takes about 30 minutes.
1. Read your hero cold, label included. If the category only lives in a 14-pixel label, move it into the headline or the line under it.
2. Weigh the HTML. Anything over 155 KB usually hides inline animation, data or SVG.
3. Walk the demo path at 1,024 pixels. Count clicks and first-step fields, remove anything sales can ask on the call, and add one sentence on call length if you can keep the promise.
When we build legal tech marketing sites in Webflow, we would run these nine tests on staging before launch, as we would for any B2B site. Our marketing website service explains how we scope that work.
