Only 34.5% of busy sites running Sitecore and 33.2% of those running Adobe Experience Manager passed all three Core Web Vitals on mobile in September 2026, against 46.6% of all sites in Chrome's top million (HTTP Archive). Those two platforms ran 16 of the 27 large law, accounting, consulting, engineering and recruitment firm home pages we could fetch on 2026-10-21. The biggest firms have built on the two platforms with the lowest pass rates in this data.
No public dataset labels a website as professional services, so this page combines three things: Chrome's real-user data for the platforms and tools firm sites use, read through the HTTP Archive technology report; our own check of what 27 large firm sites run on; and Google's thresholds, the Web Almanac and the few studies that tie speed to enquiries.
The thresholds, and why most firm sites have no score
Google judges a page on three field measures from real Chrome users. A page is good when 75% of visits reach a Largest Contentful Paint (LCP) of 2.5 seconds or less, an Interaction to Next Paint (INP) of 200 milliseconds or less and a Cumulative Layout Shift (CLS) of 0.1 or less, judged separately on mobile and desktop (web.dev). Poor starts above 4 seconds, 500 milliseconds and 0.25. Search Console gives a group of URLs the status of its worst metric over 28 days. The full threshold table is in our B2B SaaS Website Speed and Core Web Vitals Statistics 2026 post.
The catch for a firm is traffic. The Chrome UX Report includes only pages and origins that are "sufficiently popular", and Google does not state the minimum. It also leaves out Chrome on iOS and other Chromium browsers such as Edge, so a buyer reading on an iPhone or a work laptop running Edge is not counted. When a group of pages lacks data, Search Console falls back to the whole origin; when the origin lacks data, the site does not appear at all.
That matters because firm sites are small. HubSpot's web trends report, citing AgencyAnalytics' 2025 benchmarks for mid-market and small business sites, puts professional services at 1,000 to 3,000 sessions a month. At that size a practice page or a partner's bio is unlikely to have its own field score, so a firm usually sees the origin average.
Mobile is where the visits are, desktop is where they convert. Unbounce found 81% of commercial and professional services landing page visits were on mobile, while desktop converted at 11.6% against 8.3% on mobile. Several platforms and tools below pass on one device and fail on the other.
What large firms run their websites on
We fetched the home page of 32 large firms on 2026-10-21 and read the server HTML for platform signatures. Five returned a bot wall: McKinsey, BCG, Bain, PwC and DLA Piper. The raw file is saved as research/OB-0213-firm-stacks-2026-10-21.tsv.
| Sector | Sites read | Sitecore | Adobe Experience Manager | Other platforms detected |
|---|---|---|---|---|
| Law | 8 | 6 | 1 | 1 not identified |
| Accounting | 7 | 0 | 5 | Kentico; Sanity with Next.js |
| Consulting | 4 | 1 | 1 | Craft CMS (inferred from a plugin tag); Contentstack with Next.js |
| Engineering and architecture | 5 | 1 | 0 | Optimizely; WordPress; Drupal; 1 not identified |
| Recruitment | 3 | 0 | 1 | Liferay; Webflow |
| Total | 27 | 8 | 8 | 11 |
Source: Octo Agency check of 32 firm home pages with curl, 2026-10-21; detection from HTML signatures such as /-/media/ and /etc.clientlibs/ paths
Six of the eight law firms run Sitecore, and five of the seven accounting networks run Adobe Experience Manager. Three Sitecore sites, Irwin Mitchell, Linklaters and Huron, render through Next.js on top of it. Google Tag Manager appeared in the server HTML of 19 of the 27 pages and OneTrust's consent banner in 13. The home pages carried a median of 30 script tags, from 7 on Arup's to 74 on Azets'. This count misses anything a tag manager injects later, so it is a floor.
Pass rates by platform
The HTTP Archive detects technology on each origin with Wappalyzer rules and joins it to Chrome's field data. The table shows September 2026, the latest month published when we read the data on 2026-10-21.
| Platform | Mobile origins | All three, mobile | All three, desktop | Good TTFB, mobile | All three, mobile, top 1M |
|---|---|---|---|---|---|
| All origins | 8,569,446 | 52.8% | 60.8% | 45.9% | 46.6% |
| Duda | 42,026 | 85.8% | 87.8% | 73.6% | 64.2% |
| Wix | 177,709 | 80.9% | 87.1% | 74.3% | 52.5% |
| Umbraco | 4,332 | 76.4% | 73.0% | 74.1% | 70.9% |
| Webflow | 59,483 | 71.3% | 82.6% | 73.1% | 49.8% |
| Optimizely CMS | 1,820 | 71.1% | 64.1% | 63.1% | 62.6% |
| Squarespace | 89,052 | 71.0% | 76.1% | 76.6% | 56.3% |
| Drupal | 93,193 | 64.7% | 66.1% | 45.6% | 58.6% |
| HubSpot CMS | 13,904 | 60.0% | 75.1% | 67.3% | 45.2% |
| WordPress | 2,658,175 | 48.3% | 53.7% | 22.9% | 50.1% |
| Sitecore | 5,890 | 39.3% | 48.7% | 27.2% | 34.5% |
| Adobe Experience Manager | 19,174 | 33.9% | 44.3% | 44.1% | 33.2% |
Source: HTTP Archive Core Web Vitals technology report, all geographies, read through its public API on 2026-10-21. TTFB is time to first byte; top 1M is the rank filter for the million most visited origins.
Enterprise platforms trail at every size. Sitecore and Adobe Experience Manager sit 12 and 13 points below the top-million average on mobile and more than 10 points below it on desktop. Optimizely CMS and Umbraco, which also serve large organisations, pass well above it. These averages describe how sites on each platform are typically assembled, with personalisation, search and analytics on top, not a limit of the software.
Hosted builders answer fastest. Duda, Wix, Squarespace and Webflow serve pages from managed hosting, and 73% to 77% of their mobile origins have a good time to first byte. Only 22.9% of WordPress origins do. The WordPress page builders common on small firm sites do worse: Elementor origins passed at 36.3% on mobile, WPBakery at 38.7% and Divi at 41.4%, with good server response on 15% to 18% of them. We build marketing sites in Webflow, so we are not neutral on that row; note that its lead over the average shrinks from 18.5 points across all origins to 3.2 points in the top million.
Watch desktop as well as mobile. Scorpion, a marketing platform sold to law, home services and franchise businesses, passed on 83.0% of its 669 mobile origins but on only 23.3% of desktop ones, because just 23.3% had a good desktop layout shift score. Divi shows the same pattern, 41.4% on mobile and 31.2% on desktop.
The trend is upward almost everywhere. Between January 2024 and September 2026, Sitecore's mobile pass rate rose from 31.5% to 39.3%, WordPress from 39.0% to 48.3%, Squarespace from 54.0% to 71.0% and Wix from 58.7% to 80.9%, while all origins moved from 46.8% to 52.8% (HTTP Archive, same report).
The tools firm sites add
Firm sites share a recognisable kit: call tracking, because 52.6% of professional services conversions in Ruler Analytics' 2026 data came by phone, a chat or receptionist widget, a review badge, a booking link, a consent banner and sometimes an accessibility overlay. The HTTP Archive tracks each one.
| Tool | Mobile origins | All three, mobile | All three, desktop | Good INP, mobile | Median mobile JS (KB) |
|---|---|---|---|---|---|
| All origins | 8,569,446 | 52.8% | 60.8% | 79.4% | 714 |
| Google Tag Manager | 2,547,920 | 46.5% | 56.0% | 77.7% | 1,174 |
| CallRail | 20,251 | 45.0% | 53.7% | 77.6% | 1,586 |
| CallTrackingMetrics | 5,709 | 49.3% | 57.8% | 84.7% | 1,523 |
| Calendly | 32,730 | 54.9% | 61.8% | 87.9% | 935 |
| LiveChat | 24,147 | 27.8% | 43.7% | 54.5% | 1,587 |
| ApexChat | 911 | 37.8% | 58.8% | 93.9% | 1,630 |
| Podium | 7,124 | 35.7% | 40.7% | 64.1% | 2,664 |
| Birdeye | 1,494 | 37.1% | 50.5% | 89.3% | 1,644 |
| Trustpilot | 57,962 | 56.5% | 62.0% | 82.9% | 1,704 |
| OneTrust | 107,248 | 42.5% | 44.3% | 75.6% | 1,560 |
| Cookiebot | 135,894 | 58.0% | 63.2% | 88.8% | 1,139 |
| UserWay | 57,263 | 57.5% | 57.1% | 90.1% | 1,114 |
| AccessiBe | 35,055 | 50.3% | 55.8% | 85.1% | 1,779 |
| AudioEye | 28,293 | 33.7% | 23.7% | 60.1% | 2,229 |
Source: HTTP Archive Core Web Vitals and page-weight technology reports, all origins, read 2026-10-21. KB is 1,024 bytes; JavaScript is the median per page across origins with the tool.
These are correlations, not tests: a site with Podium or LiveChat usually carries other scripts too. Chat and messaging widgets sit among the lowest, 27.8% to 37.8% on mobile, and their pages carry up to 3.7 times the median page's JavaScript. ApexChat keeps a good INP score but fails on layout: only 57.5% of its mobile origins have good CLS.
Not every tool drags. Calendly, Cookiebot, Trustpilot and UserWay origins pass at or above the average. The way a tool is loaded matters more than its name.
Accessibility overlays are the sharpest case. AudioEye sites passed on 23.7% of desktop origins, with good layout shift on only 28.7%. Overlays do not buy compliance either: in April 2025 the US Federal Trade Commission made accessiBe pay $1 million over claims that its widget could make any website WCAG compliant. We would fix the template's accessibility in code and drop the overlay.
Chat and booking widgets can wait for a click. Google's Lighthouse guidance names live chat among the embeds that suit a facade, a static button that loads the real script only when a visitor uses it; its example swaps a 540 KB YouTube player for a 3 KB placeholder. Web Almanac 2025 puts the median mobile page's total blocking time at 1,916 milliseconds in lab tests, up 58% in a year, so script deferred is responsiveness gained.
What speed does to enquiries
No study ties speed to professional services enquiries. The closest measured lead-generation forms; our Website Speed Statistics 2026 post covers the ecommerce tests.
Deloitte and 55 for Google. The study followed 37 European and American brands and more than 30 million sessions, monitoring mobile load times for 30 days at the end of 2019. On lead-generation sites, a 0.1 second improvement across four speed measures came with a 21.6% rise in visitors reaching the form submission page and a 5.5% rise from the home page to the first form step. It is correlational, and one of its four measures, First Meaningful Paint, is now deprecated.
The referral check. Hinge's 2014 survey of 1,028 buyers of legal, accounting, technology and consulting services found over 99% check a firm out before hiring and over 80% visit its website to decide. Hinge sells marketing to the firms it surveys, and the data is old. A referred prospect who taps a slow site on a phone is still judging the firm, not the platform.
Rankings. Google's page experience documentation, updated in September 2026, says Core Web Vitals "are used by our ranking systems", that there is no single page experience signal, and that Search shows the most relevant content even when page experience is poor. Speed breaks ties; it does not replace a better practice page.
What this means for a professional services site: a five-page test order
Field data tells you which pages fail; a lab test tells you why. We would test in this order, because it follows the path a referred buyer takes and where a slow page costs an enquiry. It complements the conversion test order in our Professional Services Website Conversion Rate Statistics 2026 post.
| Order | Page | Why it goes here | What to check first |
|---|---|---|---|
| 1 | Contact or consultation page | Every enquiry passes through it, with the form, call tracking and chat loaded | INP on mobile with every script accepted |
| 2 | Top practice or service page | Organic and paid searches land here | LCP on mobile, hero image priority |
| 3 | People and bio pages | Referred buyers check who they would work with | Image weight, layout shift from photos |
| 4 | Home page | Web Almanac finds 45% of mobile home pages pass against 56% of secondary pages | Hero media, overlay and chat scripts |
| 5 | Most read insight article | Can carry much of a small site's traffic, and so the origin score | Embeds, consent banner shift |
Source: Octo Agency recommendation, based on the HTTP Archive and Web Almanac findings above
For each page, read the 28-day field data first, then run a lab test on a throttled mobile profile twice: once as a new visitor with the consent banner, once with every script accepted, because call tracking and chat often load only after consent. Web Almanac 2025 also found 62% of mobile pages leave at least one image without dimensions, a common cause of layout shift on photo-heavy pages such as bios.
On Sitecore or Adobe Experience Manager, we would set a script budget before considering a replatform. On a smaller firm's site, the builder buys a head start that a chat widget and an overlay can spend in a week. On our four-week marketing-site process we would write every third-party script into the brief with an owner and a reason. Our marketing websites service page explains how we scope a firm's site.
