Table of contents
The person deciding where to send a family member is on a phone at 11 PM. That session is where technical SEO is won or lost in this category - and the average mental health website takes 11.0 seconds to render its main content against a 2.5-second threshold.
Key Takeaways
- Average LCP across 173 mental health sites: 11.0 seconds, versus Google's 2.5-second good threshold.
- 78% of those sites score below 70 on PageSpeed, and 33.9% are in the critical 0 - 49 band.
- Only 62.1% publish any JSON-LD schema; 74.1% declare a canonical tag.
- 77% of sites have three pages sharing the same title, and 29.9% have no sitemap.
- Only about 42% of websites globally pass all three Core Web Vitals.
- Bounce rate climbs 32% at three seconds and roughly 7% of conversions are lost per additional second.
- One audited facility went from 5.8s to 1.9s LCP and gained 31% organic traffic in 90 days.
- 72.4% of addiction treatment site traffic is mobile, with a 64.5% average bounce rate.
- Single-page applications average 47% fewer indexed pages without SSR or dynamic rendering.
- An indexation gap above 20% points to a technical fault, not a content one.
How behavioral health websites actually perform
A 173-site diagnostic crawl of the mental health cluster gives the clearest baseline available.
| Technical signal | Share of mental health sites | Read |
|---|---|---|
| PageSpeed 0 - 49 (critical) | 33.9% | Losing traffic to faster competitors |
| PageSpeed 50 - 69 (needs work) | 44.3% | The largest single band |
| PageSpeed 70 - 89 (decent) | 14.9% | Competitive |
| PageSpeed 90 - 100 (good) | 6.9% | The addressable advantage |
| Main content slow to appear | 82.8% | Highest-impact single fault |
| Three pages sharing one title | 77% | The location-page duplication problem |
| Broken links present | 44.3% | Crawl waste and trust loss |
| No sitemap found | 29.9% | High-impact, cheap fix |
| No robots.txt found | 21.8% | High-impact, cheap fix |
The cluster benchmarks come from 173 full diagnostic crawls of mental health websites. The distribution matters more than the average: with only 6.9% of the competitive set in the good band, page performance is one of the few remaining places a treatment centre can out-execute rather than out-spend. Our rehab SEO benchmarks cover the organic picture those sites are competing in.
Core Web Vitals: thresholds, healthcare context and the 75th percentile trap
| Metric | Good | Needs improvement | Poor |
|---|---|---|---|
| Largest Contentful Paint | Under 2.5s | 2.5s - 4.0s | Above 4.0s |
| Interaction to Next Paint | Under 200ms | 200ms - 500ms | Above 500ms |
| Cumulative Layout Shift | Under 0.1 | 0.1 - 0.25 | Above 0.25 |
| Measurement point | 75th percentile of loads | Mobile first | Field data, not lab |
| Realistic treatment-site LCP target | 1.8s - 2.2s | 2.5s is the minimum | Above 4s loses sessions |
| Realistic treatment-site INP target | 100ms - 160ms | 200ms is the ceiling | Above 500ms is broken |
| Global sites passing all three | ~42% | - | - |
Thresholds and behavioral health targets are documented in Core Web Vitals benchmarks for treatment centre websites, with the global pass rate and speed research compiled in 2026 technical SEO statistics and Core Web Vitals statistics. A site averaging 2.1 seconds can still fail if the slowest quarter of loads exceeds 2.5 - and that quarter is disproportionately the older phone on a weak connection at midnight, which is the exact session this category depends on.
Healthcare comparison: who is slow, and how slow
| Site category | Average LCP or pass rate | Sample |
|---|---|---|
| Mental health websites | 11.0s average LCP | 173 sites |
| Hospitals and urgent care | 13.4s average LCP | Diagnostic crawl sample |
| General medical | 9.7s average LCP | Diagnostic crawl sample |
| Local medical practices | 66.5% pass CWV on mobile | 158 practices, 32 US metros |
| All websites globally | ~42% pass all three CWV | Chrome UX Report data |
| Google top-10 results | 1.65s median load time | Page speed study |
| Mental health sites using SSL | 95.4% | 173 sites |
The local-practice pass rate is reported in a healthcare website speed study of 158 practices across 32 metros, and the category averages in the Borah Labs crawls above. Small, focused healthcare sites beat large institutional ones comfortably - which is good news for a single-facility operator and a warning for a multi-location group running a heavy template.
What makes treatment centre sites slow
- Auto-playing cinematic hero video is the most common single cause of LCP failure on treatment centre homepages - frequently 5 to 15 MB, blocking render and adding 3 to 5 seconds.
- Chat widgets, tag managers, A/B tools, heatmaps and pixels each add 100 to 500 milliseconds.
- Generic shared hosting produces 1 to 2 seconds of TTFB before rendering starts; appropriately tiered HIPAA-eligible hosting typically lands under 400 milliseconds.
- Sites built before 2022 fail disproportionately - page-builder and jQuery-era patterns struggle against post-2024 thresholds, particularly INP.
- JavaScript weight now averages 507 KB compressed per page, with third-party scripts accounting for 54% of it.
- Client-side rendering indexes roughly 32% less content, and single-page applications average 47% fewer indexed pages without SSR or dynamic rendering.
- Moving scripts to async or defer improves LCP by about 28% on average.
- The average web page is now 2.3 MB, up 18% since 2023 - the baseline is getting heavier, not lighter.
Causes and remediation patterns are set out in technical SEO guidance for behavioral health websites, with rendering and JavaScript data in the 2026 technical SEO statistics compilation. Platform choice compounds all of this - see our rehab WordPress benchmarks.
Crawl and indexation faults that recur across audits
| Issue | Site frequency | URL-level frequency |
|---|---|---|
| Core Web Vitals failure on high-traffic pages | 91% | 34% of crawled URLs |
| Pages exceeding the 2 MB HTML crawl limit | 78% | 4% of URLs |
| Schema validation errors on eligible pages | 69% | 22% of marked-up URLs |
| Soft 404s on discontinued content | 64% | 8% of URLs |
| Redirect chains three or more hops deep | 58% | - |
| Orphan pages with traffic potential | 52% | - |
| Inconsistent canonicalisation | 47% | - |
| Long-tail CWV pass rate | 51% | vs 84% on homepages |
Aggregate issue frequencies come from 2026 aggregate technical SEO issue data, with crawl-waste analysis in the 2026 crawl waste report. The 84% homepage versus 51% long-tail pass rate is the pattern that hurts treatment centres most: the homepage gets optimised and the city and programme pages - the ones that rank for commercial intent - do not.
The duplicate location-page problem
- Location pages that copy programme descriptions across cities trigger duplicate-content filters - the structural fault behind the 77% same-title finding.
- Level-of-care pages duplicated with minor swaps compete with each other rather than consolidating authority.
- Category and tag archives generate near-duplicate thin pages that consume crawl budget.
- An indexation gap above 20% between site pages and indexed pages points to technical causes.
- Audit remediation options are rewrite, merge, correct canonicals, deindex or delete - chosen per duplicate cluster, not site-wide.
- Excluded pages, crawl errors and CWV failures are the three diagnostics that separate a technical problem from a content problem.
- Sites shipping a full remediation list typically see 10% to 25% organic improvement in 60 to 90 days, concentrated on long-tail URLs.
Duplicate-cluster methodology is described in rehab technical SEO audit guidance, with the diagnostic sequence in a treatment-site audit framework and indexation-gap thresholds in a 2026 rehab ranking diagnostic. Local visibility depends on it too - see our rehab local SEO benchmarks.
What speed costs in conversions, not rankings
| Speed or behaviour metric | 2026 figure | Why it matters here |
|---|---|---|
| Bounce increase at 3 seconds | +32% | Versus a 1-second baseline |
| Bounce increase at 5 seconds | +90% | Most treatment sites live here |
| Conversion loss per extra second | ~7% | Compounds across the funnel |
| Mobile share of addiction treatment traffic | 72.4% | Mobile is the site, not a version |
| Average bounce rate, addiction treatment | 64.5% | Above healthcare average |
| New visitors | 78% | First impression is the only impression |
| Documented audit: mobile sessions bouncing under 3s | ~47% | Before remediation |
| Documented audit: organic lift after fix | +31% in 90 days | 5.8s to 1.9s LCP |
Vertical traffic and bounce benchmarks come from 2026 addiction treatment marketing benchmarks. With 78% new visitors and a 64.5% bounce rate, there is no second session in which to recover a slow first one.
Structured data and on-page hygiene: the cheap advantage
- Only 62.1% of mental health sites publish any JSON-LD - schema is still a differentiator in this vertical.
- 73% have a populated meta description and 74.1% declare a canonical; both should be 100%.
- 73% have Open Graph tags, which governs how a link previews when a family member shares it.
- 90.2% declare a mobile viewport - meaning roughly one in ten still renders at desktop zoom on a phone.
- 95.4% use SSL, so the remaining sites are showing browser warnings to people in crisis.
- 95.6% of Google top-10 pages use HTTPS, rising to 97.2% at position one.
- Schema validation errors affect 69% of sites that publish structured data - implementing it is not the same as implementing it correctly.
- Mobile-friendliness failures are now below 5% site frequency; the bar has moved to speed and structure.
Structured-data adoption patterns are analysed in a 5,000-site schema markup audit, alongside the mental health cluster figures above.
A 90-day technical remediation order for a treatment centre site
| Window | Work | Target |
|---|---|---|
| Days 1 - 10 | Field-data audit: CWV at 75th percentile, indexation gap, crawl errors | Baseline documented |
| Days 10 - 25 | Kill hero video autoplay, compress and preload the LCP element | LCP under 2.5s |
| Days 20 - 35 | Audit third-party scripts; defer or remove; fix INP handlers | INP under 200ms |
| Days 30 - 50 | Resolve duplicate location and level-of-care clusters | Same-title pages eliminated |
| Days 45 - 65 | Sitemap, robots.txt, canonicals, redirect chains, broken links | Crawl waste removed |
| Days 60 - 80 | MedicalOrganization, FAQ and breadcrumb schema, validated | Zero validation errors |
| Days 80 - 90 | Re-measure field data; verify long-tail pass rate, not just homepage | 10% - 25% organic lift |
Sequence matters: fixing schema on a site with an 11-second LCP changes nothing. Technical delivery sits in data intelligence, and if you want the audit run against your own field data, get in touch.
Frequently Asked Questions
How fast do rehab and behavioral health websites actually load?
Badly, on average. A diagnostic crawl of 173 mental health websites found an average Largest Contentful Paint of 11.0 seconds against Google's 2.5-second 'good' threshold, an average PageSpeed score of 57 out of 100, and 78% of sites scoring below 70 - with 33.9% in the critical 0 to 49 band. For context, a sample of 158 local medical practices passed Core Web Vitals on mobile 66.5% of the time, and hospital and urgent-care sites averaged a 13.4 second LCP. Behavioral health sits toward the slow end of healthcare, which is itself the slow end of the web.
What are the 2026 Core Web Vitals thresholds?
LCP under 2.5 seconds, INP under 200 milliseconds and CLS under 0.1, all measured at the 75th percentile of real page loads on mobile. The 75th percentile matters more than most site owners realise: a site averaging 2.1 seconds still fails if a quarter of loads exceed 2.5 seconds, and that slowest quartile is typically older devices on weaker connections - which in this category is often the family member searching late at night. Only about 42% of sites globally pass all three metrics.
Why do treatment centre location pages cause technical SEO problems?
Because they are usually near-duplicates. Multi-facility operators need a page per city to capture local intent, but copying the same programme descriptions across cities, duplicating level-of-care pages with minor swaps, and generating tag and category archives produces thin, competing URLs. Audits routinely find three pages sharing the same title on 77% of sites in the mental health sample. The diagnostic is an indexation gap: compare indexed URLs with total pages, and a gap above roughly 20% points to a technical cause rather than a content one.
What does fixing Core Web Vitals actually deliver for a treatment centre?
One documented audit took a facility from a 5.8-second LCP and 340-millisecond INP - with roughly 47% of mobile sessions bouncing inside three seconds - to a 1.9-second LCP, 140-millisecond INP and 0.04 CLS. Organic traffic to the affected cluster rose 31% over the following 90 days, with admit-attributed conversion rising alongside it. Aggregate crawl data supports the pattern: sites shipping a full technical remediation list typically see 10% to 25% organic improvement over 60 to 90 days, concentrated on long-tail URLs.
Is schema markup worth the effort in behavioral health?
Yes, and adoption is still weak enough to be an advantage. Only 62.1% of mental health sites in the 173-site sample published any JSON-LD, 74.1% declared a canonical tag and 73% had a populated meta description. Meanwhile schema validation errors appear on 69% of sites that do use structured data, affecting 22% of their marked-up URLs. For a treatment centre the useful set is MedicalOrganization or LocalBusiness with accreditation and licence data, plus FAQ and breadcrumb markup on programme pages - implemented once, validated, and monitored.
Sources
Mental health website benchmarks (173 sites)
Core Web Vitals benchmarks for treatment centres
Technical SEO for behavioral health websites
Rehab website ranking diagnostic 2026
Healthcare website speed study, 158 practices
2026 aggregate technical SEO issues
2026 crawl waste report
Technical SEO statistics 2026
Core Web Vitals statistics 2026
Schema markup adoption: 5,000-site audit
Rehab technical SEO audit guidance
Rehab centre SEO audit framework
