SGA Dental — Performance Audit Evidence

periomem.com — Mobile Performance Review

Reference data for a collaborative scope discussion
Practice: Periodontal Associates of Memphis Audited: 2026-05-19 Tool: Google PageSpeed Insights API (Lighthouse, mobile)
Scope context. The current website was built by Infogenix under a January 2018 SOW covering custom design, content migration, CTAs, video production, CMS, and an optional SEO retainer (meta tags, on-page content, keyword pages, link-building, ranking reports). Performance optimization, Core Web Vitals, image format conversion, CDN, and modern caching configuration were not in that original SOW — and Google did not make Core Web Vitals a ranking signal until 2021, three years after the build. This sheet is reference data for a scope conversation, not a delivery complaint.
PSI ran successfully and returned Performance 34/100 on mobile. The vendor reply correctly notes the test produces no errors — that's right, the tool works fine. The 34 is the score, not a failure of the tool. Below: the live metrics, three points worth clarifying together, the technical evidence, and a punch list of fixes — with owner suggestions for each item.

1. Current mobile performance baseline

Metric (mobile)MeasuredGoogle "Good" thresholdStatus
Performance score34/10090+Poor
Largest Contentful Paint (LCP)11.0 s< 2.5 sPoor
Speed Index7.7 s< 3.4 sPoor
Time to Interactive12.5 s< 3.8 sPoor
Total Blocking Time200 ms< 200 msBorderline
Cumulative Layout Shift1.314< 0.1Severe
Accessibility67/10090+Needs work
Best Practices96/10090+Good
SEO92/10090+Good

2. Three points to align on

To clarify

“We have tested PSI multiple times and have not encountered any failures or errors.”

Agreed — PSI itself runs without errors. The 34/100 is the result of a successful run, not a tool failure. We're aligned on the data; the question is what to do about it.

Partly true

“The current hosting provider is InMotionHosting. We do not have access to the account.”

Understood — the 2018 SOW offered Infogenix hosting, so it's plausible the practice migrated to InMotion direct at some later point. Measured TTFB is 1.06 s, consistent with shared WordPress hosting. Host explains TTFB; it does not explain LCP 11 s or CLS 1.3 — those are front-end issues that can be fixed without host access.

Worth checking

“The SEO team has not installed performance optimization plugins.”

W3 Total Cache is currently loaded on every page — likely installed at the original 2018 build, never reconfigured. It's not emitting cache headers, so it appears inactive. Worth a quick check from your team to confirm whether to configure it or remove it cleanly.

3. Live evidence (reproducible from any terminal)

Plugins currently loading on the homepage — output of curl https://www.periomem.com/ | grep wp-content/plugins:

/wp-content/plugins/cleantalk-spam-protect/
/wp-content/plugins/nextend-smart-slider3-pro/
/wp-content/plugins/supreme-modules-for-divi/
/wp-content/plugins/w3-total-cache/

Server / caching headers — output of curl -I https://www.periomem.com/:

HTTP/2 200
server: nginx
cache-control: max-age=2452, public
content-encoding: gzip
(no x-cache, no cf-cache-status, no x-w3tc-*, no CDN headers — W3TC appears installed but inactive)

Time-to-first-byte sample:

DNS=0.003s  Connect=0.076s  SSL=0.144s  TTFB=1.056s  Total=1.381s

Page weight & image strategy:

Raw HTML size:           614,508 bytes (uncompressed)
Image tags on homepage:  42
Images with loading=lazy: 6 of 42  (36 images load eagerly)
Modern formats (WebP):   0  (all JPG/PNG — e.g. AdobeStock_487012808-min-scaled-resized.jpg)
Hero:                    Nextend Smart Slider 3 Pro (render-blocking, 7-slide carousel)
Theme:                   Divi (heavy framework; built-in perf toggles available)

4. Proposed punch list — who owns what

1
Audit the existing W3 Total Cache configuration
Confirm whether to activate it (page cache + browser cache + minify + WebP converter) or remove cleanly if no longer needed.
Suggested owner: Infogenix (existing install, theme familiarity)
Low effort
2
Replace the Nextend hero slider with a static hero
Sliders are the #1 LCP killer on dental sites. Single optimized hero image should cut LCP by 6–8 seconds.
Suggested owner: Infogenix (Divi/Nextend work)
Low effort
3
Enable Divi's global lazy-load + performance toggles
Divi → Theme Options → Performance: Critical CSS, Defer JS, Dynamic CSS, lazy-load. Native to the existing stack.
Suggested owner: Infogenix (Divi config)
Trivial
4
Convert images to WebP
~25–35% image weight reduction. W3TC ships with a converter; Divi supports WebP output.
Suggested owner: Infogenix (paired with #1)
Low effort
5
Put Cloudflare's free tier in front of the site
Fixes TTFB and adds a CDN without touching InMotion. Single DNS change at the registrar. Free.
Suggested owner: SGA (can be done without vendor or host access)
Med effort
6
Remediate Accessibility (currently 67/100)
Includes the existing viewport tag that disables pinch-zoom (a Lighthouse violation). Improves both accessibility and SEO signal.
Suggested owner: Infogenix (theme template edits)
Med effort

5. How to verify any of this yourself

  1. PSI score: https://pagespeed.web.dev/report?url=https%3A%2F%2Fwww.periomem.com%2F&form_factor=mobile
  2. W3 Total Cache present: open view-source:https://www.periomem.com/ and search for w3-total-cache
  3. TTFB: run curl -o /dev/null -w "TTFB=%{time_starttransfer}s\n" https://www.periomem.com/
  4. Lazy-load count: curl -sL https://www.periomem.com/ | grep -c 'loading="lazy"'