Bracket A · Standard
Single site
Up to 30 page templates · one locale · one brand.
$3,500
- Pay
- 50 / 50 · Net-7
- Timeline
- 14 days
- Guarantee
- Miss metric · skip back-half
§ 03.02 — Sprint 02 of 06 · open now
Sub-1s LCP on mobile. Two weeks. Public proof on delivery.
A senior WordPress engineer rebuilds the slow parts of your site — plugin teardown, CSS / JS triage, image pipeline, caching, edge — and ships you a public PageSpeed Insights URL on the last day. Fixed from $3,500. If we don't hit the agreed metric, you don't pay the back half.
- Service
- Performance Rescue
- Starts at
- $3,500
- Timeline
- 14 days · fixed
- Ideal for
- Live WP site · LCP > 2.5s mobile
- Deliverable
- Live build · public PSI URL · Loom
- Status
- ● Open · next start 09 Jun
§ 03.02.A
In 50 words
The Performance Rescue is a 14-day, fixed-price WordPress build that targets a sub-1s mobile LCP. A senior engineer strips plugin sprawl, rebuilds the critical render path, and ships a public PageSpeed Insights URL on day 14. From $3,500. Outcome-tied — miss the metric, skip the back-half invoice.
§ 03.02.B
Why teams book this
Mobile LCP is 4.2 s on your homepage. Lighthouse mocks you. Marketing is asking why bounce rate is climbing. Your developer says "buy faster hosting" — you bought faster hosting and it moved 0.3 s. The page-builder vendor says "enable our optimizer" — you enabled it and now your CLS broke. Every option pitches a quick fix that doesn't actually move the metric.
Performance work has one honest path. Measure on a real device profile at the 75th percentile. Cut the JavaScript you don't need. Inline the CSS you do. Stop rendering anything offscreen until it's near the viewport. Serve images in the right format, at the right size, with the right priority. Cache aggressively, evict precisely. That's the work. Two weeks, ranked, dated, public.
Recent client outcomes
- — Cyberhive · LCP 4.1s → 0.7s
- — Penrose Wealth · LH 62 → 96
- — Argentum · CLS 0.21 → 0.02
- — Scott Fleary · INP 287ms → 92ms
Live numbers · /benchmark · refreshed monthly.
§ 03.02.C
What you get · Day by day
D1Kickoff · 90 minRepo + hosting + analytics access. We pick three target pages and agree the metric we will move (LCP / INP / CLS / Lighthouse).
D2BaselineReal-device profile at p75 mobile + p75 desktop. CWV trace. Waterfall. PSI. Lab + field. Numbers published privately to you.
D3 — D4Plugin teardownEvery active plugin weighed — by JS, by CSS, by render impact, by abandonment risk. Removed, replaced, or contained.
D5 — D6CSS + JS triageCritical CSS inlined. Non-critical CSS deferred. Render-blocking JS lazy-loaded. Third-party scripts gated to user intent.
D7 — D8Image pipelineLCP image priority-hinted. Below-fold images lazy + sized + WebP/AVIF. Hero pre-loaded. No layout-shift on load.
D9 — D10Caching + edgePage cache, object cache, fragment cache. Edge config tuned. Eviction policy documented. CDN configured per asset class.
D11VerificationSame real-device profile. Same target pages. Same lab + field tools. We diff against D2 numbers.
D12Edge casesSlow-connection profile · Lighthouse trace on form / checkout / search. Mobile-only regressions reviewed.
D13DocumentationRunbook: what we changed, why, how to roll back. Loom walkthrough · 30 min · section-by-section.
D14Handover · publicPublic PageSpeed Insights URL signed off. Listed on /benchmark the same week unless your SOW asks otherwise.
§ 03.02.D
Sample outcome
Sample · PSI · mobile · p75
FIG. 02 — Verified outcome
Public PSI · Dec 2025
What moves the metric
- ▸ Plugin diet · 27 removed
- ▸ Critical CSS inlined · 8KB
- ▸ Render-blocking JS · 0 KB
- ▸ LCP image priority hint
- ▸ Hero pre-load · skip waterfall
- ▸ WebP / AVIF fallback ladder
- ▸ Edge cache · stale-while-revalidate
- ▸ Object cache · Redis · documented
§ 03.02.E
What's out of scope
- × Visual redesign (markup may change · pixels don't)
- × New features or page templates
- × Hosting migration (separate SOW)
- × Copywriting / messaging
- × Off-WordPress sites (Webflow, Shopify, etc.)
- × Backend-only sites with no public LCP
Need a rebuild instead of a rescue? See the Launch Sprint. Need just accessibility? See the Accessibility Sprint. Unsure where to spend? Start with a 7-day Audit Sprint — its fee credits 1:1 against this build.
§ 03.02.F
Pricing · Three brackets
Bracket B · Multi-locale
Localized
2 — 4 locales · WPML or Polylang. Adds 3 days.
$4,800
- Pay
- 50 / 50 · Net-7
- Timeline
- 17 days
- Guarantee
- Miss metric · skip back-half
Bracket C · Enterprise
Multi-site
WP Multisite or 5+ properties. Bespoke quote.
From $7,200
- Pay
- 40 / 40 / 20
- Timeline
- 21 — 28 days
- Guarantee
- Per-site metric in SOW
Outcome-tied pricing: the metric, the page, and the percentile are written into the SOW on D1. If the verification run on D11 misses the agreed target, the back-half invoice is voided — you keep what we built and pay only the deposit. The discipline of writing the target down in advance is part of why this works.
§ 03.02.G
FAQ · Five
Q1
What if my hosting is the bottleneck — can you still hit sub-1s?
Usually yes. Of 12 engagements, hosting was the binding constraint on one — and we still moved LCP from 6.2 s to 1.3 s by attacking everything else. If on D2 your TTFB is past 800 ms, we name it in writing and either accept a less aggressive metric or pause for a hosting move first. No surprises on D11.
Q2
What happens to my existing plugins?
Removed if unused or replaceable by 10 lines of code. Replaced if the role is needed but the implementation is bloated. Contained (lazy-loaded, scope-restricted) if the role is irreducible — payment processors, analytics under explicit consent, etc. Every removal is documented; rollback is one command.
Q3
Will this only help mobile?
Mobile is the harder target, so we optimize for mobile p75 and desktop comes free. Lighthouse desktop typically lands 98+ after a rescue that hits sub-1s mobile LCP. The SOW names both, but if it lists only one, it's mobile.
Q4
Do I need an Audit Sprint first?
No. If you already have a clear performance complaint — slow LCP, embarrassing PSI score, marketing pressure — book the Rescue directly. The Audit is for buyers who want to know the priority across performance, accessibility, plugins, security, workflow before they spend. Both work; pick by what you already know.
Q5
Will my site look different afterwards?
No. Performance work is markup-and-asset-level. Pixels stay where they are unless you specifically ask for a change. If a layout shift was the cause of a CLS regression, the visible fix is removing the shift — not a redesign.