Challenge
Mobile slow enough to hurt rankings and conversion, on a platform where the advice is "accept it."
Solution
Worked the render path, not the plugin list.
Result
Lighthouse 61→88; blocking time 3,370ms→40ms; main-thread 11.5s→2.2s.
Challenge
A regional home improvement client had a mobile site slow enough to affect both rankings and conversion, on a platform where the usual advice is to accept it.
Solution
We worked the render path rather than the plugin list: what runs before the page becomes usable, what blocks the main thread, and what can be deferred or removed without changing what the visitor sees. On a separate client we removed a 9.3 MB background video from 31 of 37 pages and took those pages from 12.4 MB to between 2.9 and 5.3 MB, which is the same principle applied to the asset weight instead of the script weight.
Result
Mobile Lighthouse performance went from 61 to 88. Total blocking time went from 3,370ms to 40ms, and main thread work from 11.5 seconds to 2.2. We also told the client what the score does not mean: lab numbers move the day you ship, field data is a rolling 28-day window, and the metric Google actually uses lags the fix by weeks. Setting that expectation up front is part of the work.
Work like this is delivered through our Technical Services and, where it is ongoing, our Platform Management bundles.
< Related Engagements >
More from this category
Reporting Configuration Under Privacy Constraints
Result: Reporting fixed, privacy intact, years of follow-on support.
Read the full overview →AEO / AIAnswer Engine Visibility, Measured Honestly
Result: A report of actual AI-to-sales conversion, including a $45,000 order, now in analytics reporting rather than anecdote.
Read the full overview →← All case studies · Next: Migrating a Live Site Off a Proprietary Platform →
Have a problem shaped like this one?
Bring us the messy version. The free audit tells you what is actually broken and what it takes to fix it.
Bring Us Your Challenge