Compression and Format Conversion Gate (QA)

Compression and Format Conversion Gate (QA)

1. 01

   Pull the client's current Core Web Vitals baseline before touching a single asset

   Record LCP and total page weight from PageSpeed Insights or the client's analytics so the before/after delta can be shown on the next retainer review.

2. 02

   Cap the compression pass at the point where artifacts appear in a 100 percent zoom check

   Automated pipelines such as Imagify's Smart Compression balance quality against file size, but a human still has to eyeball hero images and product shots at full resolution before approving the output.

3. 03

   Convert to WebP or AVIF only where the delivery stack supports it end to end

   Confirm the CMS, CDN, and email platform can serve the modern format with a fallback; a converted asset that renders as a broken image on one client channel costs more to fix than the bytes saved.

4. 04

   Verify that resized derivatives match the actual rendered dimensions on each breakpoint

   Serving a 2400px image into a 600px container wastes bandwidth; serving a 600px image into a retina hero looks soft. Check the three most-trafficked templates, not every page.

5. 05

   Route any asset that fails the quality check back to the creative owner, not the optimization queue

   Compression cannot repair a badly exported source file. Send it back with the specific defect noted so the retoucher or designer fixes the root cause once.

6. 06

   Log the final file size, format, and page-weight delta in the client's asset register

   A running record turns a routine QA pass into evidence for the quarterly performance conversation, and it flags when a client's library is drifting back toward unoptimized uploads.

7. 07

   Schedule a re-check 30 days after handoff to catch uploads that bypassed the pipeline

   Client-side editors and ad hoc campaign uploads routinely skip the gate. A monthly sweep of the ten heaviest pages catches the regression before it shows up in a speed report.