Setting Baselines and Comparing Before/After

Purpose

Explain how to set a baseline and compare before and after performance when evaluating Speed Layer.

Step 1: Choose the pages that matter

Pick a small set of pages that represent the main shopper journey:

• Homepage
• SRP
• VDP
• Lead form pages

Step 2: Choose the metrics

Use a mix of:

• Core Web Vitals (LCP, CLS, INP)
• Supporting metrics (FCP, TTFB)

Google reference:

• Core Web Vitals https://web.dev/vitals/

Step 3: Choose a baseline time window

A baseline should be long enough to smooth day to day variance. Many teams use a week or more.

Step 4: Compare using percentiles

Percentiles such as p75 are useful for understanding typical user experience. Use the same percentile for before and after.

Step 5: Keep changes isolated when possible

To attribute improvements clearly:

• Avoid making major site changes during the test period
• Track third party tag changes separately
• Keep installation and configuration consistent

Step 6: Segment your results

Break down by:

• Mobile vs desktop
• Page type

Related pages

• Reading Trends vs One-Time Scores
• Common Misreads of Performance Data (and How to Avoid Them)
• Why Real User Experience Matters More Than Scores