Measure the arrival, not a trophy.

Campaigns do not land on an abstract store. They land on exact URLs under real device, network, geography, cache, and consent conditions. We record the profile and test the buyer path that matters on that page.

• URL

  The actual destination used by the ad, including its template and campaign-specific behavior.

• Device

  The agreed mobile profile, viewport, network conditions, and test geography where relevant.

• Question

  What delays the first useful content or action for this paid arrival?

• LCP

  Largest Contentful Paint: when the main visible content likely appears. Google currently labels 2.5 seconds or less “good” at the 75th percentile of field visits.

• INP

  Interaction to Next Paint: how promptly the page responds across user interactions. 200 milliseconds or less is the current “good” field threshold.

• CLS

  Cumulative Layout Shift: how much visible content moves unexpectedly. 0.1 or less is the current “good” field threshold.

Thresholds and definitions: web.dev Core Web Vitals. Passing a threshold is not a commercial forecast and does not replace page-specific judgment.

Lab tests are controlled and repeatable. They help isolate causes and compare a proposed technical change.

Field data reflects eligible real visitors over a rolling period. Chrome User Experience Report data can take time to reflect a change and may be unavailable at an exact URL.

We do not blend the two, substitute one for the other, or present a lab score as proof of buyer behavior.

• Measurement

  What the tools and page behavior directly showed under the stated conditions.

• Recommendation

  The technical change our specialists judge most likely to remove the measured friction.

• Forecast

  A testable expectation with limits, never a guaranteed conversion, ranking, revenue, or ad-cost result.

Score hacks that improve a report while weakening the buyer path are outside the method.