Core Web Vitals and Conversion: A Speed Check Before Black Friday
Peak traffic exposes slow pages. What the three Core Web Vitals measure, how to read them, the usual causes of poor scores and a seven-week plan before Q4.
Every autumn the same thing happens. Teams prepare offers, campaigns and emails for the busiest weeks of the year, and then traffic arrives and exposes the one thing nobody checked: pages that are slow, jumpy or unresponsive under load. Visitors who came with intent leave because a button did not react or a banner pushed the content away at the last moment.
Black Friday falls on 27 November this year, which leaves about seven weeks from now. That is enough time to find and fix the typical problems, if you start soon. This article explains the three Core Web Vitals, how to read them, what usually causes poor scores and what to do each week.
Why speed belongs in a conversion discussion
A page that loads slowly or reacts late costs attention at the worst moment. A visitor who wants to buy or sign up is also the visitor with the least patience for a stalled button. Speed is not the only factor in conversion, and the effect differs from site to site, so do not expect a fixed percentage. It is, however, a factor that you can check and fix, and one that tends to get worse as pages collect more scripts, banners and widgets.
The three metrics
Google’s Core Web Vitals describe the experience of real visitors with three numbers. web.dev publishes the thresholds. A page is rated “good” when the value is at or below the limit for at least 75 percent of page loads:
| Metric | What it measures | Good | Poor |
|---|---|---|---|
| Largest Contentful Paint (LCP) | How quickly the main content appears | up to 2.5 s | over 4 s |
| Interaction to Next Paint (INP) | How quickly the page reacts to clicks and taps | up to 200 ms | over 500 ms |
| Cumulative Layout Shift (CLS) | How much content jumps around while loading | up to 0.1 | over 0.25 |
In plain words: does the page show what I came for quickly, does it respond when I touch it, and does it stay still while I read.
Read the right data
There are two kinds of data, and they answer different questions.
Field data comes from real visitors. It tells you what people actually experienced, on their devices and networks. You find it in the Core Web Vitals report in Google Search Console and in PageSpeed Insights, if your site has enough traffic.
Lab data comes from a test in a controlled setting, for example Lighthouse in Chrome. It helps you find causes and check fixes quickly, but it can be better or worse than reality.
Use field data to decide what matters and lab data to find out why. Check the numbers for your most important pages: the home page, the main landing pages, pricing, the sign-up flow and the checkout. And check mobile first, since that is where scores are usually worse.
Common causes and fixes
Slow Largest Contentful Paint
The largest element is often the hero image or the headline block. Typical causes:
- Oversized images. Serve the right size for the screen and use modern formats such as WebP or AVIF.
- Hero image loaded late. Do not lazy-load the main image, and make sure the browser can find it early.
- Slow server response. Caching and a content delivery network help.
- Render-blocking resources. Large stylesheets and scripts in the head delay the first paint.
If you use video in a hero, load it after the page and keep a still image in place, as we do on this site.
Poor Interaction to Next Paint
A page that is slow to react usually has too much JavaScript running at the wrong time. Typical causes:
- Third-party scripts: tag managers, chat widgets, A/B testing tools, tracking pixels, review widgets.
- Long tasks that block the main thread while the visitor taps.
- Heavy event handlers on buttons or filters.
An audit of third-party scripts is the highest-return step. List every script, who needs it and what it costs. Remove what nobody uses, load the rest later, and do not let a widget compete with the checkout. We wrote about this specifically for widgets in Integrating Feedback Widgets Without Hurting Site Speed and Lightweight, Powerful and Fast.
Layout shift
A button that moves just as you tap it is the experience this metric is meant to capture. Typical causes:
- Images and videos without set dimensions, so the browser does not reserve the space.
- Banners that appear late, such as cookie notices or promotion bars that push the content down.
- Web fonts that swap and reflow the text.
- Widgets injected into the page without reserved space.
Reserve space for everything that appears after the first paint, and show notices as overlays that do not move the content.
A seven-week plan
Weeks 1 and 2: measure and list. Pull field data for the key pages. List the worst offenders and the scripts on them.
Weeks 3 and 4: fix the biggest items. Images, the hero, the scripts that cost the most and layout shifts from banners. One change at a time, with a before-and-after measurement.
Week 5: check the journey. Run through the complete path from landing page to purchase or sign-up on a mid-range phone and on a slow connection. Look for forms and buttons that react late, as discussed in Form Optimization.
Week 6: freeze and prepare. Stop changing the site except for planned offers. Check that tracking works, that error pages are in place and that your team knows who to call if something breaks. Prepare the questions you want to ask visitors during the peak.
Week 7 and the peak: watch and learn. Check key numbers every day and keep an eye on field data. Write down what you see, so the next year starts from your notes.
More than speed: a short Q4 checklist
- Message match. Do your campaign pages continue the promise of the ads and emails? See The First Ten Seconds.
- Forms. Are checkout and sign-up forms lean and clear?
- Trust. Are delivery, returns and privacy visible where people hesitate? Building Trust Online has pointers.
- Mobile. Is the journey comfortable on a small screen?
- Tracking. Do your key events fire correctly? Test before the traffic comes, not after.
- Feedback. Is a short question ready for visitors who leave the checkout? During peak weeks it reveals problems fastest.
Measure the effect
Speed work is easy to start and easy to oversell. Measure what changes: conversion rate, bounce rate on the pages you improved and, if you can, field data a few weeks later. Compare against a period of similar traffic, and keep in mind that campaigns and the calendar change visitors too. Where you can, test properly. Kohavi, Tang and Xu explain how controlled experiments separate the effect of a change from everything else happening on the site.
The goal is not a perfect score. It is a site that does not get in the way when the visitors you paid for finally arrive.
Frequently asked questions
What are the Core Web Vitals?
Three measures of page experience: Largest Contentful Paint (loading), Interaction to Next Paint (responsiveness) and Cumulative Layout Shift (visual stability). Google publishes thresholds for each.
What are the thresholds for a good score?
At the 75th percentile of page loads: Largest Contentful Paint up to 2.5 seconds, Interaction to Next Paint up to 200 milliseconds and Cumulative Layout Shift up to 0.1.
Does a faster site always convert better?
Not automatically. Slow pages frustrate visitors and speed is a fair place to look for problems, but effect sizes vary by site and the link is not guaranteed. Measure conversion before and after a speed change instead of assuming it.
Where do I see my real-user scores?
In the Core Web Vitals report in Google Search Console and in PageSpeed Insights, which shows field data from real Chrome users where enough traffic exists. Lab tools such as Lighthouse help you find causes.
Sources and further reading
- How the Core Web Vitals metric thresholds were defined , web.dev (Google)
- Trustworthy Online Controlled Experiments: A Practical Guide to A/B Testing , Kohavi, Tang & Xu, Cambridge University Press, 2020