How to Run a Core Web Vitals Audit (No Developer Needed)

Most guides to Core Web Vitals assume you already know what a render-blocking script is. This one doesn't. If you own or market a small business site and just got a vague "poor" warning in Search Console, here's the walkthrough: which free tools to open, in what order, what the numbers mean, and what to fix first.

What a Core Web Vitals Audit Actually Checks

A core web vitals audit measures how real visitors experience your site's loading speed, responsiveness, and visual stability — not just how fast a page loads in a single lab test. That distinction matters more than any single score.

The audit centers on three metrics:

• Largest Contentful Paint (LCP) — how long it takes the biggest visible element (usually a hero image or headline block) to render. It's your loading-speed proxy.
• Interaction to Next Paint (INP) — how long the page takes to visibly respond after someone clicks, taps, or types. It replaced First Input Delay (FID) as the official responsiveness metric in March 2024, so any guide still citing FID is outdated.
• Cumulative Layout Shift (CLS) — how much content jumps around while the page loads. Think of a button shifting down just as someone taps it.

A real audit combines lab data (a controlled, one-time test) with field data — real measurements from actual visitors, aggregated by Google's Chrome UX Report (CrUX). Field data is what Google uses for ranking purposes under the Page Experience update, so it's the number that matters most, even though lab data is easier to act on.

The 2026 Thresholds: What Counts as Good, Needs Improvement, or Poor

Each metric has three bands, and Google scores your page based on the 75th percentile of real user visits — meaning at least 75% of measured page loads must hit the "good" threshold before Google considers the page good overall. A handful of fast visits won't save you if most users have a slow experience.

Current core web vitals thresholds for 2026:

• LCP: Good is under 2.5 seconds. Needs Improvement is 2.5–4 seconds. Poor is anything above 4 seconds.
• INP: Good is under 200 milliseconds. Needs Improvement is 200–500ms. Poor is above 500ms.
• CLS: Good is under 0.1. Needs Improvement is 0.1–0.25. Poor is above 0.25.

These figures are confirmed by both corewebvitals.io and RoastWeb's 2026 breakdown; conflicting thresholds elsewhere likely come from a page that hasn't been updated since the FID-to-INP switch.

How to Run the Audit: A 4-Step Manual Process

This order avoids confusion, because each tool answers a different question.

Step 1: Search Console's Core Web Vitals report. This is your field data — real visitor experiences, grouped into Good/Needs Improvement/Poor URL groups. Start here because it tells you which pages actually have a problem in the real world.

Step 2: PageSpeed Insights, per URL. Plug in each flagged URL. It's a genuinely free core web vitals audit tool that shows both field data (if the URL has enough CrUX traffic) and a lab test side by side, letting you confirm the Search Console flag and spot likely causes.

Step 3: Lighthouse in Chrome DevTools. Open DevTools, run a Lighthouse report on the specific page, and read the diagnostics — render-blocking resources, unoptimized images, unused JavaScript. This lab data, generated on your machine, gives developer-adjacent detail without needing a developer to run it.

Step 4: Cross-check against a second lab tool like GTmetrix if scores still seem inconsistent. Optional, but useful for confirming a stubborn issue isn't a fluke of one test run.

This sequence answers "how to do a core web vitals audit" without guessing which tool to trust — you're moving from real-world evidence to a lab-level diagnosis, not the other way around.

Why Your Scores Don't Match Across Tools

This is the most common source of panic, and it has a simple explanation: lab data and field data measure different things. Lighthouse and a fresh PageSpeed Insights lab run reflect one simulated visit, on one connection speed, at one moment. Search Console's field data reflects an aggregate of real visits over the past 28 days, across every device and connection type your actual traffic uses.

So a page can pass Lighthouse cleanly and still show "poor" in Search Console, because your real visitors are on older phones, slower connections, or further from your server than your test conditions assumed. Neither number is wrong — they're answering different questions. If this gap keeps tripping up your reporting, it's worth reading how performance analytics data can be reconciled across tools so you're not chasing two contradictory dashboards every month.

Which Metric to Fix First

Once you've got real numbers, resist the urge to fix whatever's easiest. Use this triage order instead:

1. Fix whichever metric is flagged "Poor" first, regardless of which one it is — poor scores are actively hurting rankings and conversions.
2. Then INP, since sluggish interactivity is the fastest way to lose a frustrated visitor mid-action.
3. Then LCP, because slow loading is usually fixable with image compression, caching, or removing render-blocking scripts.
4. Then CLS, often the cheapest fix (reserving space for images and ads) but the lowest-impact if the other two are still poor.

This is your core web vitals checklist in one line: worst-first, then INP, then LCP, then CLS. Once you know what's broken, the harder question is how to fix it — and whether that's a one-off dev task or a tool you need running continuously. That decision is covered in the guide comparing a speed optimization tool versus hiring a service. It's also worth situating Core Web Vitals inside your broader technical SEO audit checklist and the UX framework it feeds into — vitals are a symptom, and UX is usually the root cause.

How Often to Re-Audit — and When to Stop Doing It by Hand

For most SMB sites, a monthly check of Search Console's Core Web Vitals report is enough to catch drift before it becomes a ranking problem, plus an extra check any time you ship a redesign, swap themes, or add a new third-party script (chat widgets and ad tags are frequent CLS offenders). Assign one owner — usually whoever manages the site or marketing stack — so it doesn't fall through the cracks between the marketing and dev sides of a small team.

The manual process above works, but it doesn't scale well past a handful of pages, and nobody on a small team wants a recurring calendar reminder to re-run Lighthouse forever. That's the point where core web vitals monitoring stops being worth doing by hand. Evra bundles this tracking into the same subscription that already covers your other marketing tools, watching field data continuously and flagging regressions before they cost you rankings — instead of you finding out during next month's manual pass. For a deeper look at how automated auditing catches what manual checks miss, see Evra's guide to AI-driven Core Web Vitals audits.

Frequently Asked Questions

What's a good core web vitals score in 2026?

A good score means all three metrics meet their "Good" threshold at the 75th percentile of real visitor data: LCP under 2.5 seconds, INP under 200 milliseconds, and CLS under 0.1. Google requires at least 75% of measured page visits to hit these marks — a few fast lab test runs don't count.

Is PageSpeed Insights the same as a core web vitals audit?

No, it's one tool within a full audit. A complete audit combines Search Console's field data, PageSpeed Insights' per-URL scores, and Lighthouse's lab diagnostics to both confirm a problem and diagnose its cause.

Why does my site pass Lighthouse but fail Core Web Vitals in Search Console?

Because Lighthouse measures one simulated visit under lab conditions, while Search Console reports aggregated field data from real visitors on real devices and connections over the past 28 days. A page can look fast in a controlled test yet perform poorly for visitors on slower phones or networks.

How often should I re-run a core web vitals audit?

Monthly is a reasonable baseline for most small business sites, with an additional check after any redesign, theme change, or new third-party script. Sites with frequent content or template changes may need it checked more often.

Do Core Web Vitals really affect Google rankings?

Yes, they're part of Google's Page Experience signals, and pages with poor field data can be disadvantaged against otherwise similar competitors. They also affect conversions directly, since slow or jumpy pages tend to lose visitors before they take action.

Can I audit core web vitals without a developer?

Yes, the four-step manual process using Search Console, PageSpeed Insights, and Lighthouse is designed for non-technical site owners. Fixing certain issues may still require developer help, but identifying and prioritizing them doesn't.

Running this audit yourself costs nothing but time, and it's worth doing at least once so you understand what the numbers mean for your own site. Once that becomes a monthly chore competing with everything else on your plate, Evra folds continuous Core Web Vitals monitoring into the same platform already running your other marketing tools.