How to Read a Lighthouse Report (For Ecommerce Stores)
Learn to interpret your Lighthouse report for ecommerce. Understand Core Web Vitals, lab vs field data, and prioritize which metrics to fix first for better conversions.
Your Lighthouse report shows a score of 58. Your competitor shows 89. But which store actually converts better?
The answer isn't always what you'd expect. A perfect Lighthouse score doesn't guarantee sales, and a mediocre score doesn't mean you're losing customers. What matters is knowing which metrics affect your revenue and which ones you can safely ignore.
The short version: Focus on three (LCP, INP, ) because they affect Google rankings. Prioritize field data over lab data. Fix product pages before blog posts. Image optimization is usually your quickest win.
If you're here because your store feels slow, conversions dropped after a redesign, or you just installed a new app and everything got worse, Lighthouse is where the problem shows up first.
Only 48% of mobile websites pass Core Web Vitals according to HTTP Archive data. The numbers matter: improving page speed by just 0.1 seconds can boost conversion rates by 8.4% according to Deloitte's research.
This guide focuses on reading and interpreting Lighthouse reports for ecommerce stores. You'll learn what each metric means for your business, which numbers to prioritize, and how to make sense of conflicting data. For fixing the issues you find, see the site speed optimization guide.
What Is a Lighthouse Report
Lighthouse is a free tool developed by Google that audits web pages for performance, accessibility, , and best practices. When you run a Lighthouse audit, it simulates a user visiting your page and measures how quickly content appears and becomes interactive.
You can run Lighthouse in several ways:
- PageSpeed Insights at pagespeed.web.dev (easiest for most store owners)
- Chrome DevTools by pressing F12 and selecting the Lighthouse tab
- Command line for developers who want automated testing
The report generates scores from 0-100 for each category, with performance being the most relevant for ecommerce conversion rates.
Understanding the Performance Score (0-100)
A low score often means slow-loading product images or too many apps running in the background.
What the Score Represents
The performance score is a weighted average of six metrics. According to Chrome for Developers documentation, the current weights are:
| Metric | Weight | What It Measures |
|---|---|---|
| Total Blocking Time (TBT) | 30% | How long the page is unresponsive |
| Largest Contentful Paint (LCP) | 25% | Main content load time |
| Cumulative Layout Shift (CLS) | 25% | Visual stability |
| First Contentful Paint (FCP) | 10% | When something first appears |
| Speed Index | 10% | How quickly the page looks complete |
TBT and LCP together account for over half the score. If your score is low, these two metrics are often the cause.
The scoring uses real-world data from millions of websites. A score of 50 means your site performs better than roughly 25% of all websites. A score of 90 puts you in the top 8%.
Why the Score Alone Can Mislead You
Two stores can both score 65 for very different reasons. One might have a single metric dragging everything down, while another has moderate issues across the board.
Look at individual metrics instead of fixating on the overall number. A store with an LCP of 6 seconds and perfect scores everywhere else needs different fixes than a store with moderate problems in every category.
Scores also fluctuate between tests. Network conditions, server load, and even the testing device affect results. Google's own documentation notes that Lighthouse scores vary due to inherent variability in web and network technologies. Run multiple tests and look at the pattern rather than any single result.
The 7 Metrics That Matter for Ecommerce
Core Web Vitals (Google Ranking Factors)
Three metrics directly affect your Google search rankings. These Core became a ranking factor in 2021 and remain part of Google's page experience signals.
Largest Contentful Paint (LCP)
If LCP is bad, customers see a blank screen instead of your products.
LCP measures how long until your main content appears. For product pages, this is typically your hero image or main product photo.
| Rating | LCP Time |
|---|---|
| Good | Under 2.5 seconds |
| Needs Improvement | 2.5 to 4 seconds |
| Poor | Over 4 seconds |
For ecommerce, LCP directly affects whether customers see your products or a blank screen. Product pages can see 40-50% lower conversion rates when comparing a 2-second LCP to a 4-5 second LCP.
When Vodafone improved their LCP by 31%, they saw an 8% increase in sales.
Interaction to Next Paint (INP)
If INP is bad, buttons feel broken when customers try to add items to cart.
INP replaced (FID) as a Core Web Vital in March 2024. It measures how quickly your page responds when customers click, tap, or type.
| Rating | INP Time |
|---|---|
| Good | Under 200 milliseconds |
| Needs Improvement | 200 to 500 milliseconds |
| Poor | Over 500 milliseconds |
For ecommerce, INP affects every interaction: clicking "Add to Cart," selecting size options, typing in search boxes, opening product galleries. A slow INP means frustrated customers who click buttons that seem unresponsive.
The key difference from the old FID metric: INP measures all interactions throughout the visit, not just the first one. This matters for ecommerce because the most important interactions (adding to cart, proceeding to checkout) often happen later in the session.
RedBus improved their INP by 72% and saw a 7% increase in sales.
Cumulative Layout Shift (CLS)
If CLS is bad, customers misclick because elements jump around during checkout.
CLS measures visual stability. A high CLS means elements jump around as the page loads.
| Rating | CLS Score |
|---|---|
| Good | Under 0.1 |
| Needs Improvement | 0.1 to 0.25 |
| Poor | Over 0.25 |
For ecommerce, layout shifts cause misclicks. A customer aims for "Add to Cart" but hits "View Details" because a banner loaded late and pushed everything down. This frustration compounds when it happens during checkout.
Common causes on ecommerce sites: images without dimensions, dynamically loaded reviews, promotional banners, and chat widgets that appear after initial load.
Supporting Metrics
Four additional metrics help diagnose specific problems but don't directly affect rankings.
First Contentful Paint (FCP)
FCP measures when the first piece of content appears. Good is under 1.8 seconds. For ecommerce, this is the moment customers know your site is loading rather than broken.
Total Blocking Time (TBT)
TBT measures how long the main thread is blocked by JavaScript. Good is under 200 milliseconds. This metric often exposes problems with like chat widgets, review apps, and analytics tools. Our analysis of 1,533 Shopify stores shows which specific services block the main thread the most.
Time to First Byte (TTFB)
TTFB measures . Good is under 800 milliseconds. A slow TTFB often indicates hosting issues or database problems that affect everything else.
Speed Index
Speed Index measures how quickly the page visually completes. Good is under 3.4 seconds. It captures the overall perception of load speed.
Lab Data vs Field Data: Which to Trust
Field data is what Google actually uses for rankings. Lab data helps you debug.
Understanding Lab Data
Lab data comes from Lighthouse running a simulated test on a controlled machine with throttled CPU and network. Every time you run PageSpeed Insights, the "Lighthouse" section shows lab data.
Lab data is useful for debugging. It's consistent and reproducible, which helps when testing changes. But it simulates one specific scenario that may not match your actual customers.
Understanding Field Data (CrUX)
Field data comes from real Chrome users visiting your site. Google collects this through the Chrome User Experience Report (CrUX) and displays it in the "Discover what your real users are experiencing" section of PageSpeed Insights.
Field data is aggregated over a 28-day rolling period. The 75th percentile determines your Core Web Vitals rating, meaning 75% of your visitors had an experience at least that good.
Here's the important distinction: only field data affects your Google search rankings, not lab data. The gap can be dramatic: in our Shopify Speed Benchmark, only 1% of stores passed LCP in lab tests, but 85% passed in CrUX field data.
When Lab and Field Data Disagree
It's common to see different results. A few scenarios:
Good lab, poor field: Your simulated test looks fine, but real users experience slower performance. This often means your actual customers have slower devices or connections than the test simulates. Mobile users in particular may have older phones or spotty connections.
Poor lab, good field: The test conditions are harsher than your actual audience experiences. Your customers might have faster devices or better connections than average. This is more common for stores targeting tech-savvy or higher-income demographics.
Different LCP elements: The image Lighthouse identifies as your LCP element in lab tests may not match what real users see. , A/B tests, or can cause this discrepancy.
Reading the Distribution Bars
Field data includes distribution bars showing what percentage of users fall into each category. If you see:
- 40% Good (green)
- 35% Needs Improvement (orange)
- 25% Poor (red)
This means a quarter of your visitors have a poor experience for that metric. Even with an overall "pass," that 25% represents real customers who may bounce or abandon carts.
Why Lighthouse Alone Isn't Enough for Ecommerce
Lighthouse tells you what's slow. It doesn't tell you which slow pages cost you the most money.
A product page loading in 4 seconds matters more than a blog post loading in 4 seconds. Lighthouse treats them the same. Your revenue doesn't.
EcomHint's speed diagnostic adds the business context: which pages drive conversions, which issues affect checkout, and where to focus first.
How to Prioritize What to Fix
The Priority Framework
Google recommends fixing "Poor" URLs first, then "Needs Improvement." Within each status, prioritize by:
- Number of affected URLs: A problem affecting your entire product catalog matters more than one affecting a single landing page
- Importance of URLs: Checkout and product pages drive revenue more directly than blog posts
- Metric impact: Core Web Vitals affect rankings; supporting metrics affect user experience but not SEO directly
Mobile First
Google uses mobile-first , so your mobile scores matter most for rankings. Desktop performance affects customer experience but has less SEO impact.
Around 48% of mobile pages pass Core Web Vitals compared to 56% of desktop pages according to HTTP Archive data. Mobile is harder to optimize because of weaker CPUs and variable network conditions.
Ecommerce-Specific Prioritization
For interaction-heavy ecommerce sites, consider this order:
- LCP on product pages: Customers need to see products quickly
- CLS on checkout: Layout shifts during payment entry cause abandoned carts
- INP on add-to-cart flows: Unresponsive buttons feel broken
- TBT across all pages: Third-party scripts accumulate
- TTFB if consistently slow: Hosting issues affect everything else
The Effort vs Impact Matrix
| Fix | Impact | Effort |
|---|---|---|
| Image optimization | High (LCP, Speed Index) | Low |
| Remove unused JavaScript | High (TBT, INP) | Medium |
| Add image dimensions | Medium (CLS) | Low |
| Upgrade hosting | High (TTFB) | High |
| Third-party script audit | High (TBT) | Medium |
Start with high-impact, low-effort fixes. Image optimization often provides the fastest wins.
Common Issues in Ecommerce Lighthouse Reports
Across 190 audited ecommerce stores, image optimization is the most common performance issue. Around 68% fail this check, while only 28% have problems severe enough to fail (EcomHint audit data). Knowing what issues are most common helps you benchmark your own report.
"Reduce Unused JavaScript"
This appears when your pages load JavaScript that isn't used immediately. On ecommerce sites, every app, plugin, and tracking script adds JavaScript.
Shopify stores commonly see this because each installed app adds its own scripts. A store with 15 apps might load JavaScript from all of them on every page, even when most functionality isn't needed.
The fix involves auditing your apps and deferring non-critical scripts. Deferring scripts can lead to meaningful Lighthouse improvements, sometimes doubling performance scores.
"Properly Size Images"
Product photos uploaded at full resolution cause this. A 4000x4000 pixel image displayed at 400x400 pixels wastes bandwidth and slows LCP.
Ecommerce sites have hundreds or thousands of product images, making this a common issue. The fix involves resizing images to display dimensions and using responsive image techniques.
"Eliminate Render-Blocking Resources"
and JavaScript files that block rendering delay when content appears. Theme files often cause this because they load everything before showing anything.
(inlining the styles needed for above-the-fold content) and deferring non-critical scripts address this issue.
"Serve Images in Next-Gen Formats"
JPEG and PNG product photos can often be 30-50% smaller as or AVIF. Many ecommerce platforms now support automatic format conversion.
FAQ
How often should I check my Lighthouse score?
Check weekly during active optimization work. Monthly for ongoing monitoring. The store speed test is a quick way to check your current scores without leaving the browser. Field data updates daily in PageSpeed Insights, but represents a 28-day rolling average, so major changes take about a month to fully reflect.
Why does my score change every time I run the test?
Lighthouse scores vary due to network conditions, server load, and test environment variability. Run 3-5 tests and look at the median rather than any single result. Differences of 5-10 points between runs are normal.
What's a good Lighthouse score for an ecommerce store?
Aim for 90+ on desktop and 50+ on mobile for performance. Mobile scores are typically 20-40 points lower than desktop for the same site due to simulated throttling. More important than the overall score: ensure all three Core Web Vitals pass the "good" threshold. For ecommerce-specific benchmarks, see what these scores mean for online stores.
Should I focus on mobile or desktop scores?
. Google uses mobile-first indexing, so mobile Core Web Vitals directly affect rankings. Desktop affects customer experience but has less SEO impact.
My field data says "No data available." What does that mean?
CrUX requires sufficient traffic over 28 days to generate field data. New sites or low-traffic pages won't have field data. Use lab data for optimization while building traffic. Field data will appear once you have enough visitors using Chrome with usage statistics enabled.
What to Do Next
Your Lighthouse report tells you what's wrong. The numbers show which problems cost you the most conversions.
Start with your Core Web Vitals. If any show "Poor," fix those first. Then work through "Needs Improvement" ratings, prioritizing by business impact.
Remember: field data matters more than lab data for rankings. A passing grade means 75% of real visitors have a good experience. That other 25% still represents potential lost revenue.
Use PageSpeed Insights to check your Lighthouse metrics, or use the EcomHint speed test for ecommerce-specific context alongside your scores. Once you know what's wrong, see which fixes to prioritize first in 2026 or follow our complete site speed optimization guide for step-by-step fixes.
Jakub is the founder of Ecomhint, an AI-powered ecommerce audit tool focused on UX and conversion optimization. He helps online stores identify friction points across product pages, cart, and checkout using CRO best practices and original research data.
