A buyer on your site taps "Add to cart." Nothing happens. They tap again. Half a second later, two of the item land in the cart, the quantity field jumps, and they start wondering whether the order will be right. They didn't wait for a page to load. They waited for a click to respond.
That gap between a click and a visible response is what Interaction to Next Paint (INP) measures. Since March 2024 it has been one of Google's three Core Web Vitals, replacing First Input Delay. Many enterprise commerce teams still watch load time and not much else, and on the pages that do the most work, INP is often where they fall short.
Quick answer: INP measures how quickly a page visibly responds to clicks, taps and key presses, judged on one of the slowest interactions during a visit. Google rates INP of 200 milliseconds or less as good and anything over 500 milliseconds as poor, measured at the 75th percentile of real user visits. Ecommerce sites usually fail INP on interaction-heavy pages such as filters, variant selectors, quick order forms and carts, because too much JavaScript runs on the browser's main thread. Fixing it means cutting and breaking up that JavaScript, not just making pages load faster.
Why INP matters more for commerce than for content sites
A blog post is mostly read. A commerce site is used. Buyers filter, sort, pick sizes, change quantities, apply codes and update carts. Each of those is an interaction, and INP judges your site on how one of the slowest of them performed.
That makes INP a business metric as much as a technical one:
- Conversion. In Deloitte's Milliseconds Make Millions study, a 0.1-second improvement in mobile site speed increased retail conversions by 8.4%. Slow responses at the moment of decision make buyers hesitate, double-click or leave.
- Trust. On B2B sites, buyers are placing large, repeat orders. A filter that freezes or a quantity field that lags makes them doubt the order, and some will phone it in instead.
- Search. Core Web Vitals are part of Google's page experience signals. They won't outrank relevance, but when competing pages are close, a poor experience is a disadvantage you don't need.
Where ecommerce sites usually fail INP
1. Filters and faceted navigation
Every filter click can trigger a product query, a full grid re-render, URL updates and analytics events, all at once. On large B2B catalogs with dozens of facets, that easily blocks the main thread for hundreds of milliseconds.
Fix: Show immediate feedback first, such as a checked box and a loading state, then fetch and render results. Re-render only the products that change, and debounce rapid multi-select clicks. Our guide to faceted navigation and crawl waste covers the SEO side of the same pages.
2. Variant and configuration selectors
Choosing a size, finish or pack quantity often recalculates price, availability, images and part numbers in one blocking script.
Fix: Update the selected option and the price first, and defer image swaps and secondary content. Precompute variant data on the server so the browser isn't doing the lookup.
3. Add to cart and cart updates
Add to cart frequently fires a cart API call, mini-cart refresh, recommendation widget and several tracking pixels before anything visible changes.
Fix: Confirm the action visually straight away, then run the API call and refresh. Move tracking and recommendations off the critical path so they run after the buyer sees a response.
4. Quick order and bulk entry forms
B2B quick order pads that validate every keystroke against your catalog or ERP are a common INP problem, especially on mobile devices used on job sites and warehouse floors.
Fix: Validate on pause or on blur instead of every keystroke, and batch lookups. Keep the input responsive even while the validation request is in flight.
5. Third-party scripts
Tag managers, chat widgets, personalization tools, A/B testing scripts and review widgets all compete for the same main thread. Each one is small. Together they're often the biggest cause of poor INP.
Fix: Audit every third-party script against what it actually earns. Load non-essential tools after interaction or on idle, remove duplicates and retire tests that ended months ago.
How to find your INP problems
- Start with field data. Open the Core Web Vitals report in Google Search Console, or PageSpeed Insights, which shows real Chrome user data. Lab tools alone can't tell you how real buyers on real devices experience your site.
- Group by template. Look at category, product, search, cart and quick order templates separately. INP problems almost always cluster on a few page types.
- Find the slow interactions. Use Chrome DevTools' Performance panel, or add real user monitoring with Google's web-vitals library, which reports which element was clicked and how long each phase took.
- Test on mid-range phones. Your developers' laptops hide the problem. Throttle the CPU or test on the kind of Android device your buyers actually carry.
- Fix the top three interactions first. Prioritize by traffic and revenue: filters on top categories, add to cart on best sellers, and the cart itself.
Speed also affects AI visibility
INP is measured in browsers, but the habits that cause it, heavy client-side JavaScript and content that only appears after scripts run, also affect AI search. Many AI crawlers don't execute JavaScript, so products, prices and specs that render only on the client may never be seen. A leaner, server-rendered page tends to be faster for buyers and easier for AI answer engines to read and cite.
FAQ
Q: What is a good INP score? 200 milliseconds or less at the 75th percentile of page visits. Between 200 and 500 milliseconds needs improvement, and over 500 milliseconds is poor.
Q: Did INP replace First Input Delay? Yes. INP replaced FID as a Core Web Vital in March 2024. FID only measured the delay before the first interaction was handled, while INP covers the full response time of interactions throughout the visit.
Q: Do we need to replatform to fix INP? Usually not. Most INP problems come from theme code, third-party scripts and how specific components are built, and those can be fixed on your current platform. Replatforming makes sense only if the platform itself forces heavy client-side rendering you can't change.
The bottom line
Load time gets buyers onto the page. INP decides whether the page feels like it works once they start using it. Find the interactions that matter most, make them respond instantly, and move everything else out of the way.
Echidna reviews site performance as part of our UX and SEO/GEO assessments. Start a UX website review, an SEO/GEO review, or both, and we'll show you which interactions are slowing your buyers down.