Storefront Speed Journal

Reviewed November 17, 2026

Your Store Passes LCP and Fails INP: The Interaction Problem Nobody Audits

Check INP separately, because it fails for different reasons than LCP. Interaction to Next Paint measures the time from a click, tap, or key press to the browser painting the response, and Google's good threshold is under 200 milliseconds at the 75th percentile. Images do not cause INP failures. JavaScript does: app scripts, tag managers full of marketing pixels, and heavy event handlers all queue on the main thread, and when a shopper taps add to cart, their tap waits behind whatever your tags are doing. The audit is a list of what runs on interaction, not another image pass.

What the main thread is doing while your customer waits

Every interaction starts a race on one thread. The browser has to run the page's event handlers, and anything else already executing, before it can paint the result. A single long task over 50 milliseconds is enough to feel sluggish; a stack of them from chat, reviews, personalization, and analytics scripts turns a tap into a visible pause. INP reports the worst typical interaction across the visit, so one bad handler on the add to cart button can define the score for the whole page.

This is why INP failures cluster on stores that pass LCP comfortably. The image work was done, the hero is fast, and then three years of marketing tags were layered onto a tag manager with no budget and no owner. Each tag was small. The sum is the metric.

The audit: list, time, and cut in that order

Open your tag manager and your app list side by side and write down everything that loads on product and collection pages. For each one, ask what fires on interaction versus on page load, and whether it needs to. Consent tools, chat widgets, and review platforms are frequent offenders because they attach handlers broadly and load eagerly for a feature a minority of sessions use. Loading chat on first interaction intent, deferring non essential pixels until after the main content settles, and removing the duplicate analytics tag from the store that migrated platforms two years ago are the standard first cuts.

Measure with field data, not feelings. The Chrome User Experience Report splits INP by page type in Search Console's Core Web Vitals report, and the pages flagged there are where the audit starts. Lab tools can reproduce a slow interaction with throttling, but the 75th percentile of real shoppers is the grade, and it updates on a 28 day window, so a fix takes weeks to show its full effect.

What good looks like after the cuts

A healthy storefront keeps interaction handlers light: the add to cart click does its own work and paints feedback, a spinner or a state change, immediately, while the heavier consequences (cart drawer rendering, upsell queries, tracking calls) happen after that first paint or in the background. That ordering is the entire discipline. The shopper's tap gets an answer inside 200 milliseconds, and everything else queues behind the answer instead of in front of it.

Put a budget in writing when you finish: how many third party scripts may load before interaction on a product page, and who approves a new one. Without that line, the tag collection regrows at the rate of one marketing request a month, and next year's INP audit is this year's audit with different names in the list. Speed work that is not defended by a rule is a loan, not a fix.

Related reading

Related reading on this site: Unused CSS in Shopify Themes: Finding the Styles Your Pages Load and Never Use, Your Store Is Not Slow on Your Phone: Testing on the Mid Range Device Your Customers Use.

Frequently asked questions

What is INP and what is a good score?

Interaction to Next Paint measures the delay between a user's click, tap, or key press and the browser painting the response. Under 200 milliseconds at the 75th percentile of real visits is Google's good threshold. INP replaced First Input Delay as a Core Web Vital in March 2024.

Why does my store pass LCP but fail INP?

They measure different systems. LCP is about loading the largest element, usually an image. INP is about JavaScript competing on the main thread when a shopper interacts. A store can have a fast hero image and still make every tap wait behind app scripts and marketing tags.

Which scripts usually cause INP problems on Shopify?

Chat widgets, review and personalization apps, consent tools with broad handlers, and tag managers carrying years of marketing pixels, including duplicate analytics tags from old platform migrations. Audit what loads on product pages and what fires on interaction, then defer or remove what does not earn its place.

How long until an INP fix shows in my scores?

Field data is a rolling 28 day window at the 75th percentile, so improvements phase in over weeks as new visits replace old ones. Use lab tests to confirm the interaction is faster today, and judge the official metric after a full window has passed.