Blog August 24, 2026 7 min read

A large geometric button split down the middle, one half lit warm amber and responsive, the other dull slate and delayed.

The Interactivity Gap: How Slow Button Clicks Cost You 14% of Your Sales

A potential patient lands on a Tampa Bay chiropractor’s website. She scrolls the homepage, finds the booking section, and taps “Schedule an Appointment.”

Nothing happens.

She taps again. Still nothing, for about a quarter of a second.

Then the form opens.

She has already decided the site is broken. She closes the browser and goes back to Google.

She did not bounce because the page was slow to load. The page loaded fine. She bounced because the page failed to respond the moment she touched it, and on a smartphone, a quarter-second of silence reads as a malfunction.

This is the Interactivity Gap. And it is one of the most underdiagnosed revenue problems in small business web performance.

What INP Actually Measures

INP stands for Interaction to Next Paint. In plain business terms: INP measures how long your site takes to visually respond after someone taps or clicks.

Load time and interactivity are different measurements. A page can be fully loaded, images rendered, text visible, no spinner, and still take 200 or 300 milliseconds to respond when a user presses a button. To a human finger, that gap is perceptible. To a mobile user, it looks like the site is hung.

Google formally replaced its previous interactivity metric (FID, or First Input Delay) with INP in March 2024. INP is stricter: it measures not just the first click, but every interaction throughout the session. A site can pass FID and fail INP badly, which is why a significant share of sites that thought they were “good” on performance found themselves failing Google’s new standard.

Google’s current threshold for good INP is under 200 milliseconds, measured at the 75th percentile of real users visiting your site. Anything above 500ms is classified as “poor.”

The Evidence: 500 Million Sessions, One Metric

In September 2025, Yottaa published an interactive Web Performance Index built on 500 million shopper sessions across 700-plus ecommerce brands, their largest dataset to date, with daily updates. Among its findings: reducing INP from 200 milliseconds to 50 milliseconds tracks with a 14% lift in conversion rate.

That figure requires an honest label. This is a correlation, not a controlled experiment. Yottaa observed the relationship between INP scores and conversion rates across a very large dataset. Sites with better INP also tend to be better-built overall, so some of the conversion lift may come from those correlated improvements. Yottaa’s data shows the pattern; it does not isolate INP as the sole cause.

But a 14% correlation across 500 million sessions from 700-plus brands is not a coincidence or an artifact. At that scale, the pattern is strong enough to guide a decision.

A separate, smaller study published on Google’s web.dev platform in 2023 provides a useful complement. Contentsquare analyzed 997 customer sites across May through July 2023 and found that users experiencing “good” INP converted at 2.5%, while users experiencing “poor” INP converted at 2.0%, a 25% relative gap on mobile. Contentsquare notes explicitly that this is a correlation study: “correlation does not equal causation,” and users who bounce quickly may not even generate an INP reading. That caveat is real. The direction of the finding is consistent across multiple independent datasets.

A Named Company, a Specific Fix, a Measurable Outcome

redBus is a large Indian bus-ticket booking platform. In 2023, their engineering team documented a specific technical change to their search results page, improvements to how the page handled JavaScript interactions, that reduced INP by 72%. The change was measured in real user monitoring data before and after.

Following that INP improvement, redBus recorded a 7% increase in overall sales.

The redBus case study, published on Google’s web.dev platform in 2023, is one of the strongest pieces of evidence available for INP’s direct business impact. It is not a controlled randomized experiment. But it is a named company, a specific documented code change, a before/after measurement with real user data, and a revenue outcome. That is a substantially higher standard of evidence than most industry benchmarks.

For a chiropractor booking page in Tampa Bay, the technical mechanism is the same: if the JavaScript handling a form submission or a calendar widget is slow to execute, INP suffers, the interaction feels broken, and users who would have converted leave instead.

The Dollar Math for a Local Service Business

Take a hypothetical Tampa Bay law firm with 1,000 monthly visitors and a 3% rate of consultation-form completions. At a $2,000 average case value, conservative for a personal injury or estate planning practice, that’s 30 consultations a month, or $60,000 in monthly revenue connected to the website.

Now apply the 14% conversion-rate association from Yottaa’s 2025 dataset. A site with poor INP running at the 200ms threshold versus one optimized to 50ms shows a 14% conversion gap in the data.

14% of 30 consultations = 4.2 additional consultations per month.

4.2 × $2,000 = $8,400 per month.

Over a year: $100,800.

This is illustrative arithmetic, not a guarantee. The Yottaa figure is correlational and measured across ecommerce behavior, not legal-services inquiry forms. But the principle is sound: if your site takes a quarter-second to respond to every tap, a meaningful share of people who intended to contact you won’t complete the action.

”But My Site Loaded in 2 Seconds, I Passed”

This is the most common misconception about INP: that load time and interactivity are the same thing.

They are not.

Load time, typically measured as LCP, or Largest Contentful Paint (how quickly the main visible content loads), tells you how long before the page looks ready. INP tells you how long after a user acts before the page responds. A site can load in under 2 seconds and still have an INP of 400 milliseconds on a critical button.

CLS, Cumulative Layout Shift, which measures how much the page jumps around as elements render, is the third Core Web Vitals metric, and it too is independent of load time. All three must pass simultaneously for Google to classify your site as “Good.”

The HTTP Archive Web Almanac 2024, real-user data from 16.9 million websites, collected in June 2024, found that only 43% of mobile sites pass all three Core Web Vitals under the current INP-era rules. At the 90th percentile of all mobile sites, INP is 275 milliseconds, well above Google’s 200ms threshold. The average struggling mobile site is not close to “good” on interactivity.

If you ran PageSpeed Insights last year and got a passing score, that score may have been measured under the old FID metric. Under INP, your status may have changed in March 2024 without any change to your site.

How to Find Your INP Score Right Now

Google’s PageSpeed Insights tool at pagespeed.web.dev will show you INP, but only in the “Field Data” section, real measurements from Chrome users visiting your actual site. The Lighthouse score at the top of the report is a lab simulation. Both are useful; Google uses the field data for ranking purposes.

Look for the “Core Web Vitals Assessment” block. You need all three metrics, LCP, INP, and CLS, to show green at P75 (the 75th percentile). If INP shows yellow or red, you have a documented interactivity problem. The question then is where in the page’s JavaScript the delay originates, which requires a technical audit to identify.

What Fixing INP Actually Involves

INP problems almost always trace to specific JavaScript execution on specific interactions. Common culprits: third-party chat widgets, cookie-consent scripts, form libraries, and analytics tags that all compete for the browser’s attention at the moment a user clicks. The fix is surgical, identify the blocking scripts, defer or replace them, measure the result.

This is not a site rebuild. The redBus team fixed a 72% INP improvement with a documented, scoped code change. The work is technical, but it is bounded.

The Free Assessment

If you’ve read this far, you want to know what your site’s INP score is and what it’s costing you. That number is available, and it’s specific to your traffic and your revenue, not a generic benchmark.

iServU’s free assessment pulls your actual CrUX field data, identifies failing metrics including INP, and translates the gap into an estimated revenue figure based on your own numbers. No account manager, no handoffs, Gabriel Lopez-Seco runs the assessment directly.

Book a free assessment at iServU →