Editorial · INP · The newest Core Web Vital

INP Monitoring Cost 2026

Interaction to Next Paint (INP) became a Core Web Vital in March 2024, replacing FID. Every modern RUM tool captures INP, but tooling depth and attribution detail vary widely. Here's what to look for.

Direct answer

What is INP and why does it cost money to monitor?

INP measures the latency between a user interaction (click, tap, keypress) and the next visual update. Threshold: under 200ms is 'good'. INP requires real user data, synthetic monitoring can't simulate the variety of user interactions. RUM tools that capture INP charge per session or per beacon; cost ranges from free (Cloudflare) to $4.50/1K sessions (Datadog).

Why INP is harder than the other Core Web Vitals

LCP and CLS are page-load metrics, captured once per page view. INP is an interaction metric, captured every time the user clicks, taps, or types. A long session with many interactions produces many INP measurements; the worst is reported. This means:

INP scales with engagement, not just page views.
INP regressions can hide on low-engagement pages (no interactions = no INP data).
INP attribution requires capturing the long task, the synchronous render work, the event handler that caused the delay, which is more expensive than capturing a simple paint timestamp.

RUM tools with the best INP capture

SpeedCurve — performance-budgets feature includes INP regression alerts.
Datadog RUM — captures INP with long-task attribution; you can see which event handler caused the worst INP.
Sentry — captures INP with profiling integration for attribution.
New Relic Browser — captures INP, dashboards it, alerts on regression.
Cloudflare Web Analytics — captures INP as part of Core Web Vitals, but no attribution beyond URL/country/browser segment.

Tools with weaker INP capture

Older tools that haven't updated for INP's 2024 promotion to Core Web Vital may still rely on FID or only capture INP without attribution. Check each vendor's INP-attribution documentation before committing, vendor pricing pages don't always disclose this.

INP-specific cost considerations

Per-event vendors (AWS CloudWatch RUM) bill each interaction event separately. Heavy SPA workloads with many interactions per session generate more events and cost more.
Per-session vendors (Datadog, Dynatrace, Sentry) bill once per session regardless of interaction count. Cheaper for app-style SaaS with many interactions.
Per-pageview vendors (Atatus, SpeedCurve) bill once per page load regardless of interactions. INP captured passively.

The INP-driven cost increase

Pre-2024, many teams ran RUM with sampling at 10-25% and only captured page-load metrics. Adding INP capture at 100% to existing 10% sampling increases the event count by 5-10x on engaged applications. This produces unexpected RUM bill increases mid-2024 onwards for teams that didn't notice the change.

For the broader Core Web Vitals context, see Core Web Vitals monitoring. For tool-specific INP capture detail, see the per-vendor pricing pages linked above.

Last verified June 2026