Pricing model deep-dive
Per-Pageview RUM Pricing Models 2026
Some vendors meter cost per individual page load rather than per session. A session typically contains 2-4 page views, so per-pageview rates appear cheaper but are roughly equivalent when normalised.
Direct answer
Is per-pageview pricing cheaper than per-session?
Not really, a typical session contains 2-4 page views, so a $1/1K pageview rate is roughly equivalent to a $2-4/1K session rate. Per-pageview pricing is more predictable for workloads with stable pages-per-session ratios. Atatus is the main published per-pageview RUM vendor.
The conversion math
A page view is one URL load by one browser. A session is a contiguous period of activity by one visitor, typically containing multiple page views before the visitor leaves (defined by 30 minutes of inactivity or session expiry).
To convert per-pageview rates to per-session-equivalent rates, multiply by the average pages-per-session for your workload:
- Content/blog sites: 1.5-2.5 pages/session
- E-commerce sites: 3-5 pages/session
- App-style SaaS: 5-15 pages/session
- Marketing landing pages: 1-1.5 pages/session
So an illustrative $0.49/1K-pageview rate is roughly $1-2.50/1K-session on a content site, $1.50-2.50/1K-session on an e-commerce site, or $2.50-7.50/1K-session on an app-style SaaS.
Vendors using per-pageview pricing
- Atatus — historically metered per pageview; as of 2026 it publishes no tier rates and quotes custom pricing.
- SpeedCurve — tiered usage including page views and synthetic checks. Starter $90/mo.
- Pingdom RUM — priced by page-view tier through a self-serve calculator (100K to 1B page views/month); no static rate is published.
When per-pageview pricing wins
- Predictable pages-per-session. A content site with stable ~2 pages/session can forecast pageview cost more accurately than session cost.
- App-style sites that don't pay extra per page visit. A SaaS dashboard with 15 pages/session can be expensive on per-session billing if vendors meter every page transition as a session event.
- Avoiding the session-boundary game. Per-session vendors define a session by inactivity timeout (typically 30 minutes); a 4-hour visit with three 30-minute inactivity gaps becomes 4 billable sessions. Per-pageview billing avoids this boundary edge case.
When per-pageview pricing loses
- Single-page applications (SPAs). Modern React/Vue/Angular SPAs may register only one “page view” per actual user session, with all subsequent navigation happening as in-app route changes. Per-pageview billing would dramatically under-count actual user activity unless the SDK also counts SPA route changes as pageviews.
- Spiky-traffic content sites. A blog post that goes viral can generate millions of pageviews from short-bounce visitors; per-pageview billing punishes this more than per-session billing does.
Where to go next
- For per-session pricing structure, see per-session pricing models.
- For per-GB ingest pricing structure, see per-GB ingest pricing models.
- For vendor-specific pricing detail, see Atatus pricing and SpeedCurve pricing.
Last verified September 2026