Why this metric lost relevance
Even before Alexa's shutdown, Page Views Per Million Users had significant limitations that made it unreliable for business decisions.
Sample bias: The metric only captured behaviour from users who had installed the Alexa toolbar, a self-selected group that did not represent the broader internet population.
Bot inflation: Automated traffic and bots easily skewed raw page-view counts, making per-million ratios unreliable.
Single-page app blindness: Modern web design, including single-page applications, infinite scroll, and mobile browsing, means users often find what they need without triggering multiple page loads. Raw page-view volume understates real engagement.
No intent signal: High page views do not indicate that a user is satisfied, engaged, or converting. They can just as easily indicate a lost user who cannot find the right content.
What to track instead
Modern web analytics focuses on engagement quality and business outcomes rather than raw volume. These metrics offer more actionable signals.
| Metric | What it measures | Why it matters |
|---|---|---|
| Page Views | Total pages loaded across all sessions | Baseline traffic volume; best used alongside session depth |
| Engaged Sessions | Sessions lasting 10+ seconds, with a conversion, or with multiple interactions | Filters out low-quality traffic |
| Conversion Rate | Percentage of sessions that result in a target action | Directly ties traffic to business value |
| Bounce Rate | Percentage of single-page sessions with no interaction | Signals content relevance and landing page quality |
| Average Session Duration | Mean time users spend per session | Proxy for content depth and engagement |
Google Analytics 4 (GA4) and other modern analytics platforms provide all of these metrics with far greater accuracy and segmentation than Alexa's toolbar-based sampling ever could.
Should you reference this metric?
No. Page Views Per Million Users has no current application. Alexa's data collection ended in 2022, and any historical Alexa data should be treated as non-comparable to current web analytics outputs due to differences in methodology, sample composition, and tracking technology.
If you are auditing legacy reports or benchmarks that reference this metric, replace them with engagement and conversion metrics sourced from GA4 or a comparable first-party analytics platform.