I want to use the time on page as a quality metric for a set of URLs. If the difference between both is like:

Avg Time on Page = Time on Page / ( Pageviews - Exits)

Which metric is more suitable to serve as the quality metric?

  • Why not try checking the time-on-page(individual page) and time-on-site(time-on-page overall for given session) directly https://saleemkce.github.io/timeonsite with high accuracy that gives a good picture about content significance of each URL?
    – webblover
    Dec 22 '21 at 7:07

Well, first, your formula doesn't make much sense cuz you seem to be mixing an aggregated scope with non-aggregated. You probably meant total time on page divided by pageviews.

It's a common misconception that time spent on page is a positive indicator of the quality of the site/content. It heavily depends on the type of the site. For example, this metric could be interpreted as a positive (more=better) on a gaming site, or, arguably, on a news/article site. But it would be interpreted as a negative (more=worse) on, say, e-commerce sites where time spent on page would rather indicate confusing interface, errors and slow loading of elements than users being charmed by the content.

Having said that, time spent on page is normally treated as a vanity metric , a debugging metric or an SEO metric (which is the same as vanity at this point).

Now to actually measure quality, you first have to decide quality of what you're measuring: traffic, conversion funnel, articles, support, etc. In most cases, the quality is measured by either partial or complete conversions. And the definition of a conversion in each case depends on the subject of the quality to be measured.

  • It goes about a page type, well matching by time on page as a quality factor - there are knowledge sharing articles, where the scroll depth measured by events acts like additional quality factor.
    – Evgeniy
    Nov 26 '21 at 22:30
  • Seems to me you're looking for wrong things. Scroll depth on knowledge articles would likely indicate low quality of the articles. What you want for an ideal knowledge sharing article is no scroll at all, very high bounce rate and main source - search. That would indicate that people have no issues searching for solutions, they find the solutions quickly from the very first page they visit and it's strict to the point, so they don't need to continue navigating through the site.
    – BNazaruk
    Nov 26 '21 at 22:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.