Time-on-Site: Understanding the sweet spot for conversions

Time-on-Site: Understanding the sweet spot for conversions

Time-on-Site: Understanding the sweet spot for conversions
From Anthro136k Spring2011/Flickr
I’ve written before about time-on-site but in that post we focused on how site search in particular and generally revenue is influenced by time-on-site. This post will focus more on the metric itself rather than external factors that influence it. Time-on-site-after-search is the duration of time a visitor is on the site after seeing the search results page. It is normally distributed like the curve in the chart below. Time-on-Site: Understanding the sweet spot for conversions If the search results are of poor quality, then the time spent on the site will be very short. For eg., in the above chart, in the first 0 to 8 mins, low conversions could be due to irrelevant results. Such a low time-on-site will also result in high rates of exits. The next segment, between 8-12 mins, will be the segment where most conversions take place. This is the conversion sweet spot. At this point, it is certain that search results are relevant for the visitor. They will now look at other parameters of the results such as price, availability, shipping time, etc. The average time-on-site for a majority of the visitors will be similar, resulting in the sweet spot at which point the maximum conversions take place. Most visitors want to quickly identify their product of interest and then make the purchase. Visitors who spend more time on the site, i.e. in the segment right after the sweet spot, are either:
  • window shopping, or;
  • persistent enough to continue looking for products even when they are not able to find relevant products through search.
For window-shoppers, the trick is in understanding their interests to increase the potential of purchase. In the case of poor search results, another set of techniques will have to be used to judge the effectiveness of search.

More To Explore

Learning To Rank with eCommerce data

Learning to rank with eCommerce data

What, Why and How of Learning to Rank: Learning to Rank (LETOR) is used in the information retrieval (IR) class of problems, as ranking related