In this post, I will be talking about measuring search quality through a metric called Search Exit Rate. You may have observed that a lot of such metrics may not be available in a single place and that it takes some effort on part of marketing managers or product managers to get all the relevant metrics on a single dashboard. Since we make it our job to be everything for search- and merchandising-oriented marketing on e-commerce sites, you’ll find the Unbxd Search dashboard a rare instance where you can track all these metrics in a single place. Let us start by first defining search exits. Search exits or rather the more formal term %age Search Exits, is that share of search sessions where a visitor closed the browser window or navigated away from your website after being presented the search results. SER can be tracked either as a single data point every day or as a curve over time to judge the success of the search results page in retaining and converting visitors.
Suggested Read: – How to Ensure Relevant Results for Long Tail Ecommerce Search QueriesHigher the Search Exit Rate(SER), higher are the chances that the search results page does not appeal to visitors. They came to the site expecting a certain set of products in the search results or certain prices for the presented results but exit soon after being shown the results. Most important aspect here is that the metric tracks exits from the search results “page”, and is not necessarily a remark on the search results themselves. If the SER is high, here are some things that you should investigate further:
- Search results relevance Ensure that overall the results being presented to the visitor are relevant for his or her search query. To dig deeper, you may wish to track SER for individual search queries and determine whether results for those specific queries are relevant.
- Layout of search results, especially whether there are enough products from the results above the fold. There may be a problem with the layout of the search results page. Especially if large part of your traffic is from one or two browsers then you will observe that a sub-optimal layout of search results page will increase the SER. One instance where this can happen is when the results are not displayed above the fold for a particular browser or OS.
Suggested Read: – How to Measure Effectiveness of Your E-commerce Search
- Experiment with product prices on the search results page If the results page displays product prices, it is possible that the visitors are leaving the web page after looking at the prices while comparing with other sites. You may want to investigate competitor prices and re-evaluate your prices for the products with the highest SER.
- Chart a trend line of SER for the last month and determine if recent changes related to search results page pushed the SER higher.
- Experiment with product variants on the search results page The products in your catalog may contain many variants. At the same time, a visitor’s perception of which variant is most relevant for their search query is different from the results your site search is displaying. It will be worth investigating which variants are actually resulting in lower SERs.
- Measure average time taken to render the search results page If the search results page takes too long to render on visitor’s browser, he may simply be closing the browser window or moving on to a different page. This might be contributing to high SERs.