Using your site search to harvest data

You know that search bar at the top of your website? Turns out it’s not just useful for your customers it’s a really useful feature for you too.

Depending what kind of company or industry you are in, the data collected by this search function may pull up all kinds of stories and angles that will generate a content idea that publishers will love.

You can find this information in Google analytics under ‘Behaviour’ and ‘Site search’

In a past campaign for a private number plate business, we analysed over 4 million site searches for personalised number plates. We found the following data that could be used as hooks for all kinds of publishers.

  • Explicit terms are very popular, even more so than ‘mum’ or ‘dad’
  • LFC is the most popular football team for a number plate related search
  • Site searches mirrored popular trends, Paul Pogba inspired ‘dabbing’ was on the rise whereas TOWIE’s ‘Reem’ was on the decline.
  • There was a big postcode war between the more affluent areas of London such as Notting Hill and Chelsea
  • People were actually searching for ‘Trump’ and ‘Brexit’ related number plates

You can guess at what angle the Mail Online went with….

Using internal site search to plug gaps in content

As above, your site search history will basically show what your users are interested in finding. This information can be used as a data piece of content but even more so it should be used to create content that satisfies these popular internal searches.

In Google analytics under the ‘search terms’ tab you will find the following:

  • Total unique searches
  • Results pageviews
  • %Search exits
  • %Search refinements
  • Time after search
  • Average search depth

You can use the above metrics to determine whether a popular and common search on your site is actually being satisfied or not. If the % search refinements is high they may not have found what they were looking for on the initial search. Likewise, if the time after search and average search depth are low figures it may suggest that that nothing really came up after the user’s search.

Take this information and start to weave in useful content that will show up when a search query is undertaken. For example, if you are receiving 2,000 searches a month for ‘Jobs in Papua New Guinea’ but there is limited content that is provided after a search. You can start thinking about useful content that plugs this gap in content such as ‘Everything you need to know about living and working in Papua New Guinea’

It’s about satisfying that need for certain content on your site. If you do this in volume you will see a more fruitful interaction with your site and you can expect to start increasing time on site, assisted conversions and reducing bounce rate. You can even take these learning’s into improving your sites user experience as a site search can often indicate the user was not able to find what they were looking for logically on the navigation.