Guest Comment: Using web behaviour data to drive follow up marketing


Jun 19, 2008

Working out why some email ads fail can be crucial to improving future campaigns. Andrew Robinson, managing director at Lyris UK, looks at how web analytics can help marketers get the most out of their communications.

There is a lot to be learnt beyond open and click-through rates on email marketing campaigns. By employing a web analytics tool to analyse the behaviour of respondents, valuable insight can be gained into just how effective an email campaign is – or isn’t.

Many marketers consider ‘conversion’ the holy grail of campaign success. An email generating a lead or a sale is considered successful; one that doesn’t is the opposite. But a lack of conversion doesn’t necessarily constitute a failure. By examining two key metrics – visitors’ average time on site (ATOS) and campaign exit rate – it is possible to understand where the conversion process needs improvement, and use those valuable insights in follow up campaigns.

Focused analysis

Segmentation is the key to gaining actionable insight on any marketing activities, including email campaigns. Marketers need to be able to focus on visitors who share common behaviour when responding to a particular campaign. Using a web analytics program to segment campaign respondents will provide valuable information in formulating follow up campaigns.

Are website visitors interested…

Time on site measures a web site’s ability to successfully maintain each visitor’s interest. It’s also a good indicator of how well visitors’ expectations are being met – what they thought they were getting when they clicked through, and what they found once they arrived. The average time spent on the site by the users that came in via an email campaign can be analysed and compared with the benchmark value of the average time on site for ‘all visitors’.

We all like to be sure we’re getting the most for our money – and on the web it only takes a few clicks to move from store to store or to a review site to validate (or decide to rethink) our choice. In general, higher than average time on site is a pretty good indicator of interest. However, it can mean that respondents are struggling to do what they want on the website – or what was promised in an email.

It’s also important to remember that a short time on site and high exit rate on the right page, such as basket/purchase confirmation, can be a good sign. But if large groups of visitors have a high ATOS but few conversions, email marketers need to examine their campaign offers and build a reason for quick action into the follow up campaign. Lingering window-shoppers need to be given a catalyst to convert.

…and are their expectations met?

The campaign exit rate is another effective way to gauge whether or not visitors’ expectations are being met. For example, an email message offering free shipping must be matched by a landing page that clearly restates the offer that got the recipient to click through in the first place.

The first action of online marketers dealing with high exit rates needs to be to see how well the the landing page reflects the promise made in the email campaign. Using the same (or at least similar) graphics and promotion wording is a good start.

However, the time on site must also be correlated with the number of pre-defined goals (such as making a purchase, submitting enquiry forms or signing up to email) completed. A long time on site is a good sign if a lot of these goals are achieved, but indicates a need for improvement if few or no goal activities are undertaken.

Triggering action

Having put time and effort into analysing users’ web behaviour, it is important to use this information to enhance the online marketing programme. One particularly successful tactic is to implement a triggered email function. This uses web analytics to track people who spend a lot of time on a particular section of a website as a result of receiving an email, but then don’t reach a goal. A follow-up email 24 hours later reminds them of what they were looking at with the aim of encouraging them back to the store to make a purchase. For example, someone who doesn’t buy anything despite having spent 30 minutes browsing on shoes could be sent a follow up message the next day with a reminder such as: “Did you know you can return up to four pairs of shoes in one order if they don’t fit you?”.

Final judgment

It is also important to note that no one metric should be looked at in a vacuum. Instead, a holistic view of metrics – including ROI, conversion, and visitor volume/response rate – should be adopted to ensure an accurate picture of the value of each campaign.

By combining the email provider’s metrics with a few key metrics from web analysis software, the experience for customers can be improved, both in their inboxes and on the website.

By Andrew Robinson

Managing Director, Lyris UK