Written by Sophie Teh.
If you are a marketer familiar with Adwords or any Demand-side Platform (DSP) campaigns, chances are you would have come across discrepancies between the number of clicks and the sessions recorded on Google Analytics for the exact same ads.
But… isn’t this an old debate settled a long time ago? You are right. I’m bringing this subject matter up for discussion at this moment for 2 reasons. Firstly, because some of our business partners mention the discrepancy on occasion. And secondly, because there is no specific discussion on the phenomenon for metasearch engine traffic.
For some time now, it is accepted that the number of clicks and sessions do not add up for a good reason, because they are fundamentally very different metrics. Furthermore, marketers also know that:
- The difference between the numbers is influenced by many factors, including type of website
- There is no universal benchmark for what a “good” ratio for clicks to sessions should be; i.e. traffic that has a 10% discrepancy doesn’t necessarily mean it is better than traffic with a 70% discrepancy
- There are more suitable metrics for measuring quality of traffic, such as Conversion Rate (CR) and Cost per Lead (CPL), which are more direct indicators of return on investment (ROI) on traffic acquisition
Since sessions will continue to be prominent on the Google Analytics dashboard, I’d like to use this opportunity to talk about how this discrepancy relates to traffic from metasearch engines to show you a new way to look at it.
Discrepancies can indicate good user intent
Metasearch engines, commonly known as aggregators, gather information from multiple sources in one place for users to search for. Skyscanner aggregates flights from hundreds of airlines for search on one platform. Booking.com aggregates hotel stays from millions of hotels… In our case, Trovit aggregates listings from hundreds of websites in 3 categories – real estate, cars and jobs. Our job is to help users find the most relevant ads in one platform. And, by doing that, bring qualified traffic to our clients, i.e, search portals. By qualified traffic, we mean users who have serious intent and have a high probability to convert into a lead.
“Users who search more actively in one session are higher quality users with clear intent and are closer to the end of the conversion funnel.”
Consider the following 2 scenarios. Let’s use a real estate search platform like ours as an example:
User A is looking for a home to buy in a certain area. She spends her lunch hour searching and viewing home advertisements aggregated from multiple portals on Trovit. Over a continuous 30-minute period, she clicks and views 10 properties on her phone and contacts 2 of the realtors. So happens 5 of the 10 properties viewed were from portal ABC.com. The 2 realtors contacted were also through ABC.com. On ABC.com´s Google Analytics dashboard, there would be a record of only 1 session from User A, although User A clicked on 5 advertisements within that session. The discrepancy between clicks and sessions in this case is 80%.
User B is also looking for a home to buy in the same area. He views 2 properties over 5 minutes and does not contact any realtors. Say 1 of the 2 properties clicked on was also from ABC.com. On ABC.com’s dashboard, 1 session was recorded with this 1 click. The discrepancy in this case between clicks and sessions is 0.
In this example, for a real estate portal looking to monetize leads like ABC.com, User A was more valuable than User B. She trigger 2 conversion events. But even if she did not, her user behaviour meant that she was further along the conversion funnel that User B. In retail marketing, intense search behaviour shows that the user is very likely to purchase in the next days. Similarly, users who search more actively in one session are higher quality users with clear intent and are closer to the end of the conversion funnel.
But on Google Analytics’ dashboard, User A’s behaviour generates a higher discrepancy between sessions and clicks compared to User B. Some marketers believe that discrepancies mean missed opportunities but, as you can see, the opposite is true in this case. High discrepancy points to a higher proportion of qualified visits.
Form Follows Function
As mentioned earlier on, the discrepancy between clicks and sessions depends highly on the type of website. Type of website is driven by type of business. For a meta search engines like ourselves, the very nature of our value proposition dictates that this discrepancy for the traffic we send to a destination site is going to be high compared to other non-search engine referral sites.
Let me explain what I mean.
When traffic comes from a referring link via a news site, blog, or email marketing, it is unlikely that the user will return to the destination site again from the referring site.
On the contrary, search engines have the capability to generate new search results and more links to destination sites for a user. Over the course of each search session, a search engine can send the same user to a destination site multiple times if the destination site has multiple search results matching the user’s query. Therefore the more effective we meet the needs of our users, the more efficient we are at channeling traffic to our clients in a short period of time. And the higher the discrepancy between clicks and sessions. Furthermore, at the same time, we are also channeling quality traffic of users who are further along the conversion funnel to destination sites.
There is no one-size-fits-all solution in performance marketing for every business. There are possibly a hundred different metrics and indicators in the world of digital marketing to look into, but not all of them are going to be relevant to everyone. As marketing guru Neil Patel said, “Numbers alone don’t tell the whole story of who’s visiting your website… You must understand the user’s intent and act on that information. Obviously, intent is a lot more “fuzzy” and a lot less quantifiable than the crisp, exact numbers that you see in Google Analytics.” To look behind and also beyond some numbers would allow us to understand our users more and get more results from our marketing efforts.