An example analysis of Groupon's traffic showing a signficant proportion of direct traffic could be SEO-related
Gaining organic traffic from SEO is a significant inbound marketing investment for many businesses, so making the business case and demonstrating returns from SEO is important. The typical approach for this is naturally to use Google Analytics to report on the organic traffic segment. Since Google Analytics can no longer reliably provide SEO keyword data due to the growth in keywords marked 'not provided' we have written about how to use Google Webmaster Tools keyword data to determine keywords.
We're alerting readers to this example, since it shows that you should make colleagues (or clients) aware that significant additional SEO traffic may be hidden as a referrer.
Case shows that 60% of direct traffic is actually organic search
This research summary of Groupon visitor referrers from Conductor summarises the importance of SEO to Groupon based on their standard analytics. They comment:
'Groupon recently published a study showing 60% of their ‘direct’ traffic is actually organic search.
The original study showed that 47% of visits were organic search; the updated study suggests that organic search traffic might be responsible for closer to 64% of website traffic.'
This example is unusual since you wouldn't expect a large brand like Groupon to disclose their natural search. We're fortunate that, in this case, Gene McKenna, Director of Product Management at Groupon where he leads organic search was open to sharing and discussing his test. You may have seen his original post explaining the test, also unusual in that, as Gene puts it "for the sake of SEO science, we deindexed ourselves completely for about 6 hours". This was for part of the site focusing on long URLs they were investigating.
You can see that during this time, as expected organic traffic fell, but direct visits also fell significantly.
The reason for the decline in direct isn't clear, but discussion in the post suggests it's likely mainly due to browsers not sending information about the search referrer through to site or redirects client or server side. Gene concludes:
'Our testing shows that, for a site getting in the ballpark of 50% mobile web traffic, the 60% of the traffic to long URLs reported as Direct is probably Organic traffic from Google'.