Direct Traffic in Google Analytics

Sudden drop?

[jbnz]

On December 11th last year, I had a sudden drop in my Direct Traffic according to GA from 178 (Dec 10th) to 45 (Dec 11th). It has remained at this lower level ever since.

This surprises me as I would have thought that generally Direct Traffic would not vary much from day to day.

Can anyone think why Direct Traffic could drop so drastically? I'm assuming it's stuff like type-ins and bookmarks?

Just in case anyone's thinking those are big enough numbers to be able to say there was a drop, the Direct Traffic for November was 4892 and January was 1808.

Also the only changes to the site on that date, were creation of a new post (it's a WP site) and I made a different post the Sticky post on the home page. I just can't see how these changes would have affected Direct Traffic, Search traffic may be, but not Direct Traffic?

[Andy Langton]

Drill down into the traffic - where did it land? Any specific user-agents? What's the difference in bounce rates, pages viewed or time spent?

Direct traffic is a catch all for when analytics systems don't know the referrer - this can include everything from type-ins and bookmarks to issues with your analytics installation.

[jbnz]

Thanks Andy, yep Direct Traffic from Edge dropped from 2562 in November to 69 in January.

And it appears it now gets reported in Organic Search, having dug deeper. The site is about a piece of Windows software so I wonder if that's got something to do with it as I don't see the same shift in stats for another site I own.

Anyway, mystery does seem to be solved, just seems weird to move traffic from Direct to Search, may be something changed in GA on how it detects keyword data from Edge? But then why not a change in all sites?

I guess I shouldn't worry too much.

[Andy Langton]

I guess I shouldn't worry too much.

The implication is that Edge was previously tracked incorrectly. So that would either be an external event (e.g. they fiddled with Do Not Track or something similar) or something on your own site (landing page redirects? IE conditional stuff?). Again, you'll get closer to the answer by drilling into the data (e.g. which search engine, any keywords, which landing pages). Whether it matters depends on whether it would be a problem if something like this recurred.