Tomorrow I'll be speaking at SES Chicago on the topic of 'Showing Your Search Wins with Web Analytics'. When building out my presentation, I began to notice a trend: some of the best search KPIs to track these days are completely new, or otherwise significantly different from even one year ago. Gotta love this fast-moving industry.
Let's take a look at some of the KPIs I plan on discussing tomorrow.
Dwell TimeOne year ago, dwell time was just an internal metric for Bing (and possibly Google). Since the Panda update, it is arguably one of the most important KPIs to track for Panda-lized sites. What is dwell time?
The time between when a user clicks on our search result and when they come back from your website tells a potential story … while that’s not the only factor we review when helping to determine quality, it’s a signal we watch.Duane Forrester, Sr. Product Manager at Bing
Wait. Isn't this just bounce rate? Nope! Take these two scenarios:
- User X queries Google, clicks the first result, within seconds realizes that this result doesn't satisfy, hits back on their browser, then clicks the 2nd result.
- User Z queries Google, clicks the first result, spends 8 minutes reading the content on this page, hits back on their browser, then clicks the 2nd result.
How To Track Dwell TimeBecause Google Analytics and most other web analytics packages track users' interaction between pages on our site, they can't help us track dwell time. Remember, User X and Z above are both 'bounces', so we're not getting any time on site data from them either.
While there aren't (as far as I know) any tools for tracking dwell time specifically, there are some tools that can provide dwell time as a side effect of their core purpose. Specifically, eye tracking analytics tools like ClickTale or Crazy Egg can provide dwell time data, as they're both tracking how the user interacts with each page.
So if your site has been panda-lized, and you're working to improve user experience with bounce rate as a KPI, you might consider adding dwell time to your reporting dashboard.
Number of Unique Referring KeywordsThis is a sad evolution, unfortunately. Reporting the number of unique referring keywords as a KPI is a nice way to gauge a site's share of voice in the SERPs. A consistently increasing number would indicate that your site is being served and clicked on for more queries, and thus growing more trusted by the search engines and users.
Unfortunately, the advent of "(not provided)" muddies this number, making it a much less reliable metric to use as a KPI. To explain:
Prior to (not provided), we could easily get our count of unique referring keywords from the keyword reports in Google Analytics.
(not provided) counts as 1 unique keyword, regardless of how many unique keywords fell into that bucket during this time period. Sad Face.
Social InteractionsWhile Google Analytics enabled social interaction tracking back in June, and we could always hack together a solution, it still has been a bit of a PITA to get some good social interaction data out of GA. Fortunately, AddThis, the all-in-one social buttons plugin, has made passing our social interactions into GA really simple.
With the addition of a small JS snippet just above our AddThis snippet, we can get all social interactions passed into GA in an easily digestible manner.
Community managers and other socially-minded web marketers have never been able to track social interactions so easily and reliably.
SES ChicagoThom Craver and I will be speaking on "How to Show Your Wins at Search Marketing with Web Analytics" at SES Chicago on Wednesday, November 16 at 2:30. If you're attending, make sure to say hello! Once the session is over, I'll embed my slides here.
And here we are: