Idea Exchange goodness: two new metrics in SiteCatalyst
Shortly after we launched the Idea Exchange in early 2010, a submission floated to the top and has stayed there ever since.
|This is Carmen, in case you couldn’t tell.|
Carmen Sutter asked—nay, demanded—“Make Bounce Rate a default metric!” Our customers agreed en masse, with 238 people voting for the idea. We took a step toward responding when we released SiteCatalyst 15 by removing the need to build calculated metric after calculated metric to examine bounce rate in Pages reports, but we knew we weren’t done.
The next time you log into the Idea Exchange and click “Top Ideas,” you will see a big fat “Implemented” under Carmen’s idea. Between adding Bounce Rate to most SiteCatalyst 15 reports (including eVars) and the new Total Time Spent metric with all that it enables you to do, today’s SiteCatalyst release is definitely something to be excited about. Here’s a brief rundown of what we’ve done.
Bounce Rate has been one of the most important metrics in digital analytics for years, and (admittedly) we haven’t always done a great job unearthing it for you outside of the Pages report. With this latest release, we’ve taken another step toward making Bounce Rate available throughout SiteCatalyst 15. For the first time, you will see Bounce Rate in conversion reports (including Campaigns and Products) and in Traffic Sources reports. This enables analytics around things like Product Bounce Rate, Campaign Bounce Rate, and Search Keyword Bounce Rate. Additionally, Bounce Rate is available in subrelations, so you can go another level deep into your Bounce Rate to answer questions like, “Which product pages in Campaign X have the highest/lowest Bounce Rate?” or “Which products are least interesting to my return visitors?” enabling you to focus on what is working and fix what isn’t.
You should see Bounce Rate as a standard metric in every report where it is now available. Just add it as you would any other metric. We’ve also added Bounces (i.e. single-page visits), Entries, and Exits to these reports as well.
Note that a Bounce requires that the given variable occurred on the first (and only) request of the visit. So, for example, if a user views a single page where eVar1 has a value of “men’s shoes” and then leaves your site, that would record a bounce for the eVar1 value “men’s shoes.” But if the user views three pages, and on the third page the value “apple ipad LTE” is passed into eVar1, even if that is the only eVar1 value during the visit, it is not a bounce.
A few more “things to know” about Bounce Rate in SiteCatalyst 15:
- Bounce Rate does honor eVar (and campaign) persistence. Let’s say you have an eVar that persists beyond a single visit (e.g. one month). I visit your site on June 4 and the eVar in question is populated with a value of “ABC” on the third and final page view of my visit. (By the way, that’s not a bounce, because the value occurred on a multi-page visit.) Then, on June 10, I come back to the site and leave immediately after one page view; the eVar does not receive another value. That second visit would be a bounce associated with “ABC” because this value had persisted from the previous visit and had not been overwritten by a new value for this eVar. This means that campaigns and eVars that have long persistence periods will likely accrue more bounces than similar variables that expire at the end of the visit. However, it does not necessarily mean that they will have a higher Bounce Rate, because non-bounce visits are also associated with the persisted eVar value, potentially raising the denominator at a higher rate than the numerator in the Bounce Rate formula.
- Bounce Rate in SiteCatalyst 15 looks at what we could call “single-beacon visits” in its calculation. A user interaction, even if it does not lead to a new page view, can thwart a bounce. This means that if a landing page sends multiple requests into SiteCatalyst—either multiple page views, or a page view and a custom link, or a page view and a video view—it will not be a bounce. Based on feedback we have obtained from our customers, this definition is how most of you view your site/app; interaction means “no bounce.” For those of you who disagree with this definition, we are looking into the best way to give you a more suitable metric, and you can hide Bounce Rate from your users in the User Management > Groups area by restricting access to this metric at a group level.
- Can you segment reports showing Bounce Rate? Of course you can! There are all kinds of questions I can’t wait to answer with this new functionality. Are there landing pages that are stickier for first-time visitors than they are for repeat visitors? Do tablet users bounce from my site more frequently than desktop users? I could keep going, but you get the idea. Segmentation + Bounce Rate = seriously impactful analysis.
Total Time Spent
Another new metric that you’ll see in SiteCatalyst 15 beginning on May 31 is Total Time Spent. This metric represents the total amount of time, in seconds, attributable to values in your Custom Traffic (prop), Custom Conversion (eVar) reports, and Traffic Sources reports. And since you can use it in calculated metrics to create things like Average Time Spent per Visitor and Average Time Spent per Video Start, it allows you to understand engagement much better than ever before in SiteCatalyst. (Gone are the days of trying to squeeze everything into Custom Traffic reports so that you can multiple Time Spent per Visit by something and then divide it by something else.)
I’m really excited about this one. Not only does it enable all sorts of engagement-related answers to questions such as “How much longer do visitors who purchase spend on my site than users who don’t purchase?” and “Which of my campaigns are stickiest?” and “How does Time Spent vary for visitors who fit different segments/profiles?” but when coupled with new video reporting in SiteCatalyst 15 it makes a number of consumption metrics much easier to work with. In my conversations with publishers who stream live events or TV programming to their sites or apps, it is clear that metrics like Average Minute Audience are key to the future of digital in their world, and Total Time Spent makes it much easier to obtain, segment, and analyze on metrics like Average Minute Audience and others. I think you’ll find that Total Time Spent and the calculated metrics you might create with it are wonderfully applicable. It truly moves beyond the old Page Views era to one where you can understand your customers’ journey with new clarity. I, for one, have really enjoyed playing with it as I analyze my own site visitors.
A couple of things to keep in mind as you dig in to Total Time Spent:
- Right now, you’ll need to build your own calculated metrics involving Total Time Spent. Start with Time Spent per Visit ([Total Time Spent] / [Visits]) and Time Spent per Visitor ([Total Time Spent] / [Unique Visitors]). Make sure to choose the “hh:mm:ss” calculated metric to get a really easy-to-read-and-interpret output that you can share with colleagues.
- As with Bounce Rate, when you use Total Time Spent in conversion reports, it does respect your expiration and allocation settings. If you have an eVar set to “original value” (i.e., first-touch allocation), Total Time Spent will continue to accrue to that original value as long as it persists, even if that eVar receives other values subsequently prior to expiration.
- In reports where the “Average Time Spent” metric already existed, it will continue to exist, and you probably don’t need to create the Average Time Spent per Visit metric in those reports.
One more thing: “Uniques Exceeded” renamed
Last month, Matt Freestone wrote an excellent post detailing a change to the way we handle high cardinality in SiteCatalyst. By way of update, note that we have changed the label “Uniques Exceeded” to “(low traffic)” in your reports. Remember, you should only see this label for very low traffic values (hence the name) and never among the top items in unfiltered, unsegmented reports, as had been the case previously.
It’s a lot of fun to be in the office late on release nights. We get some food, fire up some music, and get stuff done. But the best part of the night is when we get to mark your Idea Exchange submissions as implemented. We’re up to 161 total ideas implemented from the Idea Exchange, but we’re just getting started. I have to tell you, Carmen, it feels REALLY good to mark the overall number-one idea as done. Be sure to let us know what you think. You can find me on Twitter at @benjamingaines or leave me a comment here on the blog!