Approximate Count Distinct function in calculated metrics approximate-count-distinct-function-in-calculated-metrics

Learn how to create a calculated metric using the Approximate Count Distinct function, which returns the approximated distinct count of dimension items for a selected dimension.

Transcript
Hi, this is Jen Lasser with Adobe Analytics product management. In this video I’m going to introduce a new function that we’ve added to the Calculated Metric Builder called Approximate Count Distinct. This is a function that our customers have asked for for quite some time. So we’re really excited to finally be able to bring into the Calculator Metric Builder.
Approximate Count Distinct is available from the left rail and it takes a dimension as its input.
What Approximate Count Distinct will do is count up the unique dimension items collected within the dimension that you’ve specified for the reporting window. So the biggest use case that we’ve heard of from our customers for wanting this function is to count the number of unique customers that are interacting with their brand. So for example if you have Customer ID collected in an eVar, you’ll now be able to count the number of unique Customer IDs collected and use that in various reports.
We call this Approximate Count Distinct because it is an approximation. It’s not an exact count of your Customer IDs. We use something called HyperLogLog Methodology which guarantees that this number will be accurate within 5 percent of the exact number, ninety five percent of the time. So it is an approximation but it is the most scalable option that we offer for counting a dimension and being able to use that in other reports as well.
You may be wondering how this compares to some of the existing count functions. So we offer Count and Row Count. Those are exact count functions, but they are only usable and local to the the dimension that you’re viewing, whereas Approximate Count Distinct is more scalable. You can count a dimension and then use that metric in any other dimensional report of your choosing.
So let me save this off and we’ll go into Workspace and use this as an example. So I’m going to call this “Approximate Customers” and I’m gonna go ahead and click save.
Back in Analysis Workspace here. I have a report for my Marketing Channels and I’ve pulled in Unique Visitors as well. Rather than looking at Unique Visitors which is essentially unique devices, I can now pull in my Approximate Customers calculated metric to see a very duplicated view of the actual customers that have interacted with each of my marketing channels across my my brand’s properties.
So this is just one example of a metric that you can create in a way that you can apply this metric within reporting. But there’s there’s lots of other use cases for this function and we hope you guys enjoy it as much as we did bringing it to the builder and hope you get a lot of use out of it.

You can use the metric shown in this video in any report to understand the count of one dimension against values of other. A very common use case would be to use this function to create a Customer IDs metric, which you could then use in any report to see how many customers apply to the different dimension values.

recommendation-more-help
b5d9c99f-be9f-4b96-8809-4e7d6ae353ba