Dynamic audiences are an advanced segmentation feature in Adobe Experience Platform which solves the scalability problems marketers traditionally face when building audiences for marketing campaigns. For more information, please see the Segment Builder documentation.
“Segments” have been renamed “Audiences” in the Platform interface. Select Create audience > Build rule to enter the rule builder featured in this video.
In this video we’ll show you how to build dynamic segments, an advanced segmentation feature availible now in Adobe Experience Platform. We’ll show you how to use dynamic segments to solve the scalability problems marketers traditionally face when building segments for marketing campaigns.
Many businesses struggle to wrangle the sheer number of segments it required to execute on seemingly simple campaigns. Say I’m a Luma marketer looking for people who made purchases from Luma stores outside their home state. A very common way of getting here would be to create one segment for every US state. And while 50 segments seems like a manageable number, this probably becomes exponentially more challenging when dealing with hundreds or even thousands of possible combinations. So why the explosion of segments for such seemingly simple business goals? Well, for many segmentation tools on the market today the trouble lies in the static nature of these segments. You see, traditionally, I’d build them by stating explicitly that I’m looking for people in a certain state, say Utah. And also for a purchase event where the state or province of the purchase is not Utah.
And then I’d do the same for the other 49 states.
In Platform, we can consolidate these 50 segments into a single dynamic segment, to eliminate all of the repetitive work. To do this, I can actually do away with this first rule all together. And then, instead of calling out a static value in the purchase event, I can drag the state field from the attribute section and drop it in the event rule. Now I have a rule that says find me a purchase where the state of that purchase is not equal to the customer’s home state. So, instead of choosing a specific state, I’m dynamically expressing the relationship between the state where the purchase occurred, and the attribute of the customer’s home state. And saving myself from building the other 49 segments.
I can also build dynamic segments which compare data from events. To illustrate this, let’s take an example of people who purchased the same product twice, but via different channels. With a reasonably sized product catalog and just three channels, the number of static segment definitions we would need for this is somewhere between 30 and 60,000, but now we can build a single dynamic segment to capture all of these. So let’s start off by adding the two purchases that we care about, you’ll notice as I bring in these events, they appear as variables, along with my other building blocks. Now, let’s add a couple of rules to a second event. First, to get the same item in both events, we’ll use SKU from the product list items.
Now, we’ll drill into that purchase one variable. We’ll get the SKU from it, and drop it on the right side here, creating a rule that says that the SKU for purchase two, is equal to the SKU from purchase one. We’ll do the same thing for channel. We’ll bring in the channel ID from the XDM schema, then get the same field from the purchase one variable. Then in this case, we want the channels to be different. And just like that, we have a single dynamic segment definition which accounts for the 30 to 60,000 individual segments we’d otherwise need to create.
Let’s take it one step further now, and move from hard to scale segments, to some which are maddeningly difficult to build outside of an analysis or BI tool. Let’s look at the example here of people who made a purchase and then spent more on a subsequent purchase than their initial one. We’ll just modify our existing segment definition, to dynamically compare the prices between the two purchase events, again using these event variables.
The segment builder makes this so easy it’s hard to believe that a segment like this is nearly impossible in other tools.
So we’ve walked you through a few examples of how you can use dynamic segments to do things that were very difficult, or impossible with traditional segment builders. We’re excited to see the kinds of deeply personalized experiences you’ll be able to deliver, using this breakthrough capability on the Adobe Experience Platform. Thanks for watching.