One of the great features you’ll find in many Customer Data Platforms (CDPs) is the ability to trigger an event on an app or website based on a user’s actions or status. So, for example, you can display an offer to buy your stock market newsletter to everyone who uses your stock finder more than some number of times.
Here’s how that works.
The CDP has code on your website that tracks visitor behavior. When a new visitor arrives, the CDP drops a cookie on that visitor’s browser and creates a user profile for that visitor in the CDP database. At this point, the visitor is known only by a unique ID in that database, and whatever fingerprint the browser leaves.
As the visitor browses your site, listeners in the CDP collect information about that visitor’s behavior and the CDP writes that into the visitor’s profile. (For more on what data a CDP might collect, see How to use a Customer Data Platform to make your site more useful and profitable.)
This is similar to what Google Analytics does with visitors, except that GA keeps the user anonymous. The purpose of a CDP is to identify that user — to move the visitor from unknown to known by collecting information that identifies the individual user.
As the visitor’s profile fills with useful information, that visitor might become enrolled in specific segments or groups. While the CDP will have some built-in segments, most of them are defined by the website owner based on their business requirements. For example, if the website owner is interested in making a particular offer to everyone who visits at least five pages on investing, they’ll first create a segment of all those visitors, then create an activation of some sort on the website — a message or an ad, for example.
This is where it gets tricky. What if Marketer A creates an activation for visitors who have viewed five investing pages, while Marketer B creates an activation for visitors who have viewed a total of ten pages? There will be some overlap between those two segments — that is, for some visitors, their 10th overall page view will be their 5th investing page view. The result is that the two activations might get in the way of each other.
That might not sound so bad, but as these rules and activations multiply, you can end up bothering the heck out of your visitors.
There are a few ways to make this less of a potential problem. There could be rules to limit the number of activations on a page, for example, or the activations could be given a priority, like low, medium, or high. That would allow the CDP to avoid conflicting activations on the same page.
That’s helpful, but what I would like to see is a collision predictor, so that when Marketer B creates a campaign, the CDP makes an estimate of how many conflicts there might be. For example, before Marketer B finalizes a proposed campaign, the CDP warns: “10 percent of the activations of your campaign are likely to coincide with the ‘buy the stock market newsletter’ campaign.” At that point, Marketer B should be able to make adjustments, such as excluding the members of the conflicting segment from the campaign.
In addition to that, there should be a bubble chart that shows likely overlaps between all activations, with a way to set rules for managing them. This could give an overall view of how and whether activations are stepping on each other.
Thoughts or comments? Please post them below.
If you have questions about how to manage your Customer Data Platform, please contact me. I’m happy to chat.
