As a DESelect Segment user, you will encounter times when you will question what approach/feature is best to use for a specific use case.
How do you choose between using a Selection Template instead of copying an existing Selection? Or should you opt for a Data Set or Master Data Extension instead?
The hesitation can happen because all these 4 features share a common principle: simplify audience generation, allowing marketers to reuse existing or recurrent structures.
In this article, we want to share some best practices to help you choose between these features when needed.
Selection Template | |
Use Case |
Example |
Campaign Type Template You want your Selection to be the base for many future Selections with a similar structure, but these future versions will need some small modifications. In this base Selection, you want to pre-set some advanced settings like Filters or Custom Values, not only the relationship between Data Extensions (since this would be possible to solve by using Data Sets instead) |
Campaign Type: get orders placed in a given country in the last 6 months Selection Template: All UK orders made by men in the last 6 months. Using this template for a new Selection: All US orders made by men in the last 6 months. In this last version, we will update only one Filter: Country Equals UK will become Country Equals US |
New runs of the same campaign You want your Selection to be a template for future versions but the original version to never be modified by anyone else, leaving a description and instructions on how the team can copy it and create future versions. Note: A Selection that has been marked as a template can only be modified by the user who created it or an administrator. This is to prevent a template from being modified by other users or to avoid deleting it by accident |
You have a campaign running monthly to give a prize to the top users. For each run, you want to generate a different Target Data Extension for archiving purposes.
|
Copy Selection |
|
Use Case |
Example |
You want to create a Selection with the same or similar structure of another Selection. This is a one/two-time copy or you need the same Filter but in another Business Unit |
You created a Selection for the first day of Black Friday, and now you want to reuse it for the second day, applying the same filter but updating the dates you are using in the Filter section. As it’s a punctual need, you can just copy the Selection of Day 1 of Black Friday, change what you need, and run it |
Note: If the structure of this Selection you are copying is useful for many future Selections, opting for a Selection Template would be more beneficial.
Data Sets & Filter Sets | |
Use Case | Example |
As a marketer, you always add the same combination of Data Extensions and filters in your Selections, and you find yourself dragging and dropping the same Data Extensions and/or filters several times a day. |
Your organization often runs campaigns based on customer orders and their order details. Additionally, there’s a set of standard filters you always need to apply: exclude customers without email, customers without email opt-in, orders older than 10 years ago, and orders of value less than 1 dollar.
In this scenario, you will create the following Data Set in the Admin Panel (a combination of 3 Data Extensions) Customers WITH MATCHING Orders Orders WITH MATCHING OrderItems
Also, you will create the following Filter Set “Opted in Customers with Valid Orders” (a combination of 4 filters mentioned above) Email (Customers) is not empty Email Opt-In (Customers) = True Amount (Orders) Smaller Than 1 Order Date (Orders) In Previous 10 years After you have this Data Set, instead of adding the Data Extensions one by one, you can just drag & drop the Data Set, and all the combinations of Data Extensions will be automatically added, and drag & drop the Filter Set, and all the filters will be automatically added. |
You have some campaign types that require building a complex Data Extension combination with a complex set of Filters. Creating these relationships and filters directly in your Selection can lead to errors, as not all the users are knowledgeable enough on the structure of your Data Model.
As an Admin, you can pre-build these combinations inside a group of Data Sets and Filter Sets. |
We have a campaign type where users need to apply filters on the products purchased in an order, which requires building a relationship between more than two Data Extensions. Non-tech users are not comfortable building it and often require support from Admin Users to help them. To simplify this campaign, the Admin can create a Data Set in the Admin Panel called “Customers with Orders and Detailed Products” with the necessary Data Extensions, and create a Filter Set in the Admin Panel called “Active Customers with Orders” so the non-tech users can use Data Set and Filter Set without knowing the structure of the Data Model. |
Master Data Extensions |
|
Use Case |
Example |
For some campaigns, you need to run queries over great amounts of data, which can lead to timeouts or very slow performance. To prevent it, you can build Master Data Extensions with the Aggregations of data you use in your campaigns, so they are pre-calculated when you needed (and refreshed nightly, for example)
|
You want to create a Selection with all customers who placed more than 3 orders in the last quarter. Create a Master Data Extension containing the ID of the customer and the number of orders in the last quarter. Whenever you need to filter out customers based on the number of orders, you can use this Master Data Extension to avoid accessing the Orders Data Extension and filter them recurringly. |
You want to simplify the day-to-day segment generation:
|
As soon as you are repeating the same Filter/combination of DEs daily, you should consider creating a Master Data Extension where you do that filter and scheduling it to run daily. After that, you can use your Master Data Extension as the base of your campaigns, knowing that the base filters or combinations are already applied |
Note: If you have a use case where real-time data is very important, then it is highly recommended to opt for Data Sets instead of Master Data Extensions since Master Data Extensions are normally scheduled to run daily/twice a day and work with static data between refreshes.
Master Data Extensions and Scheduling of Selections are a very powerful way of creating easily accessible Data Extensions, and refreshing them nightly so they are updated when you need them and they’ll keep a great performance over accessing and filtering raw data.
Comments
0 comments
Please sign in to leave a comment.