Hero Insights: Unify Data in Customer Insights

by Kylie Kiser, posted in Citizen Developer, Customer Insights, Dataverse, Hero Insights, MSDyn365, Power Platform, Series and Case Studies

If you are just joining us, I am working on a series to learn about Dynamics 365 and Customer Insights using superheroes! You can check out the full series on the Hero Insights page. Enjoy!

When we left off yesterday we had imported all our data but it was still not powering anything. So today we are going to work on the Unify process. This includes 3 parts: Map, Match and Merge. Through this process we will decide the fields we want from CDS and define them, determine how to combine the data from different sources, and then choose which attributes are duplicated in each source.

Map

The mapping step starts with us reviewing the fields we want to capture from the Common Data Service. Then we have the opportunity to link these attributes to fields on the Customer Insights Person.

Customer Insights Unify: Map
You can get here by expanding Data and selecting Unify. Then click Select Entities.
Select Entities in Customer Insights
Select all the entities that you want to use in Customer Insights. I only selected my CDS entities initially but came back later and added all of the data I imported. This can be changed at any time.
Map Fields in Customer Insights
Now we can choose the fields we want to use in Customer Insights. There are a few important things to notice: (1) Check out Select Fields in the top right. This is where you can remove items you don’t need and add our new fields. (2) You can link your fields to data points within Customer Insights but you can also leave these blank. (3) You will want to select a Primary Key for each data source. Again I used the hero’s name field.

Save your mapping and then we can move on the matching.

Match

In the Match step we will select how records are “merged”. So in our case we uploaded 3 files plus the data in CDS. Each of these files would have many of the same heroes which could look like duplicates but with different attributes. We want all these attributes to appear on one record so users do not need to understand where the data is coming from or how it is related – they just see it altogether.

Customer Insights Match Process
To get started click “Set Order”
Customer Insights Set Match Order
Choose the order records should be merged together. In this case, we always want our CDS record to be the “winning” record. We also want to include all of our CDS records. If there are additional Heroes in the other data sources, those will not be included. Only data relevant to our CDS heroes will be displayed.
Customer Insights Match Rules
Now for each dataset we need to define how it connects with our CDS data. We will need to “Create new rule” for each row.
Customer Insights Set Match Rule
Above, you can see what setting the new rules looks like. You can give it a descriptive name to help you remember what you were doing. Then you select the fields that should be compared and the precision of the match. You can add multiple conditions here as needed based on the data in both sources.
Customer Insights Match Result
If we jump to the future, this is what your Match area will look like. These rules will run and show you how many were matched and your total number of records.

Merge

Now we have brought in the data and combined this data into a unique list of records. The next step is the merge the fields within that data. For example, I had multiple data sets that contained the hero’s eye color. I do not want this listed multiple times in Customer Insights, so I can create a rule to combine those fields into one.

A few of these rules are created by default and you will also see your primary keys already called out.

I did run into a few issues on this step due to fields being different data types. My recommendation here is two-fold: (1) plan out your data and be careful which items are option sets in CDS but might not work that way in your other data sources and (2) after each field you combine, save to see if it worked. Saving after each change helps you find your errors sooner so you can focus on the items that work and figure out a plan for the villains in your data.

You can also remove fields from the customer record. This will ensure your records do not get too overwhelming. This will hide the field from any other processes we do in the future so be careful.

Customer Insights Merge Data
To merge fields of the same data together, click “Combine Fields.”
Customer Insights Combine Fields Details
When you combine the fields you have to give the new field a unique name then you select all the fields that should be flattened into this attribute. Finally you can choose which order the data sources should be prioritized. The first one with a value in their version of the field will be the winner.
Customer Insights Duplicate Fields and Keys
After you are done, you will see a few fields at the bottom of your attribute list that are merged together. You can also review the keys identified in the earlier steps.

So much progress! At this point you want to save and click Run on the top ribbon. This will take a little while to consolidate all the data we have been defining. It is a good time for a tea break.

That’s all for today but tune in next time to discover how our intrepid adventures will figure out how to set up the Search and Filter items needed to start reviewing the Customers list! Will sorting be simple? Can filtering find the heroes needed?

One thought on “Hero Insights: Unify Data in Customer Insights”

  1. Pingback: Hero Insights: Unify Data in Customer Insights - 365 Community

Leave a Reply