A long-awaited product has finally launched, and the marketing team has created a robust integrated marketing campaign that includes various messages and offers in email, search, social, new web content, videos, and more. A few days later, you, as an analyst, are asked to create a set of reports that will provide the executive team with daily performance updates on the campaigns, ultimately connecting the various interactions to purchases – both online and in-store.
You open your analytics tool, and as you sift through the data, you notice several inconsistencies in how the tracking codes were applied across and within the channels. Something is clearly wrong with how the data is coming in, and it’s going to impact both your campaign and marketing channel reporting.
What do you do?
- Option 1: Make do with the data you have, and cobble reports together as best you can.
- Option 2: Work with channel owners to update the live tracking codes going forward.
- Option 3: Create a new processing rule that adjusts the digital data as it is collected, on a go-forward basis. Request IT or engineering to reprocess and reingest any offline data affected.
- Option 4: Spend a significant amount of time creating manual workarounds to get the data into a format that can be analyzed or presented.
Depending on the severity of the issue, you may consider these four options, but each comes with its own challenges. What if as an analyst, you could make necessary changes to the data in the system of analysis, without the need to adjust your collection or reprocess the data? What if you had the ability to change the way data appears in your reporting all by yourself, and it was easy and instant?
Today, I want to introduce an industry-first capability called derived fields within Adobe Customer Journey Analytics. Derived fields empower users to apply rule-based logic to make on-the-fly updates to their data that is instant, retroactive, and non-destructive to the underlying data, reducing costly and time-intensive cycles for updates.
That means analysts can make complex changes to data in just minutes to a few hours. And the rules apply to historical data, not just data going forward, with no processing or engineering support needed.
Transform data on the fly with Customer Journey Analytics
Customer Journey Analytics is technology built specifically for continuous customer analysis. It helps users unlock omnichannel insights at a person level — no SQL or IT queue necessary — so anyone responsible for the customer experience can visualize and analyze the customer journey in full context.
With data flexibility as a core pillar, Customer Journey Analytics integrates data from various platforms and sources. It also provides real-time data manipulation and non-destructive historical data restatement in ways that standalone digital analytics and BI tools cannot.
Introducing derived fields
Derived fields allow you to perform complex data manipulations through a customizable rule builder. Advanced functions within the rule builder help you to find and replace values, create case when logic, concatenate or parse URLs from existing data to create new fields, and much more. You can then use the derived fields you create as metrics or dimensions in Analysis Workspace with all the standard visualization and filtering capabilities available.
Derived fields can save teams a significant amount of time and effort, compared to transforming or manipulating your data in locations outside of Customer Journey Analytics. This capability empowers the core users and analysts to take control of their data and make the changes they need, when they need them, to better answer the questions most critical to the business.
How to implement derived fields in Customer Journey Analytics
You can create derived fields within a data view. A data view is a container specific to Customer Journey Analytics that lets you determine how to interpret data for specific business groups and analysis use cases. It provides a customized view of your data, where you can specify all the dimensions and metrics needed for analysis with settings such as session timeout, persistence, and attribution.
Within data views you can:
- Change a metric to a dimension and vice versa — use a string field as a metric or a numeric field as a dimension.
- Create multiple metrics with different attribution models or lookback windows from the same underlying data field.
- Overwrite display names with friendly names.
- Apply formatting to data such as a decimal, time, percent, and more.
- Automatically include or exclude certain values within a specific field.
- And now, create derived fields using a rule-builder for more complex manipulations.
Any settings that you select or change in a data view are run at report time