At its core, Magento BI is here to help you answer business questions - whether you simply want to see this month’s revenue compared to last year or understand your acquisition costs for your latest AdWords campaign.
What does that path from question to answer look like, exactly? We’re so happy you asked!
To help you visualize this process, we’ve mapped out that route below. This article will shed some light on both how we approach an analytical question, and the backend logistics required to get you the data you need.
We know that you are constantly asking questions to improve your business, from increasing customer satisfaction to cutting supply costs. We will focus on how to translate your questions into analyses that help you drive decisions.
For our example, let’s assume that we want to answer the following question:
With our question in hand, it is time to identify a list of possible analyses and measurements to help answer the question. For this example, let’s focus on the following metric:
This will reveal the average time that lapses between registration date and the users' first purchase date and give an idea on how users behave at this final step in the conversion funnel
Understanding what to measure only gets us part of the way there. In order to assess the average time from registration to first purchase date per user, we need to identify all the data points that our measure is comprised of.
Let’s break down our measure into its core components: we need to know the count, or number, of people that registered; the count of people that made a purchase; and the time that elapsed between those two events.
At a higher level, we need to know where to find this data in the database, specifically:
At a more granular level, we need to identify the exact data fields that will be used for this analysis:
Creating data columns for analysis
In addition to the native data columns outlined above, we will also need a set of calculated data fields to enable this analysis, including:
That will then be used to create:
Both of these fields need to be created at the user level (i.e. on the `user` table), so that the average analysis can be normalized by users (i.e. the denominator in this average calculation will be the count of users).
This is where Magento BI steps in! You can leverage your Magento BI data warehouse to create the above columns. Simply contact our analyst team and provide us with the specific definition of your new columns and we'll create them. You can also leverage our Column Editor.
It is a best practice to avoid creating these calculated data fields in your database directly as it puts an unnecessary burden on your production servers.
Creating the metric
Now that we have the required data fields for our analysis, it’s time to find or create the relevant metric to construct our analysis.
Here we know that, mathematically, we want to perform the following calculation:
And we want to see this calculation plotted over time, or trending, according to a customer’s registration date. And here's how to create this metric in Magento BI:
This metric is now ready.
Creating the report
Now the fun begins.
With the new metric set up, we can use it to report on the average time between registration and first purchase date by registration date.
Simply go to any dashboard and create a new report using the metric created above.