In the last post of this blog series on being successful with digital analytics, I showed how to score your business requirements and determine how many could be answered today. If your scoring showed that you can only answer a subset of your full requirements list, that means you need to either fix existing data points or implement brand new things. In this post I am going to show how you can determine which data elements you should focus on first to have the largest impact on your analytics program by creating a requirements driven SDR.
As mentioned in the earlier SDR post, I am not a fan of SDRs that simply list what data points are being collected. Therefore, we are going to create a new version of the SDR using our new requirements spreadsheet as the starting point. To do this, you will list all of your digital analytics metrics and dimensions across the top of your requirements spreadsheet like this:
I normally put these in numeric order and if you want, you can source this programmatically from another tab where you list all of your metrics and dimensions. Next, similar to what you did manually earlier, I want you to indicate which variables are associated with each business requirement. But this time, you should match data points and requirements using spreadsheet formulas. This is done by setting the cell equal to the data point name in the top row as shown here:
Once you have done this, you can create a formula (i.e. =CONCAT(M4:AB4)) that will concatenate all of the variables needed for each requirement like this:
Unfortunately, this isn’t as pretty as the far-left column we did manually, so you can use a bit of spreadsheet trickery to clean it up. First, use the following formula to see how many variables are needed for each requirement and add it as a column:
=COUNTA(L4:Q4)
Then copy it to each row to see this:
Next, you can leverage the count of items to clean up the variable list using this formula:
=IF(K4=0,””,MID(SUBSTITUTE(J4,”]”,”]”&CHAR(10)),1,LEN(J4)+(K4-1)))
This formula assumes you have added a “]” (or something similar) to the end of each variable, this formula will insert a line break after each so your data points column can now be dynamically driven via formulas and you will see this:
Now you can hide the ugly concatenation column of your variables and if you ever add any variables to a row, they will automatically be added to the data points column.
Last, but not least, I like to add a count formula above each variable so I can see how many business requirements use each data point:
You can also add filters to all spreadsheet columns which will allow you to easily filter to see all requirements using a specific variable.
So, when you are done, you now have what I call a requirements driven SDR. You have one easy way to view all of your business requirements AND which data points (variables) will be used for each:
This new requirements driven SDR is something that you can use to manage and really drive your digital analytics implementation.
The only thing that is missing now are the data points that have yet to be implemented or existing ones that need to be fixed which we will address in a subsequent post.
Action Items
Your task for this post is to create your own requirements driven SDR using the instructions and formulas provided above. We will build on this document later and you will see some cool ways that it can be used during and post-implementation:
- Add columns for your metrics and dimensions as shown above
- Fill in the appropriate events and metrics for each requirement
- Add the columns required to make the Data Points column dynamic
In the next post, we will start tackling the new data points needed (in our previous example, the Chat-related data points).
