“My data scientists are building really complex models to help us understand satisfaction and reduce churn… But none of us know what to do with them.”
Nearly every week, an executive tells me this. It’s incredibly difficult to operationalize analytic insights if you can’t socialize those insights with key stakeholders in the first place.
That’s why it is critical that every piece of the Aster Customer Satisfaction Index (CSI) solution includes an intuitive “face.” Like all Aster Analytic Solutions, the CSI solution was built from the outset with operationalization as the end goal. To reach the operationalization phase of an analytic project, however, it’s important to be able to help others understand the steps that you took along the way.
Let’s start with building rules. The rules dictate the impact of each event or interaction on a customer’s satisfaction score. If I’m new to a project, I can quickly log into our CSI Guided Development Interface and view a popup for each rule to understand what is happening behind the scenes. In the screen grab above (Image 1), we are looking at a consumer financial data set. I have built a rule called “Account Booked” which will in most cases increase a satisfaction score by 30 when triggered. (The underlying data could just as easily be from a telco, insurance provider, hospitality company, auto manufacture or some other company.) A data scientist may have performed advanced modeling to determine this weight, but for others in the organization, this improved visibility into the weights behind the model is a huge leap forward.
Now let’s look at an analysis. A CSI analysis is basically a compilation of rules with a starting score and a few other parameters. In the example above (Image 2), you can see I’ve included most of the sample rules I’ve created. The two rules I excluded around the linking of external accounts have a red background. If I want to add these into the analysis, I can simply click the plus sign to the right of the rule. To see a popup with rule details similar to Image 1, I can click the “i" information icon.
Now it’s time to expose results. For those looking for an aggregate view, a few bar and line charts (Image 3) are simple to interpret. And for those looking for a much more granular view, you can dive into the results for each individual customer and see how their score has been modified as each rule has been triggered (Image 4).
Basically, if your nomenclature is straightforward, anyone in your organization can look at your scoring rules and analyses to understand what lies behind a customer satisfaction score. With Aster CSI, you can shed a bright light on much of the mystery behind the “dark art” of advanced analytics.
At the end of the day, this is why the Aster CSI solution is helping data scientists make the move from respected statistical minds to respected business leaders with seats at the executive table.
If you want to learn how you can put a face on your customer analytics with the Aster CSI solution, send me a note at firstname.lastname@example.org.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.