The modern business has a lot of touch points with its customers and the environment it operates in. All this point of interaction leaves some trails which can tell a story on the health of business to customers relations. This could be buried in different data sources ranging from social interaction data, point of sale data, supply chain data, finance data, marketing data or even log in data from various data monetization projects.
Our focus in this article is exploring a simple technique of computing login frequency. We’re assuming the position of IT manager of ‘Brand X’, which has recently launched a portal from which different subscribers can consume various industry metrics through a uniquely assigned login credentials.
Related: How to measure customer loyalty.
As the person in charge of this portfolio, I would like to understand how frequently my clients use the portal, with a broad understanding of the metrics at different categories like Segment. This could help me understand whether the portal is delivering the projected numbers at the start of the project.
Using Superstores data set, my target columns for this article is ‘Order Date’ now renamed ‘Log in Date’, ‘Customer ID’ and ‘Segment’. Where ‘Customer ID’ is a unique identifier for each customer.
Connect Superstores data set to the Tableau app and follow the guideline below.
Step 1: Compute First & Last log in date for every customer.
Using the formula below.
Step 2: Compute the time lapsed between ‘First Log in Date’ and ‘Last Log in Date’.
The time unit could vary between, days, months, years, weeks etc. However, in this article we’ll compute time lapsed in months using the formula below.
Step 3: Count number of times user logged in.
Using the calculation below.
Step 4: Compute Log in frequency.
Now, that we’ve the number of months the customer has been active and the number of time he or she has logged into the portal. We can compute the login frequency using the formula below.
Rounding our calculation to the nearest integer we’ve.
Step 5: Lets build a simple chart to present our findings.
First drag the measure field Frequency bin to the dimension area.
Drag again now dimension field Frequency bin to the Columns shelf.
Drag count distinct Customer ID to the Rows shelf.
Drag dimension field Segment to the filters, Show Filter.
Adding Table calculation by computing ‘Percent of Total distinct count of Customer ID’ we’ve.
Using the filter, I can be able to drill down to other segments. See below.
With this simple view, I can fully understand how actively my customers are using the portal, reach out to customers with fewer logins and seek to understand the challenges they face, monitor whether this problem is manifested in other customer classification categories and address them accordingly. With good understanding of my data as the IT manager, I can see crisis and respond before it’s too late.
I hope with this article you’ll find a good reason to begin questioning your data and stay informed on your business from the data point.