We're excited to celebrate the first successful quarter of Flexpertise, our new subscription service offering hours-based advice, training and support from our qualified Tableau & Alteryx consultants. We've already solved some fascinating challenges for our clients, including churn, elastic pricing, CRM utilisation and many more.
We're passionate about solving business problems and Tableau and Alteryx are some of the ways we deliver business value using data. This month we've enabled our customers to score and predict their customers' likelihood of leaving the customer base so that they can focus on keeping customers happy and securing important revenue streams.
"The churn rate, also known as the rate of attrition, is the percentage of subscribers to a service who discontinue their subscriptions to that service within a given time period."- Investopedia
The definition of churn varies from company to company, but our client wanted to know which percentage of customers had reduced their total spending by 25% year on year. We did this all in Tableau using calculated fields that first identify transactions in a dynamic time frame, then using level of detail calculations that aggregate and compare transaction totals at the customer level from year to year. This method allowed us to show the total year on year growth in spending for every customer, and identifying the percentage of customers that show a significant fall in revenues. We were even able to improve the speed and efficiency of a pre-existing churn dashboard created by the client by reducing the number of calculated fields and data stored by around 20%. Thanks to our Flexperts, our client can spend less time waiting for insights and more time pleasing customers.
Ever wanted to know if you'd make more money by dropping the price of your product? Is it possible to predict how much more you would sell if you offered a discount?
What you want to know is what economists call the Price Elasticity of Demand. Without going into a full economics lesson here, it's a dimensionless number that tells you the percentage response in quantity bought resulting from a percentage change in price. So for a relatively price elastic product with PED of -1.5, a 10% decrease in price will predict a 15% increase in quantity sold. Handy Huh?
This number isn't constant because it can vary depending on many factors like the consumer segment, the time, location and the product. To know how it applies to your company, you will need to use your transactional data and fit a model controlling for price, time, product, customer segments and other relevant dimensions. We were able to do all of this in Alteryx, including graphical data investigation using Alteryx's powerful R-based predictive tools.
One of our delivery projects revolves around enriching feedback reports for use in a data discovery platform. We were able to build a custom Python script which infers the gender of whom the report is talking about by searching for keyword clues in the language used. This means that the client will be able to search, aggregate and find correlations between a greater number of dimensions in the corpora available.
For more information about how you can extract more value from your data,