Given that the COVID era began and prevented persons for a prolonged interval of time from eating in at eating places, shoppers just about everywhere have significantly relied on cafe ordering and shipping apps to put foodstuff on the desk for themselves and their people.

To tackle the shake-up in food-intake dynamics, Yum! Brands’ electronic and technologies teams invested significantly in the advancement or improvement of this kind of applications for our dining establishments, such as KFC, Pizza Hut, Taco Bell, and The Behavior Burger Grill.

For KFC-United States particularly, the idea of owning a restaurant ordering app was somewhat new. To really encourage KFC consumers to down load and use the application, we desired to guarantee that it was “relevant, easy, and distinctive”—or, Purple, as our earlier CEO, Greg Creed, preferred to say.

But to actually be certain that it was Red, we wanted metrics. We needed to know if the app was in truth producing the procedure of ordering fried hen easier. Ended up persons contented with the application? Were there recurring styles amid prospects who beloved the app (or didn’t really like the app)? Did selected application release versions carry out superior than other people?

All those were being amongst the inquiries we had to uncover answers to. Even though both Apple and Android offer obtain to customer scores and testimonials, they do not offer a deep dive into what opinions necessarily mean for a products. So, we turned to Domo, and the device that has come to be our secret sauce: Jupyter Workspaces.

Jupyter Workspaces gives us the ability to access and assess this qualitative info. In my experience with other company intelligence platforms, textual content assessment has been confined to term counts and phrase clouds.

Sample of a Domo/Jupyter Notebook project carried out on Doordash Critiques

Jupyter Workspaces, on the other hand, requires text assessment to the following level, permitting practitioners to merge Python’s highly developed Normal Language Processing (NLP) capabilities with datasets proper within of Domo. It also allows Jupyter Notebooks to be scheduled as DataFlows to quickly refresh your facts. By making use of Python and Domo in tandem, KFC can now do the subsequent:

Python Domo
Import purchaser assessments instantly from Apple and Android retailers and merge them into a single dataset Plan the Jupyter Notebook to mechanically refresh every day
Use Normal Language Processing versions to determine the customer’s emotion toward the app in each evaluate Develop a dataset that can be shared throughout the business
Extract vital metrics these kinds of as when the overview was prepared and the user’s star-amount score Illustrate effects and metrics in a fascinating way, working with corporation branding and interactive visuals

All of these features add to deriving insights for KFC’s mobile app crew. Now, the team can detect what performs for shoppers and what does not, and cultivate strategies for long run application improvements—which all goes to clearly show that when KFC prospects speak, we pay attention. And that, of course, is key to extended-phrase brand name and product or service achievements. 

Leave a Reply