Yum! Brands’ secret Domo sauce: Jupyter Workspaces
Because the COVID period commenced and prevented people for a long time period of time from eating in at restaurants, shoppers everywhere have increasingly relied on cafe ordering and supply applications to place food on the table for by themselves and their family members.
To deal with the shake-up in food items-usage dynamics, Yum! Brand names’ digital and know-how teams invested drastically in the development or enhancement of this kind of apps for our places to eat, like KFC, Pizza Hut, Taco Bell, and The Pattern Burger Grill.
For KFC-United States specially, the principle of obtaining a restaurant ordering application was relatively new. To persuade KFC clients to down load and use the app, we desired to guarantee that it was “relevant, straightforward, and distinctive”—or, Pink, as our previous CEO, Greg Creed, preferred to say.
But to definitely assure that it was Crimson, we desired metrics. We wanted to know if the app was in fact earning the course of action of ordering fried hen a lot easier. Had been folks happy with the application? Ended up there recurring designs amongst shoppers who loved the app (or didn’t appreciate the app)? Did specified app release versions complete much better than other folks?
Those have been among the the thoughts we experienced to obtain answers to. Whilst both equally Apple and Android present access to buyer scores and evaluations, they do not present a deep dive into what evaluations signify for a product or service. So, we turned to Domo, and the resource that has develop into our top secret sauce: Jupyter Workspaces.
Jupyter Workspaces presents us the skill to obtain and examine this qualitative data. In my expertise with other business enterprise intelligence platforms, textual content examination has been limited to word counts and phrase clouds.
Jupyter Workspaces, on the other hand, takes text analysis to the future degree, making it possible for practitioners to incorporate Python’s sophisticated All-natural Language Processing (NLP) capabilities with datasets suitable inside of Domo. It also permits Jupyter Notebooks to be scheduled as DataFlows to mechanically refresh your information. By making use of Python and Domo in tandem, KFC can now do the next:
| Python | Domo |
| Import consumer reviews specifically from Apple and Android merchants and merge them into a single dataset | Plan the Jupyter Notebook to automatically refresh day by day |
| Use Organic Language Processing designs to discover the customer’s emotion toward the app in each review | Make a dataset that can be shared throughout the corporation |
| Extract crucial metrics these as when the evaluate was prepared and the user’s star-level rating | Illustrate results and metrics in a fascinating way, utilizing enterprise branding and interactive visuals |
All of these characteristics contribute to deriving insights for KFC’s cellular app team. Now, the staff can determine what performs for consumers and what doesn’t, and cultivate ideas for foreseeable future application improvements—which all goes to exhibit that when KFC customers communicate, we pay attention. And that, of class, is critical to very long-term brand and solution results.
