Coles Employs Generative AI to Analyze 40,000 Customer Comments Weekly

Li Xia
Last Updated:
November 19, 2024

Coles Group has integrated Generative AI (GenAI) to process around 40,000 customer feedback comments each week, refining these into key actionable insights at a store level, the company revealed at the Microsoft AI summit in Melbourne this week.

The feedback, collected from a weekly survey sent to customers post-interaction, typically includes 20,000 responses ranging from numerical scores, which are consolidated into a net promoter score (NPS), to unstructured comments on customer experiences. Despite NPS's prevalence as an internal benchmark for customer satisfaction at Coles, it often falls short of providing a comprehensive view.

Silvio Giorgio and Richard Walker from Coles at the Microsoft AI Summit

"The real value lies in the customer verbatims," said Silvio Giorgio, General Manager of Data and Intelligence at Coles. He highlighted that while scores might suggest satisfaction, the comments often reveal underlying issues. 

“We’d get a 9 out of 10 score, but then have the customer saying things like, ‘I couldn’t find a spot in the car park’ or ‘my favourite brand of cereal wasn’t stocked’. We realised that we were missing out on the richness of the insights in that verbatim feedback, simply because it’s not humanly possible to analyze the vast number of pieces of written feedback every week,”  - Rich Walker GM of Operations & Finance

Historically, the retailer struggled to extract actionable insights quickly due to the data's volume and complexity. Traditional methods like word clouds, although visually appealing, failed to offer actionable results, according to Richard Walker, General Manager of Operations and Sustainability.

“When we did decide to try it, we went down the traditional route of word clouds and we invested three-to-four months in there, [but] whilst they look great and it was a different lens on performance for the week, it didnt give us anything that was particularly actionable.”

The shift to GenAI has transformed this process. Now running on Microsoft's Azure platform, the AI system digests customer verbatims along with contextual scores to highlight the top three focus areas for each store.

However, the use of GenAI is not without challenges. GenAI "is like a small child," Giorgio explained, indicating that the initial outputs were often out of context or irrelevant. Fine-tuning the AI required a thorough understanding of Coles' operational structures and product hierarchies, which was the most time-consuming part of the process.

Despite these hurdles, the integration of GenAI into Coles' operations has significantly streamlined the handling of customer feedback, allowing for quicker, more targeted responses to issues like customer service and product availability. This has been particularly effective in aligning staff rosters with actual customer service demands.

“The whole intention behind the Operations team’s digital insights journey is to be able to give store teams the right information at the right time to make their jobs easier.
We also want to understand the impact of and relationship between our operations and the customer experience,” - Rich Walker

Walker hailed the application of GenAI as a "game-changer," noting it allowed the company to do what was previously impossible without an extensive team. The AI's ability to infer multiple sentiments from a single comment has enriched the data's utility, providing deeper insights into the customer experience.

“What we can then do is ... see what customers talk about most, so you can see the relative participation of that sentiment topic within a given week, and you can then trend that sentiment over time to see an amplification based on things that we do,” Walker said.

Going forward, Coles plans to automate the uploading of customer survey data to the GenAI service, further enhancing the efficiency of this innovative feedback analysis system.