Using Voice Analytics at POS, Because Every Conversation Matters
First published in convenience.org. This article is brought to you by InStore.ai.
Convenience stores offer unique access to one of the most powerful dynamics of a shopping experience: face time with customers.
“Unlike the digital cart that powers Amazon, for instance, or the propensity of self-checkouts ushering customers through a Walmart or grocery store, the register at a c-store provides an opportunity for operators to stand face to face with a consumer and their open wallet,” said Jay Blazensky, CEO and co-founder of voice analytics tech firm InStore.ai.
“This is a uniquely special moment for the cashier to mention promotions, attempt to upsell a product, gather information about brand preferences, glean direct feedback about the store facilities, expand the store’s loyalty program and more,” he added.
At the very least, their friendliness and helpfulness can be the difference in generating return visits. Or not.
“But this crucial behavior is not reflected in any transaction log,” explained Blazensky. “In an industry where the high turnover of cashiers is a constant challenge, the inability to identify and reward top performers is a gap that desperately needs to be addressed,” he continued.
Blazensky and Marc Della Torre, COO and co-founder of InStore.ai, hail from Silicon Valley and have backgrounds in customer call centers, where voice analytics has been highly successful in monitoring agent performance and improving customer experience, Blazensky said. Now, the two are introducing voice analytics to the convenience store industry, aiming for similarly positive outcomes. They are using generative AI and large language models (LLMs) to tackle the unique challenges presented by the convenience store environment.
The InStore.ai system uses microphones placed around the store—such as at the register, near the fresh food section or at self-checkout—to capture the voice interactions between cashiers and customers. A small edge computing device under the register sends the transcript and recordings to the cloud, Blazensky explained, where the conversations are then automatically processed using artificial intelligence.
“The InStore.ai software aggregates all of the data into an intuitive and powerful dashboard that allows users to quickly identify trends and anomalies from conversations with customers,” said Blazensky. Users can receive notifications and alerts when specific events occur, and quickly listen to conversations to validate the insight before taking action.
For example, a retailer might want to analyze the specific language cashiers are using to ask if a customer is a member of its loyalty program and which phrasing results in a successful sign up. Or when a cashier successfully upsells a customer and is able to increase basket size, a retailer can save that recording for training new employees.
“From measuring and improving cashier engagement to identifying issues that affect customer experience, such as not getting a receipt at the pump, there are numerous use cases for this highly innovative technology,” Blazensky said.
The second part of this series, coming Thursday, examines the power of using semantic search to query all the conversations in every store to see how each store is performing against that topic or metric.
This is the first in a series of articles on InStore.ai.
