How Cosentino's turned personalization into repeat visits 

Cosentino's Food Group is an independent grocery retailer in the United States, operating through banners Price Chopper, Sun Fresh, and Cosentino's Market. Millions of shoppers visit Cosentino's stores every year, and as a community-oriented grocer, the company wanted to grow sales and deepen customer relationships without making discounts the sole growth lever.

Grocery margins don't leave much room for error, and personalization at this scale creates a hard problem: you have to first figure out what shoppers want next, then deliver that message to hundreds of thousands of individual shoppers every month without the system breaking down.

Cosentino's solved the first half of that problem with Goodlight AI, whose models turn loyalty and point-of-sale data into shopper-level personas and recommendations. It solved the second half with Customer.io, the platform that turns Goodlight's intelligence into millions of individual sends a month, reliably enough to make the whole approach work.

• The challenge: growing sales without discounts in a thin-margin category
• Turning loyalty and POS data into shopper personas Goodlight AI can act on
• From insight to action: Goodlight's AI agent meets Customer.io
• Delivering industry-first personalization with Customer.io
• The result: open rates from 15% to 48%, trip frequency up 20%, shopper spend up 21%

Grocery operates with extremely thin margins. Discounting, the most common promotion practice in the industry, hurts the bottom line. Cosentino's partnered with Goodlight AI, who proposed to test the thesis that they could drive repeat visits without relying on discounts by making the buying experience hyper-personalized to each shopper.

Like many grocery retailers, Cosentino's had valuable customer information spread across loyalty and point-of-sale systems. The opportunity was bigger than sending more promotions: the team needed a clearer understanding of how shoppers behaved, what they were likely to need next, and how to turn those insights into relevant customer experiences.

The goal was to make every interaction more valuable, and getting there meant solving two problems at once: building the intelligence to know what each shopper needed, and building the activation layer to act on it without slowing down.

The Goodlight AI team brought loyalty and POS data together to analyze purchase behavior by individual shoppers. Using proprietary machine learning technologies, the team generated specific personas of shoppers, based on their buying preferences, to build an overall profile of the shopper demographic. The models also predicted when each shopper would visit next and what they'd buy.

This persona generation is a first in the market for retail grocery, and it's been instrumental in giving Cosentino's a deeper understanding of its shopper base. Every profile the models built landed directly on that shopper's Customer.io profile, ready to use.

Goodlight AI combined the resulting personas and behavioral signals to generate product recommendations tailored to each shopper. Recommendations were grounded in what customers had bought, how they shopped, and what they were likely to find relevant. The personas enabled deeper, more relevant product recommendations at the individual customer level than the current industry standard.

This created a more relevant path to growth: help customers discover products that fit their needs, build more complete baskets, and give them reasons to return. Getting there depended on one more step: turning a recommendation into an actual message, in front of the right shopper, at the right time.

To generate individual, hyper-personalized marketing messages for hundreds of thousands of shoppers at once, Goodlight AI leans on an autonomous AI agent that determines the best phrasing for each shopper. The agent also identifies each shopper's cadence of store visits and determines the right time to send the message.

Once the agent makes that call, Customer.io executes it. Each decision, the message, and the moment, triggers a send through Customer.io, individually, at the scale Cosentino's needs every month. This agentic approach, paired with a platform built to activate it, allows for the highest level of personalization offered in the retail grocery industry today.

Few customer engagement platforms were built to handle the rigorous demands of Goodlight's level of personalization, where every single message sent is genuinely different. After a careful selection process, the team chose Customer.io to serve as the activation layer for Cosentino's shoppers.

The volume alone rules out most platforms. Most email service providers aren’t built to protect deliverability at this scale, but Customer.io sends more than 100 billion messages a year at 99.98% uptime, the kind of infrastructure required to make millions of one-to-one emails a month possible without dropping a single send. That reliability, backed by 24/5 support, is what lets Goodlight AI move from a working model to an always-on program.

The approach helped Cosentino's Food Group create more relevant customer interactions, resulting in stronger engagement, more frequent trips, and larger baskets. Email open rates increased from 15% to 48%, trip frequency increased by 20%, and spend per customer increased by 21%.

None of that happens without delivery. A model can predict the right message down to the shopper, but the results only show up if that message actually reaches an inbox, every time. That's what Customer.io added to Goodlight's intelligence: consistency, at a scale most platforms can't sustain.

The broader lesson is that profitable grocery growth doesn't have to start with a deeper discount. By combining loyalty and POS data, behavioral personas, tailored recommendations, and a platform built to activate all of it reliably, an independent retailer can make relevance the growth engine.

• Translate loyalty and POS data into actionable shopper personas.
• Shortlist product recommendations tailored to each customer.
• Generate individualized marketing copy, and identify the right moment to send it.
• Sync every persona, recommendation, and send decision to the shopper's Customer.io profile.
• Activate each message through Customer.io, at the volume and reliability grocery-scale personalization requires.
• Measure and report success through engagement, trip frequency, and basket growth.

If you're a retailer with a loyalty program, see how Customer.io and Goodlight AI can help you grow monthly sales without depending on discounts. Learn more about how Customer.io powers personalization at scale, or reach out to Goodlight AI to get started.