The Personalization Disconnect: What Business Leaders Expect vs. What They Actually Get

Across every grocer, personalization is viewed as a critical lever for deepening loyalty, growing basket size, and linking digital engagement to in-store sales.

From Loblaw to Sobeys to regional banners, each emphasizes the use of data and analytics to deliver curated product recommendations, personalized offers, and tailored shopping experiences. Yet, there is a significant disconnect between how business leaders with P&L accountability define personalization and how personalization is operationalized.

Take personalized offers, for example. In grocery, these offers have become ubiquitous—70% of customers have received one in the past 30 days, but 56% say the offers are irrelevant to their needs.

When business leaders talk about personalized offers, they envision a system that treats each shopper as a market of one — understanding their micro-behaviors, preferences, and intent.

In practice, however, systems are rooted in segments and follow a familiar three-step playbook:

Step 1. Define segments

Segments are constructed using existing customer signals available within the grocer’s environment. These may include a customer’s geographic location, life-stage signals, interest tags, average spend (overall and by category), basket size, deal-seeking behavior, and brand loyalty.

Step 2. Assign a fixed pool of offers to each segment

For example, Segment A might receive 20% off eggs, while Segment B gets 15% off the same item. The pool of offers is revised on an ongoing basis.

Step 3. Deliver "personalized" offers.

The system computes an offer relevancy score for each customer-offer combination within their assigned segment. Offers with the highest scores are then delivered to the customer.

What is delivered is often labeled as “personalized,” but not in the true sense — and certainly not to the expectation of business leaders. This approach falls short of realizing the full value that personalized offers can deliver because it:

1. Assumes each customer has a single willingness to pay.

In reality, willingness to pay is product-specific. A customer may be highly price sensitive when buying organic chicken, yet price plays no role in their purchase of premium coffee.

2. Ignores sequential choice behavior.

Spending $40 on poultry per month could mean one stock-up trip or four weekly replenishment visits. Same total spend, but vastly different routines — one calls for timing-based offers, the other for basket-building. Treating both customers as the same leads to mistimed and ineffective incentives.

3. Limits assortment variability.

The fixed offer pool represents a small fraction of the grocer’s full assortment, which includes tens of thousands of SKUs. As a result, grocers fail to deliver offers on products that meet a customer’s unique preferences and needs.

4. Hinders discounting scalability and exploration.

The fixed offer pool constraints the grocer’s ability to explore varying discount levels to influence each individual customer’s behaviors. Afterall, the goal is to influence these behaviors at the lowest discount possible.

5. Overlooks customer-level cannibalization and halo effects.

Every discounting decision should be evaluated based on its incremental impact on revenue and gross margin while accounting for cannibalization and halo effects — at the individual customer level.

There are levels to what it means to deliver “personalization.” Business leaders must look beyond the label and probe deeper — examining the underlying logic, data, and decision systems driving their personalization efforts. Only then can they ensure those systems are truly designed to optimize value at the individual customer level.

Move Beyond Segment-based Approaches and Embrace True 1:1 Personalization.

Schedule a free strategy session to discover how Polymatiks can help you reimagine personalization to drive customer engagement, loyalty, and sustainable growth.