The scenario is always the same. Launch day is approaching. The technical build is slightly behind schedule. Pressure is high and resources are tight. And the inevitable question comes. What are we going to drop from the scope to make the launch date?

In our work with clients who are implementing a retail personalization program, there is one thing that always seems to be on the chopping block: the offer bank. If we were planning to launch with 1,000 offers in the bank, maybe we can get away with 500. If 500 was the target, maybe 250 will do. After all, the logic goes, we don’t need all the offers in the first week. We can always build up the offer bank later.

I understand the temptation. I have been there myself. But I always urge retailers to look elsewhere if they are trying to save on time or effort. If delivering true 1-to-1 personalization to your customers is important to you, the offer bank should be sacred. Here are three reasons why…

1. The offer bank is the key driver of relevance…even if you are only sending a single offer

People often assume that there is a direct correlation between the number of offers they are planning to give customers and the size of the offer bank they will need. If you were going to send customers ten coupons a week from a bank of 500 offers, then if we only send them five a week, we should only need 250 offers, right? But this thinking is flawed.

The truth is that the breadth and depth of your offer bank is the key driver of the relevance of your offers to your customer. The best AI engine in the world can’t assign the right coupon to someone if it isn’t available in the first place.

Here is a simple way to understand the need for a deep offer bank. Take a moment and think of your favourite fruit. Now imagine you your favourite retailer wants to send you an offer for fruit, but the only offers they have been able to source from their suppliers were for apples and oranges. If you happened to choose apples or oranges as your favourite, you will be fine. But if you prefer kiwis…or blueberries…or mangoes, you are out of luck. The personalization algorithm will give you the highest scoring item from the offer bank. But the item you really want was never in the offer bank to begin with.

2. The real payoff happens further down the list

Giving customers offers on items they have bought or shown interest in is a good start. But there is another compelling reason to build a deeper offer bank. Consider the following chart, which shows the purchase patterns for three customers across three groups of products: “A” items, “B” items, and specialty products.

Personalization - civlaue

First, you may notice that if you built an offer bank of only “A” items, only Anna would have a good experience with personalized offers. Mark would get the same two offers on rotation over and over. While Sophie would get zero offers that were relevant to her.

But think also about the psychological and emotional experience of getting personalized offers. Think about the types of products that would be included in Anna’s offer set from a bank of “A” items. They are the best-selling SKUs of the most popular brands, household names that everyone will recognize. These are the items that normally find their way into Anna’s basket. But in truth she is unlikely to be too impressed. After all, these same items are regularly on promotion for everyone. The offer might be personalized. But they will not feel personalized.

On the other hand, give Sophie a set of offers on her favourite niche products, and she will be thrilled. Not only does it show that you are really paying attention to her, but most of these products are probably never featured in the flyer. She is simply not used to getting a discount on them.

The benefits are not just emotional, but also financial. Personalized offers on specialty items tend to drive much higher lift on average.

3. You are only as good as your worst offer

Finally, we must consider how the customer will perceive not just each individual offer, but the offer set as a whole. Let’s say a retail personalization program sends you 6 offers a week. If 5 or 6 are spot on, you will probably be engaged and happy…and much more likely to check your offers again next week. But what if 3 are perfect and 3 miss the mark? Not only did the retailer get half of them wrong. They also hurt your impression of the 3 that were right. Instead of feeling valued as a customer, you are much more likely to think the three that were correct to random chance.

Building a deeper offer bank is not easy, and it is not flashy. It usually involves a lot of behind the scenes work – negotiating with suppliers, chasing down images, calculating discount rates. But the work is worth it. For retailers that put the customer first and make it a priority to build a deep offer bank, the rewards can be huge.