How to build a data-driven loyalty program

Why Do Businesses Need a Loyalty Program?

What is a loyalty program? It’s a rewards system that motivates customers to make repeat purchases and helps businesses collect data on customer behavior. A loyalty program is a mutually beneficial exchange: the customer agrees to share data about their behavior, and the business provides them with useful rewards and personalized offers. When this exchange is structured effectively, it can yield much more than just repeat sales.

Statistics: The Cost of Retaining vs. Acquiring a Customer

The numbers speak for themselves. Depending on the industry, acquiring a new customer costs a business 5 to 25 times more than retaining an existing one. Other studies show an even more striking effect: increasing customer retention by just 5% can boost a company’s profits by 25–95%. Therefore, a loyalty program can yield significant returns, provided its mechanics are based on real customer data.

Which Businesses Benefit Most from Loyalty Programs

Customer loyalty programs are most effective in situations where there are regular repeat purchases and enough transactions to accumulate statistically significant data on customer behavior:

However, for businesses dealing with one-time purchases (such as real estate sales), a full-fledged CRM loyalty program is usually less relevant. In such cases, it makes more sense to focus on referrals, after-sales service, and partnership mechanisms.

Types of Loyalty Programs

Loyalty programs differ in the logic behind how customers receive benefits. The choice should be based on the purchase cycle and the product’s profit margin. What works for a coffee shop may not yield the same results in the high-end B2B segment.

Rewards System

The most common format for a customer loyalty program is the rewards model, in which customers earn points for purchases and later redeem them for discounts, products, or other rewards. The mechanics are simple to implement and easy for customers to understand from day one, which is why this is the format most often used to launch a loyalty program for an online store. The main risk is the devaluation of points if the redemption threshold keeps rising: customers notice this and lose trust faster than you might think.

Cashback

Instead of virtual points, customers receive a portion of their spending back in cash or as a bonus credit to their account balance. The advantage of cashback lies in its transparent monetary value: customers don’t need to convert points into discounts on their own. Without proper analytics, it’s difficult to tailor cashback to product margins, the behavior of different customer segments, and the company’s business goals.

Tiered System (Levels)

A current example of a tiered model is the 2026 update to Starbucks Rewards. After several years of operating without status tiers, the company reinstated three member categories with different benefits and reward accumulation rates. In this way, Starbucks aims to better reward active customers and give them additional motivation to advance to the next level. Starbucks Rewards now once again features three tiers with different star-earning rates. The lesson is simple: a loyalty system without tiers rewards everyone equally—and therefore, no one truly.

Coalition Program

Bringing together several independent brands under a single rewards currency looks appealing on paper, but in practice, it relies on a very delicate balance of interests. A prime example is American Express’s Plenti, which initially included Macy’s, AT&T, Exxon, and several other major players. The program shut down in 2018 after several key partners withdrew. Among its problems were low member engagement, insufficient brand recognition, and the difficulty of reconciling the interests of different brands. For the Ukrainian market, the coalition model may be suitable for niche but complementary businesses whose target audiences truly overlap.

Subscription

The customer pays a regular fee and, in return, receives ongoing perks, ranging from free shipping to exclusive content. This model works well when there is more to offer the customer than just a discount: the value of the subscription is based on a steady stream of clear and desirable perks.

How Data Makes a Loyalty Program Effective

The mechanics are just the surface. What transforms an ordinary loyalty card into a full-fledged data-driven loyalty program lies beneath the surface: in how a business processes information about each purchase.

Customer Segmentation

Without segmentation, all program participants receive the same offers, a significant portion of which do not match their interests or current needs. Categorizing customers by purchase frequency, average transaction value, and interest in specific categories helps ensure that relevant offers are shown to the right segments.

Personalization of Offers

Personalization of offers is a result of effective segmentation: a customer who buys children’s products won’t be offered a discount on alcohol. A prime example is the personalized recommendation system we built for the Planeta Kino movie theater chain: instead of mass mailings, the algorithm selects push and email notifications tailored to a specific moviegoer based on their viewing history, which directly boosted conversion rates across communication channels. At this stage, Excel spreadsheets are usually no longer sufficient. Implementing BI analytics helps consolidate data and track customer behavior patterns over time.

Predictive Analytics

The most advanced level is when loyalty program analytics don’t just describe the past but predict the future: which customers are on the verge of churning, who should be offered an upsell, and whose interest in the brand is already waning. Today, such analysis can be performed using off-the-shelf BI tools, CRM platforms, and machine learning models integrated with transaction and communication data.

Loyalty Program Performance Metrics

Before discussing how to create a loyalty program that actually impacts profits, it’s important to agree on which metrics to use to evaluate its success. Without this, any CRM-based loyalty system becomes nothing more than a pretty facade, with no way of knowing whether it’s actually working.

Retention Rate

The retention rate shows what percentage of customers remained active during a selected period. The formula is as follows:

Retention Rate = (customers at the end of the period − new customers) ÷ customers at the beginning of the period × 100%.

This is a basic metric to start with. An increase of just a few percentage points can significantly boost a company’s profits.

CLV (Customer Lifetime Value)

CLV indicates a customer’s projected value to a business over the entire duration of their relationship. A loyalty program directly influences this metric: the more precisely a business tailors its offers to a specific customer, the longer that customer remains active and the higher their lifetime value becomes.

Redemption Rate

This is the percentage of accrued bonuses or points that customers have actually redeemed. The global average hovers around 50%, meaning that about half of the accrued rewards remain unused. A low rate may indicate complex rules, unattractive rewards, or a threshold for redeeming points that is set too high. A very high Redemption Rate should be analyzed alongside profitability to ensure that the cost of rewards does not exceed the revenue from additional purchases.

How to Integrate a Loyalty Program with CRM

Without CRM integration, loyalty mechanics operate in isolation from the rest of the business processes: marketing sees one set of data, customer support sees another, and the business owner sees a summarized report from a month ago. Integration brings together purchase history, bonus balance, and customer communications into a single profile.

Salesforce for Loyalty Programs

Off-the-shelf solutions like Salesforce Loyalty Management are ideal for companies that need to get up and running quickly without having to build an entire system from scratch. Basic mechanics—such as point accrual, tiers, and coupons—are already available within the platform. At the same time, a ready-made platform can limit flexibility: non-standard mechanics and specific business processes must be adapted to its architecture.

Custom Development

When a business has outgrown off-the-shelf templates or has an atypical sales model, it makes more sense to build a system tailored to specific processes—either from scratch or based on a flexible framework. It can be integrated with existing CRM and e-commerce solutions. This type of development requires more time and investment upfront but allows for the implementation of mechanics not supported by off-the-shelf platforms.

How IWIS Helps: From Mechanism Development to Performance Analytics

At IWIS, we develop loyalty systems tailored to specific business processes: we select the mechanics, integrate them with the CRM, and configure performance analytics. If a company has already accumulated purchase history but isn’t yet using it for segmentation and personalization, we can implement BI analytics separately without a complete overhaul of the existing system.