CRM and loyalty 101: customer data, lifecycles, programmes and retention

CRM (customer relationship management) is the discipline of using what a company knows about each customer to decide what to say to them, when and through which channel. Loyalty is its best-known tool: a loyalty programme rewards people for coming back, and gives the company the data to know who they are.

In one line: CRM turns customers you already have into the cheapest growth you can get, if you know them and treat them well.

CRM runs on a customer record, usually held in a CRM system or a customer data platform (CDP), that joins purchases, contacts, preferences and consent. It holds first-party data the company collects as people buy and use the product, and zero-party data that people choose to give, such as preferences in a profile.

Segmentation and lifecycle

The oldest segmentation in the discipline is RFM: recency, frequency and monetary value. A customer who bought last week, buys often and spends a lot gets different treatment from one who bought once a year ago. Segmentation of this kind then drives lifecycle marketing: messages timed to where each person is in their relationship with the brand, often sent by marketing automation and adapted through personalisation.

Designing a loyalty programme

A programme is defined by a few choices. The earn rule says what members get for each purchase or action; the burn options say what they can spend it on; tiers set the thresholds for status; and the data promise says what the company will do with what it learns. Rewards that are easy to understand and reach keep members active; rewards that are hard to use leave points unspent and members indifferent. Paid memberships make the trade explicit: a fee in return for benefits worth more to the member than it costs. Coalitions spread the costs and the data across partners, which makes it vital to agree who may contact the member and for what.

Measures

• Retention rate is the share of customers at the start of a period who are still customers at the end; churn is the share who left.
• Customer lifetime value (LTV) estimates the profit a customer brings over the whole relationship; set against the cost of acquiring a customer (CAC), it says how much a business can spend to win and keep one.
• The Net Promoter Score (NPS) asks customers how likely they are to recommend the company to a friend or colleague, on a scale of 0 to 10. Those who answer 9 or 10 are promoters, 7 or 8 passives, 0 to 6 detractors, and the score is the percentage of promoters minus the percentage of detractors¹.

As with any campaign, the effect of a retention programme is best read against a holdout group of customers who did not receive it.

Modern loyalty marketing began in the air. American Airlines launched its AAdvantage programme in 1981, one of the airline industry's first frequent flyer programmes to reward members for their travel¹⁵. Retail followed: Tesco launched its Clubcard in February 1995, and the analysis of its data became a model for retail CRM¹⁶.

Bain & Company shaped how the discipline measures itself. Its partner Fred Reichheld argued that keeping customers is a profit lever⁵, and in a December 2003 Harvard Business Review article, "The One Number You Need to Grow", proposed that a single question about recommendation predicts loyalty and growth better than typical satisfaction surveys². Paid membership grew in the same years: in 2005 Amazon launched Prime, offering free two-day shipping on one million items for $79 a year¹⁷.

CRM and loyalty are used in every region, wherever a business can identify repeat customers.

• Europe. In the UK, supermarket loyalty pricing is widespread enough for the competition regulator to study it: the Competition and Markets Authority (CMA) examined about 50,000 loyalty-priced products at Tesco, Sainsbury's, Morrisons, Waitrose and Co-op¹⁴. In France, retailers asked the data protection authority, the CNIL, to clarify what loyalty data customers may take with them¹³.
• Americas. Airline frequent flyer programmes began in the US in 1981¹⁵, and paid memberships such as Amazon Prime started there in 2005¹⁷.
• Middle East and Africa. Coalition programmes, where one currency works across many brands, are a feature of Gulf retail. Majid Al Futtaim launched SHARE in the UAE in 2019, letting members pool points with family or friends and spend them across the group's brands¹⁸; in 2022 it added SHAREPay, which lets members link up to 10 bank cards to earn and spend points by tapping at the till¹⁹.
• Asia-Pacific. Loyalty programmes are common across retail, travel and telecoms under national data protection laws; this primer has no comparable sourced figures to give.

CRM and loyalty serve retention and loyalty first: keeping customers, raising what they spend and bringing back those who lapsed. They also support conversion (an offer to a customer who left a basket), advocacy and referral (promoters asked to recommend a friend), community building around members, and measurement, since customer data shows what each campaign changed.

They matter most where people buy repeatedly: retail and e-commerce, hospitality and travel, financial services, telecom and tech and B2B software, where churn is watched closely. In an integrated plan, CRM decides who receives the email, the mail pack or the message, and loyalty gives a reason to identify oneself at every purchase. The discipline's hub is CRM and loyalty; see also the marketing operations and marketing leader pages.

CRM costs come in three layers: the systems (a CRM or CDP, marketing automation, analytics), usually priced per contact or per seat; the people and content that run lifecycle programmes; and, for loyalty, the rewards: points that must be honoured, discounts given away, and benefits such as free delivery. Coalitions add fees between partners. A paid membership turns part of the cost around, since members pay to join; Amazon Prime started at $79 a year¹⁷.

Loyalty prices can be a large give-away. The CMA's 2024 review found:

92% of loyalty prices offered a genuine saving against the usual price in the same store, and shoppers could save 17% to 25% on average by buying loyalty-priced products; yet 55% of shoppers surveyed thought non-member prices were inflated, and 43% thought it unfair that members pay less¹⁴.

What retention is worth

Bain's research is the most quoted source on retention economics. In financial services, Reichheld wrote, a 5% increase in customer retention produces more than a 25% increase in profit, partly because returning customers buy more over time and cost less to serve⁵; for online businesses, Reichheld and Phil Schefter wrote that keeping 5% more customers can raise profits by 25% to 95%⁶. These are ranges from Bain's case work, not a rule for every business.

This section describes the main frameworks; it is not legal advice.

• Profiling. The GDPR defines profiling as automated processing of personal data to evaluate personal aspects of a person, such as to analyse or predict their preferences, behaviour or economic situation; anyone may object at any time to processing for direct marketing, including profiling linked to it⁷. EU guidance from the Article 29 Working Party describes profiling as gathering information about people and analysing their characteristics or behaviour to place them in a category or make predictions¹². See GDPR.
• Automated decisions (EU). Article 22 gives people the right not to be subject to a decision based solely on automated processing, including profiling, that produces legal effects or similarly significantly affects them⁷. The ICO explains that "solely" means a process that is totally automated with no human influence on the outcome, and gives examples such as an automatic refusal of an online credit application⁸.
• United Kingdom. The ICO's guidance on automated decisions and profiling is under review following the Data (Use and Access) Act, which came into force on 19 June 2025⁹; the ICO published draft updated guidance on 31 March 2026¹⁰. Under the new law, the general prohibition now applies only where special category data is used, so organisations can rely on any lawful basis, including legitimate interests, for other automated decisions¹¹.
• France. In October 2025 the CNIL clarified that, in loyalty programmes, the barcodes of products bought, when linked to the customer's identity, and the amount of promotions obtained are part of the data customers can access and take with them under the right to portability¹³.
• Loyalty pricing (UK). The CMA found very little evidence of supermarkets inflating usual prices to flatter loyalty deals, and no evidence of consumer law being broken in how loyalty data is collected and used¹⁴.

• Bribes are not loyalty. A discount for members can become the price members expect; the CMA found most shoppers suspicious of non-member prices¹⁴.
• Points without data use. A programme that collects data and does nothing with it is only a cost.
• A score is not a strategy. The evidence for NPS as a predictor of growth has been questioned: a replication found that Reichheld's original analysis used past growth, which shows a link but not that the score predicts future growth³.
• Survey bias. In a Bain example, a survey with a 20% response rate showed an NPS of 50, but research among non-responders put the true figure at about -22⁴.
• Retention claims out of context. Bain's 25% to 95% range comes from specific industries and cases⁵⁶; model your own customers.
• Profiling without care. Objections to profiling for marketing must be honoured⁷, and decisions with significant effects may need a human in the loop⁸.

• New UK rules in practice. The ICO is finalising its guidance on automated decisions and profiling after the Data (Use and Access) Act⁹¹⁰¹¹.
• Portable loyalty data in France. The CNIL's clarification means customers can take their purchase history with them¹³.
• Coalitions that pay. Gulf programmes such as SHARE are linking points to payment cards¹⁹.

Choose it when repeat purchase drives your business and you can know your customers lawfully. Pair it with email and direct marketing to reach them, with objectives and KPIs to set retention targets, and use which mix for which objective to weigh it against the other disciplines.