Loyalty that changes behaviour, not just rebrands discounts

Why “loyalty” at all

Asked “why does our company need a loyalty programme”, the answer usually reduces to “to retain customers” or “to reward the loyal ones”. Both are too general to act on. They do not let you compare what the programme actually costs against what it actually delivers.

A loyalty programme has one feature that separates it from a plain discount mechanic: it should change customer behaviour in a direction that benefits both sides. If no such change happens, the programme is a marketing variant in a cabinet-and-catalogue wrapper.

The change can take several forms. The customer uses a margin-positive service more often. The customer provides more accurate data about themselves and gets relevant offers in return. The customer recommends the operator and brings new activations. The customer moves to self-service instead of calling. Each of these is measurable, has a number, has ROI.

If the programme moves the base on none of those metrics, its honest name is “permanent promotional budget”, and it lands in the P&L as a marketing cost without a deferred effect.

What gets sold as loyalty but is not

Most active programmes at telecoms in the region fall into five types. None of them does on its own what is expected of them.

Points for top-ups. The customer tops up and earns points, which convert into discounts or gifts. The problem is that the customer would top up anyway — they need the service. The programme pays cashback for behaviour that already exists. No new action emerges.

Points for service usage. Same problem. Additional minutes or gigabytes come from the natural growth of consumption that would happen anyway. The link between the accumulated point and the customer’s decision is weak.

Welcome bonus on activation. Not loyalty by definition — it is an acquisition incentive. After activation the customer’s relationship with the operator resets to standard, no special tier behaviour follows.

Partner discount catalogue. “Show our app at this cafe — get 10%”. Mechanics like these have low usage on average, and the operator’s brand starts to be associated with scattered small discounts instead of its own value.

Referral programmes. “Bring a friend, get a bonus”. Works on a narrow set of active referrers, and the referred activations often show higher churn in the first half-year — referrals were made for a bonus, not for a really suitable service.

Each mechanic has its place in the marketing mix, but calling them a loyalty programme misleads the organisation about what it is building.

What “loyalty” really costs

Direct costs — platform, team, marketing, payouts. Without a separate P&L line, the total tends to be much higher than the budget for any single campaign would suggest.

Indirect costs are harder to measure but real. When the company permanently runs a “discount for loyalty” mechanic, marketing and product lose discipline — every new offer can be repackaged as “specially for our loyal customers” instead of being made into an honest product. The brand starts to be perceived as “the operator that hands things out” rather than “the operator that gives value”.

There is also a behavioural effect. A customer used to a steady stream of bonuses comes to see them as the norm. Trying to scale the programme back is read as retraction rather than optimisation. This dependency reduces strategic flexibility — the programme becomes non-removable.

What actually changes behaviour

A real loyalty programme usually rests on four levers, and a points mechanic is not the first.

Status that is visible in the experience, not on paper. A long-tenure customer does not get more points, they get a different service path: a direct number to the contact centre with low wait time, a flag in the dealer system, a dedicated account manager when the segment justifies it. This creates a sense that the relationship is continuity, not a sequence of one-off transactions.

Exclusive access to the new. A long-term or high-value customer gets first access to a new tariff, a new partner integration, a new feature in the app. This builds engagement and reduces the appeal of switching to a competitor.

An offer based on observed behaviour. Not “you got a 10% discount because you are a loyal customer”, but “we noticed you call Turkey often — we have an international package that pays back for your pattern”. The difference between a discount and a useful personalised service.

A bonus tied to a specific behaviour change. “Switch to digital billing — get a bonus, you save on paper invoices, we save on print”. This is a real change benefiting both sides. The bonus here is a stimulus, not a gratitude.

The first lever is the hardest to deliver — it requires the front line to actually serve the customer differently. Without that operational capability, the “premium tier” becomes a declaration, and that is worse than its absence.

Preparing for a relaunch

Behavioural baseline. Which 2-3 specific behaviours do we want to shift? Not “increase loyalty” but “raise family account adoption by 30%”, “reduce contact centre calls by 20%”, “raise device financing conversion by 15%”. Without targets the programme cannot be evaluated a year from now.

Segmentation by value, not tenure. Tenure correlates with value, but loosely. A real LTV model defines where differentiation pays back.

Operational capability for differentiation. If the premium segment must receive dedicated support, the contact centre, the dealer network and retail must be able to deliver it. Without that backing the programme falls apart in the second month.

Experimentation infrastructure. Each behaviour-change bonus is an A/B test. Without testing the operator guesses what works.

P&L owner. Someone takes responsibility for the result in numbers, not in slides. Without an owner the programme becomes a cost centre without accountability.

When it is more honest not to launch

If retention is already high and churn is low, an additional loyalty mechanic is an investment with low marginal return. First — the diagnostic of why customers stay; then — whether a separate mechanic is needed at all.

If customer data is fragmented and does not reconcile between billing, CRM and the app, personalisation cannot work at the required precision. Without a data quality foundation, a new programme will hit the same wall.

If the contact centre and retail cannot deliver a differentiated service, the premium tier remains a brochure promise.

If the P&L of the existing programme is not measured separately, launching a new one without shutting down the old just stacks costs.

If the CFO is not willing to underwrite a 12-18 month investment period before visible return, the programme will not get sustained funding and will wither in the pilot phase.

Discussion points for the committee

How much do we spend each year on existing loyalty programmes — including indirect costs, not just rewards? If hard to answer, the programme is not measured.

Which 1-2 behaviours do we want to change through loyalty? If there is no answer, the programme is not focused — it cannot be evaluated.

What is the LTV distribution by segment? That decides where differentiation makes commercial sense.

Who owns the P&L of the new programme? Without an owner the project will not start substantively, even if it starts on slides.

What are we ready to stop doing in the old programme to free resources for the new? Without that, new initiatives stack and dilute each other.

How SamaraliSoft can help

Loyalty Reset & Behavioural Programme Design — an audit of the existing programme on real P&L (including indirect costs), behavioural diagnostics to identify the 2-3 behaviours that matter most for operator margin, redesign of the mechanic around those behaviours with explicit measurement, an operating model for differentiated service where it pays back, and a 90-180 day pilot plan with go/no-go criteria.

Related reading

• /en/insights/telecom-arpu-bundles-devices/ — where ARPU grows
• /en/insights/telecom-app-monetization/ — frequency and engagement
• /en/insights/telecom-subscriber-intelligence-operating-model/ — operating model
• /en/use-cases/telecom-churn-war-room-mnp/ — retention in the MNP era

Sources

• Digital Government Portal — digital infrastructure UZ
• UzDaily — telecom market in Central Asia
• Ministry of Digital Technologies — telecom in UZ
• MNP.UZ — number portability