How Ready Is Your Brand for Rewards That Work?

Rewards Readiness Assessment — Full Reference Content

Dimension Insights

Intent — The Intent Gap (72%)

McKinsey found that up to 72% of trade promotions are unprofitable. The root cause is the same every time: the promotion did not have a defined behaviour it was trying to change. It had a budget, a creative brief, and a hope that sales would go up.

FNB eBucks members hold 3.2 products versus 1.4 for non-members. That is not an accident. It is a programme designed around a specific behaviour: product holding. The reward structure escalates with products held. The behaviour creates stickiness. The stickiness reduces attrition from 25-35% to 3-5%. Every rand spent has a defined job.

Coverage — The Coverage Gap (64%)

VodaBucks has over 14 million monthly active users. Vodacom has 39 million prepaid customers. That means 64% of the prepaid base never interacts with the programme. Meanwhile, prepaid customers declined 5.7 million in FY2025.

Shoprite Xtra Savings reaches 33.7 million households and captures 88% of turnover. Dis-Chem Better Rewards works at the till with a card swipe. No app. No data plan. No login. Coverage is not a nice-to-have. It is the difference between a programme that shifts market behaviour and one that entertains a fraction of your base.

Relevance — The Relevance Problem (99.998%)

MTN Nigeria's Mega Billion Promo: N290 million spent. 1,500 winners out of 87.3 million subscribers. The other 99.998% recharged, hoped, and received nothing. Recharge spiked during the 90 days. Then it reverted. The budget bought attention, not retention.

The same N290 million could fund millions of guaranteed lifestyle rewards. Every subscriber feels value on every recharge. Every recharge creates a data point. The behaviour compounds instead of evaporating on Day 91. That is the difference between a reward that is relevant and one that is just exciting.

Delivery — The Delivery Gap (6 steps)

VodaBucks: earn, bank before Friday midnight, choose to spend or convert or withdraw. Minimum three steps with a weekly deadline. Miss the deadline, lose what you earned. Clicks ClubCard: cashback loads automatically on the 20th. No step. Dis-Chem: discount prints on the till slip. Zero steps.

Every step between earning and feeling value is a drop-off point. In mass markets where the average customer spends R55 a month on airtime, complexity is not a design choice. It is a filter that removes exactly the customers you are trying to retain.

Measurement — The Measurement Gap (R2.4bn)

Discovery can prove that Vitality Diamond members have 57% lower mortality and 47% lower disability claims. That actuarial proof funds R2.4 billion in annual shared-value payouts. The rewards pay for themselves because Discovery can draw a direct line from behaviour change to financial outcome.

If you cannot prove that your programme changed a behaviour, every budget cycle becomes an argument about feeling. The programmes that survive budget season are the ones that can say: we spent X, we changed Y behaviours, each behaviour change was worth Z. That is the difference between a cost centre and a growth engine.

Recommendations by Dimension and Tier

Intent — Low Tier (0-2)

Your programme does not have a defined commercial trigger. This is the most expensive gap because every rand spent without a behaviour target is a rand spent on hope. Start here: pick one behaviour you want to change. Repeat purchase within 30 days. A second product added. A referral that converts. Attach a rand value to that behaviour. That value is your reward budget ceiling. Everything else follows from that single decision.

Intent — Mid Tier (3-5)

You have a sense of what you want to drive but the connection between reward and behaviour is loose. This is common. The programme was probably inherited or built by a vendor who optimised for engagement metrics rather than commercial outcomes. Tighten the link: for every reward mechanic in your programme, write one sentence that says ‘We pay X when the customer does Y, because Y is worth Z to us.’ If you cannot complete that sentence, the mechanic is not earning its place.

Intent — High Tier (6-8)

Your intent is well defined. The next question is precision. Are you rewarding the behaviour at the exact moment it happens, or are you batching rewards and losing the association? The tighter the link between action and recognition, the more efficiently each rand works. Look at your cost per behaviour change and find the mechanic with the lowest cost. Scale that one.

Coverage — Low Tier (0-2)

Your programme reaches a small fraction of your customers. In most African markets, 40-60% of volume moves through customers who do not have apps, do not have consistent data, and pay cash. If your programme requires a download, a login, or a data connection, the majority of your market is invisible to you. The fix is channel, not content. WhatsApp, USSD, SMS, and card-swipe at till. Start with the channel your customers already use every day.

Coverage — Mid Tier (3-5)

You reach a meaningful segment but the majority does not participate. The question is why. Is it awareness, access, or value? If customers know about the programme but do not join, the offer is not compelling enough for the effort required. If they do not know, distribution is the bottleneck. Dis-Chem solved this with a physical card handed to every customer. Shoprite solved it with a card that is free and instant. The activation moment matters as much as the reward.

Coverage — High Tier (6-8)

Your coverage is strong. The next question is depth. What percentage of total transactions flow through identified, programme-linked customers? Clicks is at 82.6%. Shoprite is at 88%. The difference between 60% and 85% coverage is the difference between useful data and a competitive data moat. Push for identification on every transaction.

Relevance — Low Tier (0-2)

Your rewards do not match what your customers actually need this week. Points, discounts, and lucky draws are the default because they are easy to set up. But they do not change behaviour because they do not change someone's day. In mass markets across Africa, what people need is airtime, data, groceries, fuel, transport money, and school-related costs. A R5 guaranteed dining voucher solves a real problem. A 2% discount on your own product does not feel like a reward. It feels like a price correction the customer suspects you owed them anyway.

Relevance — Mid Tier (3-5)

You offer some relevant rewards but the programme still leans on discounts or points as the core currency. The risk is that customers engage with the lifestyle rewards and ignore the rest, which tells you exactly what they value. Follow that signal. Increase the proportion of guaranteed lifestyle rewards. Reduce the points and discount mechanics. Track which reward categories drive the highest repeat rate. That data tells you what your customers actually care about.

Relevance — High Tier (6-8)

Your rewards are relevant. The optimisation opportunity is personalisation. Not everyone needs groceries. Not everyone needs fuel. The data from redemption tells you which reward matters to which segment. Use that to route the right reward to the right person. Clicks does this with ‘My ClubCard Deals’ refreshed every two weeks based on individual purchase history. That personalisation is what pushes relevance from good to exceptional.

Delivery — Low Tier (0-2)

Your delivery mechanism is the biggest bottleneck in your programme. A reward delivered three weeks after the qualifying action does not create a habit. It creates a pleasant surprise that the customer does not connect to the behaviour you wanted to reinforce. The neuroscience is clear: reward delay kills habit formation. Move to instant delivery. WhatsApp voucher within 60 seconds. Discount on the till slip. Airtime loaded directly. No portal. No redemption step. No waiting. That single change will improve programme performance more than any other.

Delivery — Mid Tier (3-5)

Some of your rewards are instant, others are delayed. The delayed ones are dragging down the overall programme. Customers who experience instant delivery have a different relationship with the programme than those who wait. Map every reward in your programme and mark it: instant, same day, or delayed. Then work to move every delayed mechanic to same-day, and every same-day to instant. Each step you compress increases the behavioural reinforcement per rand spent.

Delivery — High Tier (6-8)

Your delivery is fast and low-friction. The next frontier is zero-step delivery. Not ‘click here to redeem’ but automatic: the reward appears without any action from the customer. Dis-Chem discount on the till slip is zero-step. Clicks cashback loading on the 20th is zero-step. The customer does not redeem. They receive. That distinction matters because it removes the last friction point between the programme and the behaviour loop.

Measurement — Low Tier (0-2)

You cannot currently prove that your programme works. This is the gap that kills programmes in budget season. Without incrementality data, every review becomes a debate about feelings. Start measuring three things immediately: redemption rate (what percentage of issued rewards are actually used), cost per behaviour change (total programme cost divided by number of customers whose behaviour measurably changed), and incremental revenue per rewarded customer versus a control group. You do not need a data science team. You need a spreadsheet and a willingness to look at the numbers honestly.

Measurement — Mid Tier (3-5)

You have some measurement but cannot fully isolate the programme's impact. The most common gap here is the lack of a control group. If you reward everyone, you cannot prove the reward caused the behaviour. Start holding back 10-15% of qualifying customers as a control group for your next campaign. Compare their behaviour to the rewarded group. That comparison is the difference between ‘sales went up’ and ‘the programme caused a X% increase.’ The second statement is the one that protects your budget.

Measurement — High Tier (6-8)

Your measurement is strong. The next level is predictive: using historical cost-per-behaviour-change data to forecast the ROI of new campaigns before they launch. If you know that a R5 grocery reward drives a R22 incremental basket in your category, you can model the economics of any new campaign with confidence. That predictive capability is what turns a rewards programme from a marketing expense into a finance tool.

Cost of Doing Nothing — By Dimension

Intent (R0)

If your programme has no defined behaviour target, every rand spent is a donation to activity that may or may not move a commercial metric. At R500,000 per year, that is R500,000 in spend with no causal link to revenue. The Dis-Chem comparison: R410 million returned to customers, 10.4% revenue growth, 550,000 new shoppers. Every rand had a defined job. The gap between those two approaches compounds every month the programme runs without a behaviour target.

Coverage (60%+)

If your programme reaches 30% of customers, 70% of your base is making every purchase decision without any influence from your rewards investment. In a market like South Africa where 77% of consumers say loyalty programmes influence where they shop for groceries, the customers you cannot reach are the ones your competitor is reaching. Every month without mass-market coverage is a month where competitors are building habits you will have to break later.

Relevance (99.9%)

When 99.9% of participants in a lucky draw receive nothing, the only impression your brand leaves is a broken promise. The ones who received nothing do not think ‘better luck next time.’ They think ‘that brand promised something and gave me nothing.’ Over multiple campaigns, this creates the opposite of loyalty. It creates trained cynicism. The cost is not just the budget. It is the trust erosion that makes your next campaign harder to land.

Delivery (3 weeks)

A reward delivered three weeks after the action sits in a dead zone where the customer has already forgotten why they received it. The behaviour you wanted to reinforce has moved on. The customer has already bought from someone else, already recharged on a different network, already filled their script at a different pharmacy. Every day of delay is a day your competitor can insert themselves between the behaviour and the reward. Instant delivery is not a feature. It is the mechanism that makes everything else work.

Measurement (R0 proven)

A programme you cannot measure is a programme you cannot defend. In the next budget cycle, when the CFO asks what the rewards spend actually delivered, the answer cannot be ‘engagement went up’ or ‘customers liked it.’ Discovery can say: Diamond members have 57% lower mortality, which funds R2.4 billion in payouts, which grew normalised profit 29%. That clarity is what turns a cost centre into a growth engine. Without it, your programme is one tough quarter away from being cut.