How much of your promo spend actually drives repeat purchase?

FMCG Rewards Assessment — Full Reference Content

Dimension Insights

Intent — Campaign Proof (R8.26M)

Revenue generated from a R350K guaranteed rewards budget. 70,000 rewards issued. Every single participant received something.

Not one winner and 69,999 losers. 70,000 data points. 70,000 reasons to repeat.

Coverage — Informal Blind Spot (40-60%)

Of FMCG volume in most African markets moves through informal trade. Spaza shops, street vendors, informal wholesalers.

If your programme requires a smartphone or data connection, you are invisible to more than half your volume.

Relevance — Supplier Proof (20.9%)

Volume growth achieved by participating brands in Dis-Chem’s Better Rewards programme. Supplier-funded instant discounts.

The same co-funding model applies to FMCG. Brands fund the reward because they get measurable access to identified buyers.

Delivery — Speed Gap (5 min)

From merchandiser task completion to reward in hand. Shelf compliance verified, photo uploaded, airtime delivered.

The gap between effort and recognition is where field team motivation dies. Five minutes keeps the behaviour loop tight.

Measurement — Unit Economics (R5)

A R59 product at 25% margin. Buy-2 action gives R118 revenue, R29.50 gross profit. A R5 dining pass leaves R24.50. That is 83% margin preserved.

A behaviour driven. A data point captured. No discount. No erosion. No draw.

Recommendations by Dimension and Tier

Intent — Low Tier (0-2)

Your promotions lack a clear behavioural goal. You are spending money on promotions without defining the specific behaviour you want repeated. This means you cannot measure success, and you cannot improve what you do not measure. The first step is simple: pick one behaviour per campaign. Not awareness. Not engagement. A specific, countable action. Buy again within 14 days. Visit a second store. Add a second product to basket. Every rand you spend should trace back to that one action.

Intent — Mid Tier (3-5)

Your intent is partially defined but not sharp enough. You have some sense of what your promotions should achieve, but the goal is likely too broad. Volume increase is not a behaviour. Repeat purchase within 14 days is. Trial by lapsed buyers in specific regions is. Sharpen the intent, and the rest of the programme design becomes easier. The reward type, channel, timing, and measurement all follow from a clearly defined behaviour.

Intent — High Tier (6-8)

Your promotional intent is well-defined. You are already designing promotions around specific behaviours. The opportunity now is to layer multiple behavioural triggers. First purchase, second purchase, referral, product combination. Each one tracked separately, each one rewarded differently.

Coverage — Low Tier (0-2)

99.9% of your participants get nothing. This is the most expensive mistake in FMCG promotions. You spend R500,000 on a lucky draw. One person wins. 99,999 people walk away with nothing. No data. No reason to buy again. No behaviour change. Now compare: the same R500,000 buys 100,000 guaranteed R5 lifestyle rewards. Every participant gets something. Every participant becomes a data point. Every participant has a reason to repeat. One approach creates a single moment of excitement. The other creates 100,000 behaviour loops.

Coverage — Mid Tier (3-5)

You reward some participants, but most still get nothing. You are better than a pure lucky draw, but the gap between winners and losers is still too wide. The participants who get nothing are the ones you need most. They are the ones whose behaviour you are trying to change. Consider a tiered approach: everyone gets a base reward (R5 airtime or a dining voucher), top performers get something extra. The base reward costs almost nothing but keeps everyone in the behaviour loop.

Coverage — High Tier (6-8)

Your coverage model is strong. You are already rewarding the majority of participants. The next step is optimising the reward mix. Are you using lifestyle rewards that cost you R5 but feel like R30 to the recipient? Co-funded rewards from partners can double your effective coverage without increasing spend.

Relevance — Low Tier (0-2)

Your rewards do not match what your audience values. You are giving people things they do not want, or things that feel generic. Branded merchandise sits in a drawer. Points expire unused. Discounts on the next purchase assume someone was already planning to buy. A R10 airtime voucher delivered to a merchandiser’s phone within 5 minutes of completing a shelf check is worth more than a R100 gift card posted to their home 6 weeks later. Relevance is not about the rand value. It is about whether the reward feels like it was chosen for that person, at that moment.

Relevance — Mid Tier (3-5)

Your rewards are functional but not optimised. You are using rewards that make sense in theory but may not match what your specific audience actually wants. Have you asked them? Have you tested different reward types and measured which one drives the most repeat behaviour? Run a simple A/B test: offer one group airtime, another group grocery vouchers, a third group fuel. Measure which group repeats fastest. The answer might surprise you.

Relevance — High Tier (6-8)

Your reward relevance is strong. You have tested and you know what works. The opportunity now is personalisation at scale. Different regions, roles, or customer segments may respond to different reward types. A merchandiser in Limpopo may value data more than dining. A store owner in Durban may prefer fuel over airtime.

Delivery — Low Tier (0-2)

The gap between action and reward is killing your results. If someone does the thing you asked and then waits weeks or months for a reward, the behavioural link is broken. They do not associate the action with the reward. You get compliance without habit formation. Behavioural science is clear on this: the reward must arrive while the action is still fresh. Within minutes, not days. The technology exists. WhatsApp delivery takes seconds. USSD works on any phone. The only reason for delay is operational, not technical.

Delivery — Mid Tier (3-5)

Your delivery is reasonable but not instant. 24-48 hours is better than monthly, but it is still too slow for maximum behavioural impact. Every hour of delay weakens the association. If a merchandiser completes a shelf check at 10am and gets rewarded at 10:05am, the behaviour becomes self-reinforcing. If they get rewarded the next day, it feels like payroll, not recognition.

Delivery — High Tier (6-8)

Your delivery speed is a competitive advantage. Real-time or near-real-time delivery puts you ahead of 90% of FMCG programmes in Africa. Protect this advantage. Make sure your delivery channel matches your audience. WhatsApp for smartphone users, USSD for feature phones, airtime top-up for the broadest reach.

Measurement — Low Tier (0-2)

You are flying blind. You cannot calculate cost per behaviour change. You do not know which participants repeated, which lapsed, which rewards drove the most action. This means every promotional rand is a guess. You are not running a programme. You are running an expense. The fix is not complicated. Capture four data points per interaction: who did it, what they did, what they got, whether they redeemed it. That is enough to calculate ROI, optimise reward types, and prove value to the business.

Measurement — Mid Tier (3-5)

You have some data, but gaps remain. You track aggregate performance, but you cannot trace a specific reward to a specific behaviour for a specific person. This means you can see whether a campaign worked, but you cannot see why. Closing this gap requires participant-level tracking. When you know that Participant 47,382 completed Action X, received Reward Y, and then repeated Action X within 12 days, you have a real optimisation engine.

Measurement — High Tier (6-8)

Your measurement foundation is strong. You know cost per behaviour change and you track individual-level data. The next level is predictive: which participants are most likely to lapse, which reward type will retain them, which regions need different treatment. This is where the data becomes a strategic asset.

Cost of Doing Nothing — By Dimension

Intent (72%)

McKinsey found that up to 72% of trade promotions are unprofitable. The root cause is the same: no defined behaviour target. At your scale, every month without a clear behaviour goal is a month of promotional spend with no causal link to repeat purchase.

Coverage (99.9%)

When 99.9% of participants in a lucky draw receive nothing, the only impression your brand leaves is a broken promise. The same R500,000 could fund 100,000 guaranteed lifestyle rewards at R5 each. 100,000 people feel something. 100,000 data points captured. The economics are not close.

Relevance (40-60%)

If your programme requires a smartphone or data connection, you are invisible to 40-60% of your volume in most African markets. Spaza shops, street vendors, and informal wholesalers have a phone and WhatsApp. That is enough. The brands reaching them now are building distribution data your quarterly audit cannot match.

Delivery (5 min vs 6 weeks)

A merchandiser who gets R30 in airtime within 5 minutes of submitting a verified shelf photo will repeat the behaviour at the next store. A merchandiser who gets a cash bonus in payroll 6 weeks later does not connect the money to the action. The gap between action and reward is where motivation dies.

Measurement (R0 proven)

A programme you cannot measure is a programme you cannot defend. In the next budget cycle, the question will not be whether you ran promotions. It will be what the promotions changed. If you cannot answer with a number, the budget is at risk.