Repurchase program: step-by-step implementation checklist

A repurchase program is flow engineering, not generosity

The confusion in restaurants is thinking repurchase = discount. It does not. A discount is a PRICE REDUCTION applied to everyone; repurchase is REACTIVATION based on an OBSERVED ABSENCE. A customer who ate with you 2 years ago is not the same as one who ate 40 days ago and disappeared. The second is at risk; the first is out of reach. Masterestaurant has documented that restaurants segmenting by 'measured inactivity' (30+ days without transaction) reactivate 24–28% of that segment in 90 days, vs 8–12% with generic discounts. The difference is CRITERION: if you define well who is at risk, the incentive you send is a SIGNAL, not a gift. That is why the program is a machine: it reads operational data every day, executes a decision every week, adjusts every month. That is not improvised marketing; that is REVENUE ENGINEERING. A customer who visited weekly with USD 20 ticket is DIFFERENT from one who came every 2 months with USD 8 ticket.

Margin and segmentation are twins: they do not separate

The first, if gone, is at risk of competitor or habit shift — deserves strong outreach. The second is seasonal or occasional; reactivation requires less pressure. But there is a second filter: MARGIN. If that low-frequency customer, when they came, ate low-margin dishes (10–15%), the cost of reactivation with incentive can erode it completely. That is why the rule is SEGMENTATION BY FREQUENCY + MINIMUM MARGIN. SATE Institute, in dialogue with multilateral banks and development agencies, insists on this because a restaurant that does NOT KNOW the margin of its customers is a microenterprise WITHOUT INCOME CONTROL — and that is credit risk. Measuring the program per customer is FORMALIZATION. Email has open rate 15–25%; SMS has 40–60%; WhatsApp, if customer is on your list, 65–80%. The secret is CHANNEL AVAILABILITY, not quantity of channels. If you don't have customer's phone, SMS is impossible. If you don't have them on WhatsApp, that channel is zero.

SMS and WhatsApp: the channel people open (when it matters)

That is why the program detects: do I have SMS available for this customer? Do I have WhatsApp? And PRIORITIZES the one with higher open rate. Sending is nearly free (USD 0.01–0.05 per SMS in volume), so the program cost lives in the INCENTIVE, not communication. When customer sees the message and the offer is specific ('you visited us 4 times last year, why wait — USD 3 beverage for today'), redemption rate climbs to 40–50%. That is PERSONALIZATION THROUGH DATA, not flooding social media with ads. A program you see numbers on EVERY WEEK adjusts every 30 days; one you see every year already failed in months 2–9 without you knowing. SATE Institute underscores this for multilateral banks: a restaurant that MEASURES incremental revenue weekly is 3x more likely to endure than one that measures annually. The reason is simple: if your incentive redemption is at 15% (bad) versus 40% (good), in week 2 you already see it and in week 3 you change the incentive.

Weekly measurement is what turns repurchase into an income machine

Waiting until month 12 is leaving money on the table and eroding confidence in the program. That is why the method requires DASHBOARD: not 'reports', but LIVE ones, read every Monday what happened. If you see redemption drop, Tuesday you escalate the incentive. If you see rise, Tuesday you expand to another segment. That is OPERATIONS, not hope. Maintaining a program manually — reviewing Excel, segmenting with vague criteria, sending messages one by one — is an operation that costs USD 200–400/month in staff hours. An automatic flow reading POS daily, executing inactivity rules, and sending SMS costs USD 30–80/month in tool and ZERO in staff hours. For a small restaurant, that DOUBLES the program margin without doing anything. For a chain or franchise, it is SCALE INCREASE WITHOUT OPERATIONS COST. That is why Masterestaurant integrates repurchase into its Canvas and Dashboard: once you write the rule ('if this customer is inactive 30+ days, send Segment A SMS on Tuesday at 10am'), the engine runs it daily.

Automation removes fixed cost and speeds decision: it is the scale multiplier

The owner's decision is only MEASUREMENT and ADJUSTMENT, once monthly. That is engineering: machine + human decision, not machine = machine. For development agencies, multilateral banks, and MIPYME cartera investors, a restaurant with a FORMALIZED repurchase program is a DIFFERENT credit subject than one without it. Why: it generates verifiable operational data (customer X ate Y times, was reactivated with incentive Z, returned in N days), that data predicts future cash flow, and that reduces credit risk. The Inter-American Development Bank has shown that gastronomic companies measuring customer retention with formal criteria have 34% lower default probability in cartera. That is not marketing: that is RISK MANAGEMENT. That is why SATE Institute places this checklist as a cartera structuring tool — not to sell software, but to FORMALIZE the business. A repurchase program implemented per this method is proof the owner understands the business in data, not intuition. And that is what credit officers read when they say 'yes' to a line.

The 5 differences that drive measurable return

**SEGMENTATION BEFORE DISCOUNT:** A customer who has eaten 8 times in 2 months with average USD 18 ticket does not deserve the same incentive as one who ate 1 time 6 months ago with USD 8 ticket. The first needs a circumstantial reason (vacation, habit shift); the second needs REDISCOVERY. The offer is not the same. **MEASURABLE INACTIVITY CRITERION:** Not opinion ('seems like they don't come'). It is date + number: 30 days without POS, or 45 days if low-ticket customer. OBJECTIVE DATA is what enters an automation schedule. **INCENTIVE CALIBRATED TO MARGIN:** Offering 10% discount on an USD 8 plate (30% margin) costs USD 0.80 gross. Offering USD 2 on beverage (70% margin) if customer takes it is BETTER BUSINESS — incentivizes what leaves margin. Not all incentives are equal in cash impact. **AUTOMATION IS SCALE WITHOUT FIXED COST:** Reviewing an Excel and sending messages by hand costs time (USD 150–300/month in staff hours).

The 5 differences that drive measurable return — in practice

An automatic flow that reads POS daily and sends SMS/WhatsApp costs USD 30–80/month in tool + zero hours. The ROI of the program is not just revenue: it is operational efficiency. **WEEKLY VS ANNUAL MEASUREMENT:** A program you see numbers on every week adjusts every 30 days. A program you review once a year already failed in months 2–9. The method that works is agile: if redemption dropped, you change the incentive; if it rose, you scale. That is ENGINEERING, not hope.