Restaurant sales growth plan: traditional method vs the Masterestaurant method

What does each plan measure: customers in, or customers back?

The traditional plan measures the mouth of the funnel and the Masterestaurant method measures the bottom, and that single difference decides who actually grows.

In the Bogotá case, 14 million pesos of paid media brought 1,180 new orders in one quarter and only 96 people came back: a return rate of 8.1% that forces you to replace nearly your whole base every year. Acquisition cost landed at 11,864 pesos per guest against a contribution margin of 23,180 pesos on the first visit, so the campaign closed in the black by 11,316 pesos and not a peso more. The MR method takes that same budget and splits it 60/40 between acquisition and repeat business, because a returning guest costs zero to acquire. MR wins this criterion outright: retention turns a quarterly expense into an asset that keeps billing after you switch the ads off. This is where the comparison turns brutal, because the two curves cross.

Acquisition cost climbs; your average check does not keep up

Restaurant input costs rose 35% in food and 35% in labor since 2019 according to the National Restaurant Association (2024), and menu prices at large U.S. chains climbed 42% between 2020 and 2025, almost double the 22% general inflation (One Haus). Under that arithmetic, the traditional plan needs your average check to grow at the pace of the ad auction, and it does not grow: the Bogotá operator has been stuck at 38,000 pesos for four straight quarters. The Masterestaurant method attacks the variable an owner actually controls, which is frequency. Moving from 1.4 to 2.6 quarterly visits lifts annual lifetime value from 129,800 to 241,100 pesos per guest, an 85.7% jump with no change to the menu or the price. Verdict: MR wins, because frequency is not auctioned. Once promotion becomes the engine of the plan, effective food cost drifts away from the recipe and nobody notices until month-end.

Permanent discounting is borrowed volume and sold margin

With 22% of transactions running under discount, a menu costed at 29% truly operates at 38%, nine percentage points taken straight out of contribution margin. The Masterestaurant costing contract sets 32% as the CEILING per dish, never the target, and a plan that lives on promotion breaks that ceiling week after week. The traditional plan answers a traffic dip by cutting price, which works for one month and trains the customer for the next. The MR version uses discounts as a surgical lever for off-peak hours or high-rotation dishes, never as a permanent current. MR wins with one condition attached: if you cannot measure your promoted-transaction percentage every week, do not use the lever at all. Judge any serious growth plan by a single inventory: how many guests from ninety days ago sit in a database you can write to today. The Bogotá restaurant had zero, which is why every quarter restarted from the same line with a more expensive budget.

Your guest list versus a rented audience

The gap gets paid in owned channels: SMS drives 25% higher engagement in food and beverage (Tabular, 2025), and rewards programs raise value per customer by 23%, with 55% of operators reporting that member checks grew faster than their own menu prices (Paytronix Loyalty Trends 2024). The traditional plan rents a platform's audience and pays rising rent forever. The MR method builds the list and works it. MR wins, and the decision rule is plain enough: paid media is a faucet, your database is a well. Reviews pay better than paid media, and the average owner treats them as a complaints inbox. Every additional star in your rating is worth between 5% and 9% of revenue according to Michael Luca's Harvard Business School study of Yelp, a return that neither expires nor gets bid up, whereas a creator post lifts reservations 30% the following week before traffic settles back to baseline (Marketing LTB, 2025).

Reputation and creators: two levers with very different returns

The traditional plan chases that 30% spike because it shows up on a calendar; the Masterestaurant method buys the rating points first, since they are permanent, and brings in creators as a timed accelerant once the operation can hold the extra covers. No middle ground on this verdict: if your rating sits below 4.3, hiring creators means paying to fill a room with reviews that will scare off the next table. Back to that March statement, because the complete arithmetic is the strongest defense of this verdict. With 1,180 new guests, 1.4 quarterly visits and 23,180 pesos of margin per visit, the traditional plan yields roughly 129,800 pesos of annual value per customer and burns 11,864 acquiring them: 117,936 pesos of contribution remain, provided you repeat the 14-million investment every three months. Reallocating that same budget toward repeat business and lifting frequency to 2.6 visits pushes annual value to 241,100 pesos while acquisition cost spreads across a base you no longer replace wholesale.

The full case: same cash register, two plans, twelve months

The difference per guest sits around 111,300 pesos. Multiply by the 1,180 guests of a single quarter and the money left on the table comfortably clears 131 million a year. Assume the auction gets 100% more expensive next quarter, something that has hit entire categories without warning. The traditional plan walks into an impossible choice: holding those 1,180 new orders demands 28 million pesos and margin per acquired customer drops from 11,316 to below zero, or you halve the volume and bill half. There is no third door, because all your demand is rented. The restaurant that spent eighteen months building a base takes the same blow with half the skin exposed: its repeat traffic does not depend on the auction, so it absorbs the increase by trimming acquisition without losing sales. That is the real tension of the trade, and it resolves cleanly: paid media buys speed, repeat business buys resilience, and whoever only buys speed is driving without brakes.

What to choose according to your operating profile?

Diego F. Parra sums it up in a rule that Masterestaurant applies before a single peso of budget moves: if you are already billing, choose the MR method without hesitation.

A restaurant past twelve months of operation, with a stable check and at least 800 monthly guests, already owns the asset the traditional plan ignores, and capturing it costs a fraction of what a new customer costs. If you just opened, the calculus differs: with no history you do need to buy three or four months of traffic with paid media, though that spend only makes sense if you capture contact and consumption data for every guest from day one. The line is that clear. Your first task this week is not picking an agency, it is exporting from your POS the list of everyone who ate with you in the last ninety days, and counting them. The traditional plan measures entry, the Masterestaurant method measures RETURN.

The four differences that move cash

A restaurant counting only new guests is watching the mouth of the funnel while the bottom leaks; at 1.4 quarterly visits you must replace nearly your whole base every year just to hold flat, and that replacement is paid at ad-market prices, which in 2026 rise faster than your average check. Discounting buys volume and sells margin. When 22% of transactions carry a promotion, effective food cost jumps to 38% even with recipes costed at 29%, and the Masterestaurant costing rule is blunt: 32% per dish is the ceiling, not the target. Promotions have a place — as a lever for slow hours or high-rotation dishes — never as the permanent engine of growth. Vertical video is not brand decoration, it is acquisition infrastructure with a falling marginal cost. A scripted Reels series with a character — the line cook, the six a.m. mise en place, the supplier's story — still brings guests eighteen months after filming, whereas a paid ad ceases to exist the second the budget runs dry.

The four differences that move cash — in practice

That is the difference between buying traffic and building an asset. Ownership of the guest decides who controls the relationship. If last year's 4,000 guests live in an aggregator's database, you do not own a restaurant business, you run a production point for somebody else's; the MR method starts by recovering that list, because an owned-channel campaign at 34% open rate costs nothing in commission and answers within hours.