Why loyalty drives repeat business

A loyalty program is not a discount. It is a way of buying visits, at a price you set, from people who already know where you are. Whether that is worth doing is a question you can answer with arithmetic, and it is worth doing the arithmetic before you build anything.

All the numbers below are in whatever currency you take payment in.

What one more visit a month is worth

A café with 200 regulars, an average ticket of 5.00, each of them coming four times a month:

200 x 5.00 x 4 = 4,000 a month.

Now move those same 200 people to five visits a month. Nothing else changes.

200 x 5.00 x 5 = 5,000 a month.

That is 1,000 more a month, or 12,000 a year, out of the same shop with the same staff and the same opening hours. You did not find a single new customer.

Now the cost side. Say the program gives back 5% of spend in rewards. On 5,000 of monthly takings that is 250 of rewards at menu value, and what you actually hand over — a coffee, a pastry, a blow-dry — costs you a fraction of what it sells for. Even counting every reward at full menu price and giving yourself no credit for the margin, you are 750 a month ahead.

That is the entire case for loyalty in one calculation. Everything else in this category is about not getting the details wrong.

What it costs to replace a regular

Do this one with your own numbers rather than an industry statistic. Say you spend 200 on a promotion: flyers, a boosted post, a discount for first-timers. Forty new people come through the door, so that is 5.00 of marketing per head. Of those forty, eight come back a second time. You have just paid 25.00 for every customer you actually kept.

Compare that with the reward you hand a regular. It costs you a slice of margin, you pay it only when they come back, and you never pay it to someone who does not. Acquisition is paid in advance and in hope. Loyalty is paid in arrears and only on results.

This is not an argument for never advertising. It is an argument for noticing that the customers you already have are the cheapest ones you will ever serve, and that most businesses spend almost nothing on keeping them.

Frequency is the number that actually moves

There are only three ways to take more money: serve more customers, take a bigger ticket from each one, or see the same people more often. Most of what owners spend goes on the first, which is the hardest and most expensive of the three.

A loyalty program mostly moves the third, and it moves it because the hard work is already done. Your regular knows where you are, knows what they order, and has decided they like you. You are not persuading anyone of anything. You are giving someone a small reason to choose you on a day they might have walked past, or gone to the place with the shorter queue.

Frequency also compounds in a way a one-off promotion never does. One extra visit a month from a customer who stays two years is twenty-four extra visits out of a single decision.

Small and reliable beats big and rare

Two programs, both giving back exactly 5% of spend, both in that same café with a 5.00 ticket:

• Program A — 1 point per 1.00 spent, and a pastry worth 1.50 costs 30 points. Six visits.
• Program B — 1 point per 1.00 spent, and a hamper worth 30.00 costs 600 points. A hundred and twenty visits.

They cost you the same percentage of revenue. Only one of them changes anybody's behaviour.

People move faster when they can see the finish line, and a finish line a hundred and twenty visits away is invisible. Your customer does the sum once, decides the reward is theoretical, and stops thinking about it. Worse, the points they are holding sit on your books as a debt that never gets settled.

The working rule: your first reward should be reachable in about five to ten visits. If you want a big reward as well, put it above the small one rather than instead of it. A first rung people actually reach is what makes the rest of the ladder worth climbing.

What to watch once it is running

Your Overview has two tiles that answer the question this article opened with. Members is everyone who has ever joined. Active (30d) is how many of them did something in the last thirty days. Members climbing while Active (30d) stays flat means you are collecting sign-ups rather than visits, and the reward is almost always too far away.

Outstanding liability, the tile whose hint reads "Points customers still hold", is the other half of that picture. Liability rising while Redemptions stays flat is the same warning arriving from the other direction.

When you are ready to pick a mechanic, program types and how to choose one covers the options, and setting earn and redeem rates turns the percentage you just read about into the numbers you type into the builder.