You have their number. Now what? Five campaigns worth running on your loyalty list

Most loyalty advice stops at signup. Get more members, the thinking goes, and good things follow.

They do not follow on their own. A list of two thousand guests that never gets used is worth exactly the same as a list of zero, and it costs you more, because you paid for the system that built it. The value shows up the first time you send something to a specific group of guests for a specific reason and watch what happens.

That does not require a marketing department. It requires five campaigns, set up once, that run off what your loyalty program already knows about your guests.

Start with a trigger, not a calendar

Here is the mistake almost everyone makes first: picking a day and blasting the whole list.

Tuesday comes, somebody writes a message, it goes to everyone. Some of those people were in yesterday. Some have not been in for a year. Some spend three hundred dollars a month with you and some came once on a Groupon. They all get the same email, and it performs the way you would expect.

A trigger works differently. Instead of a date, you pick something a guest does, or stops doing, and the message goes only to the people it describes. First visit and no second one. Six visits and then silence. A birthday next week.

The message is more relevant because the group is smaller, and you can tell whether it worked because you know exactly who got it. Both of those come from the same choice.

Everything below follows the same shape: who it goes to, when it goes, and the one number to watch afterward.

The five campaigns

1. Win back the guests you already lost

• Goes to: guests who used to come in regularly and have gone quiet. Something like six or more visits in the previous few months, and nothing in the last thirty or sixty days.
• When: monthly, to whoever has newly gone quiet since the last run.
• Watch: what share of the guests you contacted came back within thirty days.

Run this one first. These are not strangers who might like your food. They are people who liked it enough to come six times and then stopped, usually for a reason that has nothing to do with you: a new job, a move, a schedule change. A message is often enough.

Lead with the fact that you noticed, not with a discount. “We have not seen you in a while” does something a coupon does not, and it does not teach your regulars to wait for a deal before coming in. If you do include an offer, keep it small. These guests were happy paying full price.

2. Turn a first visit into a second one

• Goes to: guests with exactly one visit, seven to ten days after that visit.
• When: weekly.
• Watch: what share of one-visit guests come back within thirty days.

This is the biggest group on almost every list and the one nobody talks to. Whether a first-timer becomes a regular is mostly settled in the first few weeks, and after that the odds drop off fast.

Keep it short. Thank them, tell them one thing they probably did not discover on a first visit, and give them a reason to pick a specific next occasion. A weekday lunch or a slow daypart works well here, because you are filling a seat that would otherwise be empty.

3. Birthdays

• Goes to: guests with a birthday in the next week.
• When: five to seven days ahead, never the day of.
• Watch: redemption rate, and average check on the visits it produces.

The most reliable campaign in the business, and the easiest to run badly. Sending on the morning of somebody’s birthday is too late, because they made their plans a week ago. Send early enough to be part of the decision.

The other thing worth knowing: birthday guests rarely come alone. The check on a birthday visit is usually well above your average, which means a generous offer on one item is often the cheapest table you will fill all month. You are getting a party, not a cover.

4. Fill your slow daypart

• Goes to: guests who only ever come at your busy times.
• When: two or three times a season, not weekly.
• Watch: covers in that daypart, compared with the same weeks last year.

You already know which shifts are soft. What you may not know is that a good share of your regulars have simply never thought of you for that occasion. The dinner crowd does not know you do lunch. The weekday crowd has never been in on a Sunday.

This is the campaign that changes a habit rather than a transaction, so give it time and do not run it into the ground. Discounting a shift that was already busy is the one thing to avoid here, and targeting by when guests actually visit is what keeps you from doing it.

5. Recognize your best guests

• Goes to: your top guests by visits and by spend. Usually a small enough group to count.
• When: a few times a year, and never with the same message twice.
• Watch: whether that group’s visit frequency holds steady over the following quarter.

This one is not really a campaign, and it should not read like one. Early access to something, an invitation, a table held on a night you are slammed. The goal is that your best guests feel known, which is the thing that keeps them from drifting to the new place down the street.

The version of this that works best is not an email at all. It is a tag on the guest record in your POS, so whoever is working that night sees it when the check opens. Recognition on the floor beats recognition in an inbox.

Running them without wearing out your list

A few rules that keep this from turning into noise.

• Start with one. — Win-back, most likely. Get it running for a month before you add another. Five campaigns launched at once is five campaigns you cannot tell apart.
• Cap how often any guest hears from you. — Once or twice a month is plenty. Triggers overlap, and a guest who lands in three groups at once should not get three messages in a week.
• Do not lead with a discount every time. — If every message contains an offer, you have trained your regulars to wait for one. Recognition, news and an invitation all work without costing margin.
• Measure the group you contacted, not overall traffic. — Total covers move for weather, holidays and a hundred other reasons. The only honest read is what the people who got the message did next.
• Give it thirty days. — Restaurant visit cycles are measured in weeks. A campaign judged after four days has not been judged at all.

When the holidays come around, a seasonal nudge about gift cards is one more message that works without a discount, and e-gift cards in particular tend to come back worth more than their face value.

Where to start this week

Pull one list: guests who came in six or more times in the spring and have not been in for a month.

Look at how many there are. For most operators the number is bigger than expected, and every name on it is someone who already decided they like your restaurant. That list is the entire argument for doing any of this, and it costs nothing to look at.

Send them one message. Wait thirty days. Count how many came back.