Win-Back Campaigns: How Restaurants Bring Lapsed Guests Back

Most restaurants try to bring lost guests back the same way they reach everyone else: one email to the whole list. It feels efficient. It is also the weakest message a restaurant sends.

We compared win-back emails with blasts across Bloom network restaurants in the US and Canada from January 2025 to August 2026. Win-back emails brought back 6.5% of recipients, against 0.27% for blasts. This post covers how to spot a guest who is slipping, when to send, what to offer and how to measure the result. For the background on why guests leave, see our guides to restaurant customer churn and restaurant customer attrition.

What is a restaurant win-back campaign?

A restaurant win-back campaign is an automated message, usually an email, sent to a guest whose visits have slowed or stopped, with a reason to return. Unlike a blast, it goes to one guest because of what that guest did.

A blast goes to the whole list on the date you pick. A win-back message goes to one guest on the day their own behavior says they are drifting away. The trigger is the guest, not the calendar.

That difference shapes everything else. The message can name the guest, reflect how often they used to visit, and arrive while they still remember why they liked you. A blast can do none of that.

How do you know a guest is slipping?

Measure the gap since a guest’s last visit against that guest’s own visit pattern, not a fixed number of days for everyone. A weekly regular who misses 3 weeks is at risk well before a monthly guest who misses 3 weeks.

A single rule such as “email anyone who hasn’t visited in 60 days” treats every guest the same. It reaches your weekly regular far too late, after the habit has already broken. It also nags your monthly guest too early, before they are anywhere near gone. As our post on losing regular restaurant customers explains, regulars rarely announce that they are leaving. Their visits simply space out.

Spotting that change takes one profile per guest. Visits arrive from different places: the POS, guest WiFi and online ordering. If those stay in separate systems, a guest who switched from dining in to online ordering looks lost when they are not. A restaurant customer data platform joins those visits into one profile per guest, learns each guest’s normal rhythm and flags the ones who fall behind it.

Which win-back emails work?

Win-back emails open less often than email blasts, but they bring far more guests back. In Bloom network data, win-back emails returned 6.5% of recipients against 0.27% for blasts, and earned $1,106 per 1,000 emails against $80.

Look at the first row. Win-back emails open less often than blasts, 13.3% against 19.5%. A team that grades email by open rate would call the win-back the loser and cut it. Then look at the third row. The win-back brought guests back 24 times more often. Open rates measure curiosity. Return visits measure revenue.

One caution keeps the comparison honest. Win-back emails go to guests who already have a visit history, so part of the gap reflects who receives them. The full Restaurant Email Marketing Benchmarks 2026 report explains the method and compares every campaign type.

When should you send a win-back email?

Send a win-back email when a guest goes noticeably longer than usual between visits. The earlier fix is a welcome series for first-time guests, because most lost guests never make a second visit.

Timing is about the gap since the last visit, nothing else. A guest who normally visits every 10 days and is now at 25 needs a message. A guest who normally visits every 6 weeks and is at 25 days does not, yet. Let each guest’s pattern set the clock.

The biggest churn happens before a guest ever becomes a regular. In Bloom network data, 8 in 10 first-time guests never return. A welcome series of 2 or 3 messages over a guest’s first 14 days gives that guest a reason to come back while the first visit is fresh. A guest who makes a second visit is far more likely to keep coming, so the welcome series stops churn before a win-back is ever needed.

What should a win-back offer be?

Lead with a personal note, then add an offer only if the note alone is not enough. Keep any offer simple and time-limited, so the guest has one clear reason to return soon.

A guest who used to visit often already likes the food. Often they got busy, or tried somewhere new. A note that says you noticed, and that you would like to see them again, does real work before any discount enters the picture.

• Make it personal. Use the guest’s name, and mention what they usually order if you know it.
• Keep the offer simple. One clear item or one clear amount. No fine print.
• Give it an end date. A short window turns “someday” into a plan.
• Follow up once. If the first message gets no visit, a second, warmer message can follow. Then stop.

There is no universal right discount, and we won’t invent one. Test a note with no offer against a note with one, and keep whichever brings more guests back. Our restaurant email marketing guide has more examples of messages that work.

How do you measure a win-back campaign?

Count the guests who came back and the revenue from those visits, not opens or clicks. The table above shows why: the email with the lower open rate earned more than 13 times the revenue per 1,000 sends.

Measuring return visits means linking each message to the guest who got it, then to that guest’s next visit and check. Email tools stop at the click, so they can’t see the visit. Restaurant marketing attribution closes that gap by matching the guest who received the message to the visit that followed and the money spent.

Track three numbers for each win-back campaign: the share of recipients who returned, the revenue from those returns, and how many of those guests kept visiting afterward. The last one tells you whether you won the guest back or only won one visit.

Can a win-back campaign run on its own?

Yes. Bloom Intelligence finds at-risk guests and sends win-back messages automatically, and recovers 38% of at-risk guests. No one has to pull a list or pick a send date.

Bloom’s AI restaurant marketing automation watches each guest’s visit pattern in the Bloom customer data platform. When a guest falls behind their usual rhythm, Bloom sends a win-back message written in your brand voice, then tracks whether that guest comes back and what they spend. Welcome series for first-time guests run the same way.

Restaurants on Bloom recover $53,000+ per location in year one, and the win-back program runs every day without a campaign manager.

A blast asks everyone to come back. A win-back asks the one guest who stopped.

Methodology

Email figures are original Bloom Intelligence data from the Restaurant Email Marketing Benchmarks 2026 report. The data covers Bloom network restaurants in the United States and Canada, for campaigns starting January 1, 2025 to August 31, 2026. The sample is more than 3,000 blast campaigns plus hundreds of automated campaigns. Every rate is measured per email delivered. A return is a recipient who was recognized on a visit after the send. Revenue per 1,000 emails is return visits multiplied by the returning guests’ observed spend. Win-back emails go to guests with a visit history, so the comparison with blasts includes a selection effect. The 8 in 10 first-time guest figure comes from Bloom’s published first-time guest research.

Results come from the Bloom network and may differ for your restaurant. Research and analysis for this article were produced with Bloom Intelligence’s data platform and AI tools, then reviewed and edited by Allen Graves.

By Allen Graves, Director of Marketing, Bloom Intelligence. Allen leads website, content, marketing automation and AI search strategy at Bloom, after years as a restaurant server and bartender.