When should a sauna studio send a win-back email?

Send a win-back email when a bather's own routine has stopped and you have a real reason for them to come back. "No visit in 30 days" picks out who to look at. It does not explain why each person on that list has stayed away.

Zenamu, which sells booking software to studios, says the right threshold depends on the studio's rhythm, and suggests 60 to 90 days for weekly classes.¹ Even a well-chosen threshold is still one number for every bather. It treats a Saturday regular who has missed four weekends and a bather who comes every six weeks the same way, and they can cross it on the same day. Read their history before you decide whether they need a message, an offer or nothing at all.

Compare the gap with the habit

Three hypothetical bathers on the same morning:

The threshold makes a review list, not a diagnosis. Travel, illness, cost, a timetable change or plain preference produce the same gap, and an email should not pretend to know which one applies.

Pick the threshold from your bathers' routines. A studio where most members come weekly can use a shorter gap than one where most visits are monthly. If your regulars split into weekly and monthly groups, the weekly group is the one a single threshold serves badly: by the time a monthly bather looks lapsed, a weekly one has been gone for a month.

Elara's win-back automation reads its threshold in whole days. It only considers bathers who have attended at least once, skips anyone with an upcoming booking (including a waitlist place or a booking awaiting approval) and only reaches people opted in to marketing email. It sends once when someone crosses the threshold, not every day after, and their next attended session resets the clock.

The At Risk list in reports uses a threshold of its own, 30 days by default, and shows bathers whose last visit falls between one and two times it, so 30 to 60 days ago. Set the win-back threshold to the same number and an eligible bather usually gets the email the day they first appear on that list. With a different threshold, the two cover different windows, so read them side by side with that in mind. When you first turn the automation on, it reaches back only a week, so bathers who lapsed earlier than that are not emailed.

Give the reason to return before the discount

A useful message tells the bather something new or removes a specific obstacle. If you have added a session that fixes a common complaint, say so. If nothing has changed, a short invitation to look at the schedule is enough.

When the timetable really changed

Subject: A later Wednesday session

"We've added a 7:30 p.m. session on Wednesdays. If the earlier times were hard to make, this one may fit your week. You can see open places and book from the schedule."

When nothing changed

Subject: Your next session

"If you'd like to come back for a round, the current schedule is ready to browse. If there is anything you need to check first, reply and the front desk will help."

Use the first only when the schedule changed, and the second only if replies reach someone who can answer. Skip "We know life got busy" and "We miss you": an attendance filter tells you neither.

Check each bather before adding an offer

An incentive can bring someone back, and it can also discount a session they were already going to book. Before you add one, open the customer profile: an upcoming booking, an unused visit pack or a membership they still pay for changes the right message. Do not offer a pack to someone who still holds credits.

If you do test an incentive, state what it buys, who can use it and when it expires. Judge it by contribution after the discount, not by redemptions.

Measure the return against a holdout

A booking that follows the email tells you the bather came back. It does not tell you whether they would have come back anyway. To estimate that, hold some bathers out before you send.

Elara has no holdout or exclusion control. The win-back automation emails everyone who qualifies, and a campaign audience can include customers by tag, membership status, visit frequency or spend but cannot leave anyone out. So run the test as a one-off campaign with the automation off, and split the group with tags:

1. On the Customers page, filter to At risk (lapsing), select everyone and add a tag such as winback-oct. This fixes the group, since the At Risk list changes every day.
2. Split that group at random. Elara will not do this for you: filter the Customers page by winback-oct, select everyone and choose Export CSV, then give each row a random number in a spreadsheet and take the lower half. Still filtered by winback-oct, tick the customers in that half and tag them winback-email. Everyone else in winback-oct is the holdout.
3. Send the campaign to winback-email only.
4. On the day the window you chose ends, filter the Customers page to one completed visit or more in that many days, once with winback-email and once with winback-oct. The first count is the email half's returns; the second less the first is the holdout's.

The campaign skips anyone in the email half who is not opted in to marketing email, has turned marketing email off or has an address suppressed after a bounce or complaint. The Customers page cannot filter by opt-in, so both halves will include people the email could never reach. The Reports customer export has an opt-in column, but you do not need it: count everyone in their group anyway. The comparison stays fair; it measures the effect across everyone assigned rather than only the reachable.

In this hypothetical test, 400 at-risk bathers are split at random: 200 are assigned the email and 200 receive nothing over the same 30 days. Each person counts once if they attend.

The difference is three percentage points. Applied to the 200 assigned the email, that estimates six additional returns, not 36.

Now price it. Say each additional visit contributes $30 before any discount, the email offers $5 off a session, every one of the 36 returning bathers redeems it once, and the campaign costs $120 to prepare and send. Six additional visits bring $180. The discounts given to all 36 cost $180, and the campaign $120, so the offer loses $120. The 30 bathers who would have come anyway still took the discount.

Six extra returns in groups of 200 can easily be noise, so treat one test as a hint, not a verdict. Decide the outcome and the window before sending, and count everyone in the group they were assigned to, including bounced emails. Comparing people who clicked with people who did not is not a holdout: clicking already signals interest. Don't stop the test early because the first week looks good. Keep the holdout away from the same offer on social posts or at the desk. And don't compare this month with last month instead: seasons move sauna attendance on their own.

Read Elara's attribution for what it is

Elara reports each campaign two ways. Direct attribution credits a booking made in the booking widget after arriving from a tracked link in the email. Assisted attribution notes a booking by someone who clicked a campaign link in the seven days before.² Both describe what happened after the email. Neither holds anyone out, so neither tells you how many of those bookings the email caused.

Close the loop

Measure the outcome you wanted. If it was a return visit, count attended sessions in a stated window; a booking that later cancels is not a return. If it was a new purchase, report it apart from sessions paid from an existing balance.

Read the replies too. If several bathers say the sessions start too early, that is a timetable question, not a reason for a bigger discount. Track unsubscribes and complaints alongside attendance: a campaign that brings back a few people while annoying the rest needs a different process, not a stronger subject line.

The email marketing page shows how the win-back automation and campaign reports work. Ask for a quote. When you set up, choose a win-back threshold from your own attendance pattern and read the At Risk list before you turn the automation on.