A Win Back Campaign in One Week: A Worked Example

Doubt is fair. Many people read about a win back campaign and wonder whether it works outside a demo. This post does not claim a result. It does something more useful: it shows how a small business could get a win back campaign running in one week, day by day, with an example that is made up and labelled as such.

Everything about the example shop below is illustrative. The shop, the numbers and the outcomes are invented to show the steps, and nothing here is a measured result or a benchmark. Lesson 5.6 walks you through it step by step: Win back customers who went quiet.

In a week you can define who went quiet, write three emails, set a stop rule, send, and read the first results. You cannot know the outcome in advance, so plan to measure it.

The example shop and its win back campaign

Say a small bakery sells boxed cakes online. It has a few hundred past customers. Most buy for birthdays and holidays, so the owner expects repeat orders a few times a year. The owner wants to message people who ordered once or twice and then went quiet.

For the sake of the example, assume 120 customers have not ordered for more than 90 days. That number is invented to keep the maths simple.

Day 1: Decide who counts as quiet

The owner looks at how long repeat customers usually wait between orders. Klaviyo suggests averaging the time between orders for customers with at least two purchases in the last two years. Cake buyers order every few months, so 90 days looks too short to the owner, who picks about four months instead.

This is the first lesson of the example: the cutoff is a judgement about your own buying cycle. Mailchimp says there is no set number of days after which a customer counts as inactive.

Day 2: Build and clean the list

The owner exports the customer list, filters by last order date, and removes anyone who has ordered since, who unsubscribed, or who had a complaint last year. Then the owner splits the rest into repeat buyers and one time buyers. Mailchimp recommends treating these groups differently.

The count is written at the top of a notes file. In the example it is, say, 90 people, with 25 repeat and 65 one time buyers.

Day 3: Write the three emails

The owner writes three short emails, each addressed to one imagined customer.

1. Email 1, a check in"It has been a while. Here are two new boxes this season, and a reminder you can order for a date coming up." Nothing pushy.
2. Email 2, a small offer"Free delivery on your next box until Friday." Real end date, stated once.
3. Email 3, a last call"This is the last note from me on this. If you would rather not hear from us, the link below takes you off the list. If you have a minute, tell us what we could do better."

This follows the three email pattern that both Klaviyo and Mailchimp describe, with a last email that includes an unsubscribe link and, in Mailchimp's version, an option to ask for feedback.

Day 4: Set the rules and test

Before sending, the owner checks the legal and sender basics.

• Each email has an accurate sender, a subject line that matches the content, a physical postal address and a clear opt out, as the FTC's CAN-SPAM guide requires.
• Unsubscribes will be honoured quickly. Gmail's guidelines ask for 48 hours, and Yahoo's asks for two days.
• Anyone who orders, replies or unsubscribes is removed from the rest of the series. Klaviyo's setup does this with a filter on orders since the flow started.

The owner sends all three emails to themselves and one friend to check links, the offer wording and how they look on a phone.

Day 5: Send email one

The owner sends the first email to the cleaned list. After sending, the owner does nothing for two days except answer replies. Replies are the most useful early signal, because they tell you in the customer's words what is going on.

Day 6: Read the early signals

The owner looks at replies and unsubscribes. In an illustrative scenario, a few people reply to say they moved or no longer need cakes. That is useful: it shows that some of the "quiet" group were never coming back, and that the quiet window might need adjusting. Nothing to change yet. Changing mid campaign makes the results hard to read.

Day 7: Plan emails two and three

The owner reviews replies, updates the list to remove anyone who ordered, and schedules the next two emails a few days apart. Then the owner writes down the plan for the number check: count orders, replies, unsubscribes and complaints from this group, and compare repeat buyers against one time buyers.

What the example shows, and what it cannot

It shows that a week is enough to get the first email out with a clean list and clear rules. It cannot show you how many people will buy. That depends on your product, your list and your offer. Treat any article that gives you exact numbers for "what you will get" with caution, including ones that claim a case study without showing its source.

How to adapt it to your week

• If you have few customers, shrink the plan and email by hand. See win back campaign vs the manual way.
• If your cycle is longer, spread the days. The order stays the same.
• If you want ready wording, use the free template.

Questions people ask

Is the bakery a real case?

No. It is an invented example to show the steps. The numbers are made up and are not results.

Can I really do this in a week?

The setup, yes, for a small list. The outcome depends on your own customers and cannot be promised.

What should I measure?

Orders from the group, replies, unsubscribes and complaints. Treat opens as a weak signal.

Where to go next

If you want this week run for you step by step, on your own customer data, take Lesson 5.6, Win back customers who went quiet. It covers finding people who stopped buying, writing the note in your voice, sending a small offer to the right people only and seeing who came back.