Email and customers · Lesson 5.6
The WIN Framework for a Win Back Campaign
Doing a win back campaign by feel gives you uneven results. One month you pick the wrong people, the next you forget the unsubscribe line, and you never learn which change helped. A fixed routine fixes that.
This post gives you a simple routine called WIN: Who went quiet, Invite them back, Number check. The name is ours, made to be easy to remember. It is not an industry standard. Each part below draws on how email tools and sender rules actually work. Lesson 5.6 walks you through it step by step: Win back customers who went quiet.
Run the same three passes every time: choose who, invite them with one small offer, then check the numbers before you change anything.
Why a win back campaign framework helps
Without one, every campaign is a new decision, and you cannot compare this month with last month. With a routine, you can change one thing at a time. That is how you learn what works for your own customers, which no general article can tell you.
W: Who went quiet
The first pass is about picking people, because a perfect email to the wrong list does nothing.
- Set the windowBase it on your own buying cycle. Klaviyo suggests averaging the time between orders for repeat customers and setting the first delay a little longer than that. Mailchimp says plainly there is no universal number of days.
- Pull the listEveryone whose last order is older than the window. The lesson starts at 90 days.
- Clean itRemove people who bought since, who unsubscribed, who complained, or who have an open issue with you.
- Split itAt minimum, repeat buyers and one time buyers. Mailchimp suggests separating high value customers from one time shoppers.
- Write the count downHow many people are in each group. You need this number at the end.
A good check at the end of W: could you explain in one sentence why each person is on this list? "They bought in the spring and have not been back" is a reason. "They are on my list" is not.
I: Invite them back
The second pass is the message and the offer. Keep it modest.
One reason to return. Choose one: something new, a seasonal use, a small offer, or a simple question. Mailchimp's suggested pattern is a first email on value or updates, a second email with a concrete incentive, and a third that is a last call or a feedback request. Klaviyo's pattern also uses three emails, with an incentive early.
One small offer, with an honest end date. Free delivery, a small percentage off or a bonus item. If you say it ends on Friday, it ends on Friday. The FTC's CAN-SPAM guide warns against deceptive subject lines, so make sure the subject matches the contents.
Easy exit. Put an unsubscribe link in each email. Yahoo's sender requirements ask bulk senders to support one click unsubscribe and honour requests within two days. Gmail's subscription guidelines ask for the same one click method and a 48 hour limit.
A stop rule. If someone orders, replies or unsubscribes, they leave the series. Klaviyo's setup does this with a filter on orders since the flow started.
A cap. Three emails. Klaviyo recommends keeping a win back flow to three emails per recipient.
N: Number check
The third pass happens a week after the last email, before you plan anything new. Use a small fixed set of numbers.
| Number | How to get it | What it tells you |
|---|---|---|
| Group size | The count from W | The base for every percentage |
| Purchases | Orders from the group since the first send | Whether people came back |
| Replies | Count by hand | Why they left, in their words |
| Unsubscribes | From your tool | Whether the message felt unwelcome |
| Spam complaints | From your tool, if reported | Whether you are over reaching |
| Opens | From your tool | Weak signal only |
Why is opens a weak signal? Apple's Mail Privacy Protection can prevent senders from learning when and whether someone opened an email, and downloads remote content in the background regardless of engagement. So trust purchases and replies first.
On complaints: Gmail and Yahoo both state that you should keep your spam rate below 0.3%. If you see complaints rise on a win back send, treat that as a signal to send to fewer people next time.
Run WIN every time
After the number check, write three lines in a notes file:
- What I changed this time.
- What happened to purchases, replies and unsubscribes.
- What I will change next time, and only one thing.
That last line is the discipline. Change the window, or the offer, or the subject line, not all three. Over a few rounds, your own results will tell you more than any list of tips.
A short worked pass
Say a plant shop runs WIN with invented round numbers, for illustration only. Who: 80 people who last bought more than 90 days ago, split into 30 repeat buyers and 50 one time buyers. Invite: three emails, a care tip first, a small offer second, a last call third. Number check: the shop counts orders, replies and unsubscribes and compares the two groups. Next change: try the offer in the first email for only the one time buyers. These numbers are not a benchmark. They only show the shape of one pass.
Where it connects
WIN is the compact version of the longer process in the complete guide. When you want to know what to count, read Measuring a win back campaign. For mistakes that break each part, see 5 win back campaign mistakes.
Questions people ask
Is WIN an established method?
No. It is a simple name for a three step routine, built from how email tools and sender rules work.
How often should I run it?
As often as people go quiet in your business. Many shops find a monthly or quarterly pass fits. Check your own buying cycle.
Do I need all three parts?
Yes, because skipping the number check means you cannot learn from the campaign.
Where to go next
To run WIN on your own customer data, with the note written in your voice, work through Lesson 5.6, Win back customers who went quiet. It covers finding people who stopped buying, writing the note, sending a small offer to the right people only, and seeing who came back and what to try next.