- Latestly AI
- Posts
- 💰 Bill $1,500 to bring back customers your clients already lost
💰 Bill $1,500 to bring back customers your clients already lost
By Sunday night: a workflow that sorts a lost-customer list by why each customer left, plus a personal win-back email for each group.
Latestly AI
October 31, 2026
In partnership with
🎁 $1,000+ in free AI tools. Access Offers →
Become An AI Expert In Just 5 Minutes
If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.
This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.
Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.
🧭 Build brief
⚡ In 20 seconds
Build: a workflow that groups lost customers by reason and drafts one email per group.
Time: 2–3 hours to set up. About 1 hour per campaign after that (estimate).
Sell it: $1,000–$1,500 per campaign, with results you can measure.
By Sunday night: a sorted customer list, 4 email drafts and a tracking sheet.
What will you have by Sunday night? |
How does the system work? Input: a list of customers who stopped buying, with their history |
Who should build this? Email marketers and e-commerce freelancers |
What does it improve? Recovered revenue. These people already bought once. They know the brand. |
📦 Build card
💰 Business value | 🎯 Best for | ⏱ Build time |
⭐ Weekend playbook
Step 1 Define the input · 30 min
First, define “lost.” Use twice the normal buying gap. Three common cases:
Subscriptions: cancelled in the last 6 months
Online stores: no order in 180 days
B2B suppliers: no order in 90 days
Export these columns into Google Sheets:
Customer ID and first name
Last purchase date
Number of orders and total spent
Last product bought
Cancellation reason, if one exists
Last support note or complaint, if one exists
Remove anyone who unsubscribed or never agreed to marketing emails. Email rules differ by country. In the EU, consent rules are strict.
Step 2 Build the AI layer · 40 min
The AI does two jobs. First, it sorts. Then it writes.
Sorting into 5 groups:
Group | Evidence in the row | Best message |
|---|---|---|
Service problem | complaint, refund, late delivery | Apology and what you fixed. No discount first. |
Price | “too expensive,” downgrade | One specific offer the owner approves |
Product fit | wrong size, didn’t use it | A better-matched product |
Drifted away | no reason found | What’s new, plus an easy way back |
Moved on | “no longer need it” | Don’t sell. Ask for a referral or feedback. |
Rule: if the row has no evidence, the customer goes in Drifted away. The AI must never guess a reason.
Step 3 Connect the workflow · 30 min
Trigger → AI → Filter → Action → Storage
Trigger: export the lost-customer list once a month
AI: paste rows in batches of 50 into the prompt below
Filter: your top 10% of past spenders get a personal email from the owner, not a template
Action: load each group into the email tool as its own list or tag
Storage: a tracking sheet with group, send date, replied (yes/no), ordered (yes/no) and revenue
Step 4 Test and improve · 40 min
Check 20 sorted customers by hand. If you agree with at least 17, the sorting works.
Then send to 10% of each group first. Wait 7 days. Look at replies before you send the rest.
Good output looks like this:
The email for each group matches the evidence.
Discounts appear only in the Price group.
No email mentions a reason the customer never gave.
Review by hand if: a customer complained publicly, asked for a refund, or spent far more than average.
Step 5 Deploy · 20 min
Save the prompt and the 4 email drafts.
Keep 10% of the list as a holdout group that gets no email. Comparing the two groups shows what the campaign actually earned.
Repeat monthly with newly lost customers.
You now have: a tested win-back campaign and a way to prove its results.
The best marketing ideas come from marketers who live it. That’s what The Marketing Millennials delivers: real insights, fresh takes, and no fluff. Written by Daniel Murray, a marketer who knows what works, this newsletter cuts through the noise so you can stop guessing and start winning. Subscribe and level up your marketing game.
📋 Copy & paste prompt
Run this on each batch of 50 lost customers.
ROLE
You are a customer retention specialist. You help small
businesses win back customers in an honest, respectful way.
OBJECTIVE
Sort each customer below into one reason group.
Write one win-back email per group.
INPUT
Business: [NAME AND WHAT IT SELLS]
Brand voice: [ONE LINE, e.g. "friendly, plain, no hype"]
Approved offer for the Price group: [e.g. 15% off next order, or NONE]
What's new since they left: [1-3 real updates, or NONE]
Customers: [PASTE ROWS: ID | first name | last purchase | orders |
total spent | last product | cancellation reason | last support note]
GROUPS
Service problem, Price, Product fit, Drifted away, Moved on.
RULES
Use only evidence in each row. Never guess a reason.
No evidence = Drifted away.
Offer a discount only in the Price group, and only the approved offer.
Service problem emails apologize and say what changed. No discount.
Moved on emails do not sell. Ask for feedback or a referral.
Never mention facts that are not in the row.
Emails: under 120 words, one clear call to action,
use {first_name} and {last_product} as merge fields.Add the line: "Not interested? Unsubscribe here." to every email.
DECISION CRITERIA
Confidence = HIGH if the row states a clear reason.
Confidence = LOW if the reason is inferred from weak evidence.
Flag = OWNER if total spent is in the top 10% of this batch.
OUTPUT FORMAT
Part 1: a table with ID | Group | Confidence | Evidence (max 10 words) | Flag
Part 2: one email per group, each with a subject line and body.
When to use it: for every campaign and every client. Change only the input block.
Good output looks like: every customer has a group and a short piece of evidence. Discounts appear only where price was the reason. The emails read like a person wrote them.
Reach 150.000+ AI decision makers every week. Advertise with us.
🧰 AI stack
Google Sheets holds the customer list, groups and results. Free alternative: Excel.
ChatGPT or Claude sorts customers and drafts the emails. Free alternative: Gemini.
Your client’s email tool sends each group as a separate list, with unsubscribe built in. Common options: Mailchimp or Klaviyo.
Zapier (later, optional) exports newly lost customers every month. Free alternative: n8n.
📈 Monetization blueprint
Who buys this?
Subscription businesses: boxes, gyms, memberships
E-commerce brands with repeat products like coffee, pet food or skincare
B2B suppliers and wholesalers with accounts that went quiet
What are you selling?
Not “AI email marketing.” Sell this: “We bring back customers who stopped buying, and show you exactly how much revenue came back.”
Suggested pricing
$1,000–$1,500 per campaign
Or $500 setup plus 10% of recovered revenue for 60 days
$300–$500 a month to run it every month
Find your first 5
Subscribe to the newsletters of 20 repeat-purchase brands. Buy something small from a few, or just watch the emails. Brands that only send promotions, with nothing for lapsed customers, are your prospects.
Common objection: “We don’t want to annoy ex-customers.”
Answer: “We only email people who agreed to it, in small batches. Customers who had a bad experience get an apology, not a coupon.”
First outreach message
“Hi [Name], quick idea for [brand]. Most repeat-purchase brands lose a slice of customers every year and never contact them again. I sort lost customers by why they left and send each group the right message. Want me to run a small test on 100 of yours?”
Rates shown are suggested starting points, not guarantees.
Your Search Bar Is Haunted
Shoppers type exactly what they want and get “0 results found” back. They don’t complain. They just vanish to a competitor.
Gladly understands what shoppers mean, typos and all, and turns dead-end searches into product recommendations.
Reach 150.000+ AI decision makers every week. Advertise with us.
⚠️ Common mistakes
A discount for everyone. It cuts margin and trains customers to wait for offers. Fix: discounts only for the Price group.
Emailing without consent. It risks fines and spam complaints. Fix: use opted-in contacts only, with an unsubscribe link in every email.
Guessed reasons. The AI fills gaps with plausible stories. Fix: keep the “no evidence = Drifted away” rule.
No holdout group. Without it, you can’t prove the campaign caused the orders. Fix: keep 10% unemailed and compare.
Sending everything at once. One weak email reaches the whole list. Fix: test on 10% first, then send the rest.
🔗 Worth trying
✦ Latestly AI partner or affiliate. We may earn from some links. Everything here is selected because we think it's genuinely useful.
Google Postmaster Tools a free Google tool that shows whether Gmail treats your emails as spam, worth checking before a big send.
✦ Xero scan invoice history to spot B2B accounts that quietly stopped ordering, your best win-back targets.
Mailchimp tag each reason group and send win-back emails in batches, with unsubscribe handled for you.
✦ ClickUp track every win-back campaign, its reason groups and follow-up tasks in one free workspace, with no credit card needed to start.
✦ T-Mobile for Business growing the team that runs your campaigns? New accounts that add 10 lines on SuperMobile get $10,000 in savings, $1,000 per line, applied as 36 monthly bill credits.
🏆 Weekend challenge
Export 50 customers who haven’t bought in a while, from your own business or a friend’s. Run the prompt once. Read the five groups and the emails. Send one email to one real customer. Under one hour.
💬 Latestly AI Take
Most businesses spend heavily to find new customers and almost nothing to ask old ones why they left. That’s backwards. A lost customer already trusted you once. AI doesn’t make win-back emails clever. It makes them specific, and specific is what gets a reply.
Tools referenced: Google Sheets, ChatGPT, Claude, Gemini, Mailchimp, Klaviyo, Zapier, n8n, SoftSync, ClickUp, Xero, T-Mobile for Business. This issue is a how-to guide, not a news report. Prices and free-tier limits change often, so check each provider’s site before you build. Rates shown are suggested starting points, not guarantees. Partner offers are the providers’ own terms, so check each site before you sign up. Email marketing and consent rules differ by country. This is not legal or financial advice. Latestly AI is independent. Sponsored content clearly labelled. No undisclosed paid placements.
How was this edition? |
We hope you enjoyed this Latestly AI edition.
📧 Got an AI tool for us to review or do you want to collaborate?
Send us a message and let us know!
Was this edition forwarded to you? Sign up here
Latestly AI is now operated by GHE Operating Company II LLC (174 Good Hope Road, Bluffton, SC 29909), which is responsible for your subscriber data.