2nd Purchase — the highest-leverage retention moment

Audience: Merchant·Plan: pro·Read: 8 min·Updated: 2026-08-03

In 30 seconds

• Works out your own reorder rhythm from your real orders — not an industry guess.
• Sorts one-time buyers into four windows: approaching, in-window, overdue, and lapsed.
• The 1st → 2nd purchase step is the hardest and most valuable in the whole customer lifecycle.
• Builds a ready-to-use audience plus nudge copy, and can create a targeted offer.
• Only contacts people who opted in — consent is checked before anyone is included.

In plain English

Most people who buy from you once never buy again. That's normal and it's the single biggest leak in most shops.

But here's what makes it fixable: the customers who do buy a second time are dramatically more likely to keep buying for years. Getting someone from one purchase to two is the hardest step in the whole relationship — and once they've taken it, everything afterwards gets easier.

This tool does three things:

1. Works out your shop's natural reorder gap. If your two-time customers typically come back after 38 days, that's your rhythm. It's measured from your real orders, not assumed.
2. Finds one-time buyers sitting in that window right now — the people most likely to buy again if nudged today.
3. Builds the audience and writes the message, ready for you to send.

Timing is the whole thing. Nudge someone too early and they don't need anything yet. Nudge them too late and they've already bought from someone else, or forgotten you. The window is where the money is.

How to use it

1. Click "Growth Science" in the left-hand menu, then "2nd Purchase."
2. Look at your reorder cycle at the top — that's how many days your customers typically wait before buying again.
3. Check the four groups. The "in window" and "overdue" ones are your targets.
4. Read the projected revenue figure — what these customers are worth if you reach them.
5. Click "Create segment." This builds an audience of the right people, already filtered to those who agreed to be contacted.
6. Review the suggested message. Edit it so it sounds like you.
7. Send it through email, WhatsApp, or SMS — or enrol the audience in an automated flow so this runs continuously.

⏱ ~15 min to set up · 💳 Pro+ · 🎯 Turn one-time buyers into repeat customers

Why this matters for your business

Cohort tools tell you your repeat rate is 14%. That's useful, and it's where most shops stop — because knowing the number doesn't tell you what to do about it.

The gap between measuring and acting is exactly what this fills. There is a specific list of people, right now, who bought from you once and are at the moment when a second purchase is most likely. That list is knowable, and reaching it is a straightforward email.

Most shops don't, for two reasons. They don't know who's on it, and they don't know when. So they either email everyone constantly, which trains people to ignore them, or they email nobody and hope.

The timing piece deserves emphasis. Your reorder rhythm is specific to what you sell. Coffee might be three weeks; a mattress might be seven years. Using an industry average would put your nudges in the wrong place entirely. This measures your own, from your own orders — and where you don't have enough repeat customers to measure it yet, it says so and uses a sensible default.

What this typically unlocks

What you get	Typical result
Knowing your actual reorder rhythm	Measured, not guessed
One-time buyers reachable at the right moment	Identified daily
Repeat rate after a consistent programme	Often doubles over 2–3 quarters
Time to build the audience	One click
Risk of contacting someone who opted out	Zero — consent checked first

What you actually get

Your reorder cycle

The median number of days between first and second purchase, among customers who have bought twice.

Median rather than average, deliberately — one customer who came back after two years shouldn't drag your rhythm out for everyone else.

If you don't have enough repeat customers yet to measure this reliably, the tool says so and falls back to a sensible default of around five weeks. As your repeat customers accumulate, it switches to your real number automatically.

The four windows

Every one-time buyer is placed relative to your cycle:

Window	When	What it means	Priority
Approaching	Less than 60% through the cycle	Too early — they don't need anything yet	Wait
In window	60% to 150% of the cycle	The moment. Most likely to buy now	Highest
Overdue	150% to 300% of the cycle	Slipping away, still recoverable	High
Lapsed	Beyond 300%	Needs a winback, not a reorder nudge	Lower

A 40-day cycle would give roughly: approaching under 24 days, in window 24–60 days, overdue 60–120 days, lapsed beyond that.

Work "in window" first. These are people whose need is arriving right now.

Projected revenue

The tool estimates what the reachable group is worth, based on how many are in the prime windows, a realistic second-purchase rate, and what a second order typically comes to at your shop.

Like the other estimates in the app, it's deliberately conservative — it's there to help you decide whether this is worth your time, not to promise a number.

What gets built for you

Output	What it is
A segment	A saved audience of the right customers, ready to use in a campaign or flow
Nudge copy	A suggested message, written in your brand's voice
A targeted offer	Optionally, a discount or incentive aimed at this group

The segment is a normal audience, so everything you already use — campaigns, journeys, WhatsApp, email — works with it immediately.

Consent is checked first

Only customers who have opted in to at least one channel — email, WhatsApp, SMS, or push — are included.

This isn't a setting you can turn off, and that's intentional. Nudging someone who never agreed to hear from you is both a legal problem and a good way to get your sending domain flagged. The group you're given is safe to contact.

The highest-value customers are surfaced first within the audience, so if you're testing on a small batch, you're testing on the ones that matter.

How it works (without the technical bits)

Real merchant scenarios

Scenario A — 840 people at exactly the right moment

Setup. Homeware brand, repeat rate 14%, flat revenue. Their Journey diagnostic had flagged Ascend as the weak stage.

What the tool found:

Reorder cycle       47 days  (measured from 610 repeat customers)
Approaching         1,240
In window             840   ← the target
Overdue               520
Lapsed              2,910
Projected revenue  £18,400

Action. Built the segment, edited the suggested copy, and sent a simple "you bought this 6 weeks ago — here's what goes with it" email. Then enrolled the segment in an automated flow so it ran continuously.

Result over four months:

	Before	After
Repeat rate	14%	26%
Revenue	flat	+31%
Ad spend	unchanged	unchanged

The 840 already existed. Nobody had ever contacted them at the right time.

Scenario B — Timing mattered more than the offer

Setup. Coffee subscription, 22-day reorder cycle. Ran the same message to two groups.

Group	Sent	Conversion
In window (13–33 days)	Identical email	11.2%
Approaching (under 13 days)	Identical email	1.4%

Eight times the conversion, same message, same discount. The only difference was whether the customer had run out yet.

What they'd been doing before. Emailing every one-time buyer seven days after purchase — deep in the "approaching" window, when nobody needed more coffee.

Change. Moved the nudge to day 15. Second-purchase rate went from 9% to 24% over the following quarter.

Scenario C — The default cycle, then the real one

Setup. New shop, four months old, 31 repeat customers.

Initial reading:

Reorder cycle   35 days  (default — not enough data yet)

They used it anyway and it worked reasonably.

Three months later, with 180 repeat customers:

Reorder cycle   68 days  (measured from your customers)

Nearly double the default. Their product lasted much longer than a typical consumable, so half their nudges had been going out far too early.

After adjusting, second-purchase conversion went from 6% to 17%.

The lesson. The default gets you started. Check back once you've accumulated real repeat customers.

Scenario D — Consent filtering catching a real problem

Setup. Merchant had 4,100 one-time buyers and expected an audience of roughly that size.

The segment came back at 1,850.

Why. Fewer than half had ever opted in to any channel. Their checkout had the marketing consent box unticked by default and buried below the fold.

The bigger finding. This wasn't a limitation of the tool — it was a discovery that more than half their customer base was unreachable, permanently, and they hadn't known.

Action. Redesigned the opt-in at checkout with a clear value proposition and moved it above the fold. Opt-in rate went from 44% to 71%.

Compounding effect. Every future campaign, not just this one, reaches 60% more people.

Best practices

✅ Target "in window" first. Scenario B shows the difference timing makes.

✅ Enrol the segment in an automated flow rather than sending manually. This should run continuously, not once.

✅ Re-check your reorder cycle every quarter, especially in your first year.

✅ Make the nudge useful, not just promotional. "Here's what goes with what you bought" outperforms "10% off."

✅ Treat "lapsed" as a separate winback campaign, with a different message.

✅ If your opted-in share is low, fix that first — it caps everything else you can do.

❌ Don't nudge the "approaching" group. You'll train them to ignore you before they're ready.

❌ Don't use an industry-average reorder time. Yours is specific to what you sell.

❌ Don't lead with a discount every time. You'll teach customers to wait for one.

❌ Don't run this as a one-off. New one-time buyers enter the window every single day.

Plan tiers

Capability	Free	Starter	Pro	Agency	Enterprise
Measured reorder cycle	—	—	✓	✓	✓
Four-window classification	—	—	✓	✓	✓
Projected revenue	—	—	✓	✓	✓
One-click segment build	—	—	✓	✓	✓
AI nudge copy	—	—	✓	✓	✓
Targeted offer creation	—	—	✓	✓	✓
Consent filtering	—	—	✓	✓	✓
Multi-store rollout	—	—	—	✓	✓

Frequently asked

How many repeat customers do I need for a real cycle? A modest number — below that it uses a default and tells you. It switches to your measured cycle automatically once there's enough.

Why median instead of average? One customer returning after three years would drag an average out badly. The median describes your typical customer.

Can I change the window boundaries? They're fixed proportions of your cycle, which keeps them consistent as your cycle changes. You control which windows you target.

What about customers who bought twice already? This tool focuses on the 1→2 step specifically, because it's the hardest and most valuable. Ongoing repeat customers are handled by journeys and loyalty.

Does it send anything automatically? No. It builds the audience and drafts the message. You send, or you enrol it in a flow you control.

Why is my segment smaller than my one-time buyer count? Consent filtering — only opted-in customers are included. See Scenario D.

Should the nudge include a discount? Test it. Discounts lift response but train customers to wait. Many shops get better long-term results from a helpful, non-promotional nudge.

How does this relate to cart recovery? Different problem. Cart recovery chases someone who didn't complete a purchase. This chases someone who did, and is due for another.

See also

• Growth Science overview — all eight tools
• Journey — the Ascend stage this fixes
• Offer Lab — the tripwire that feeds this
• Positioning — makes the nudge copy sound like you
• Segments & cohorts — the audience it builds
• Journeys & automations — running it continuously
• Customer intelligence — the cohort view of repeat rate
• Loyalty overview — what to do after purchase two