The post-purchase window is the only moment when a customer will reliably open anything you send, because they are waiting for information from you. Most businesses spend that attention on a single order confirmation and then go quiet for six months. The flow that fills the gap is not complicated, but it does require deciding which of its messages are receipts and which are advertisements, because the two are governed differently.
Why this window is worth engineering
Triggered flows outperform broadcasts on revenue per recipient by an order of magnitude, because they arrive when context is live rather than when your calendar says Tuesday. Klaviyo’s benchmark analysis of more than 143,000 abandoned cart flows found an average of $3.65 per recipient against a top decile of $28.89, with an average placed order rate of 3.33%. Post-purchase flows work on the same mechanism: known intent, known product, known timing.
There is also a defensive argument. The days after a purchase are when refund requests, "where is my order" tickets, and negative reviews originate. An email that pre-empts the question is cheaper than the support conversation it prevents, and considerably cheaper than the one-star review it prevents.
The sequence, and what each message is legally
| Message | Timing | Classification |
|---|---|---|
| Order confirmation | Immediate | Transactional. Keep the order details at the top of the body, above any cross-sell |
| Shipping or access notification | On dispatch or provisioning | Transactional, same rule |
| Onboarding or usage guidance | On delivery, or day one of access | Transactional if genuinely about using what they bought; commercial the moment it leads with an offer |
| Check-in | 3 to 7 days after delivery | Commercial. Needs an unsubscribe |
| Review request | After enough elapsed use for an opinion to exist | Commercial. Needs an unsubscribe |
| Replenishment or cross-sell | Timed to the consumption cycle of the specific product | Commercial. Needs an unsubscribe |
The classification column is not pedantry. Under the CAN-SPAM Rule at 16 CFR 316.3, a message mixing commercial and transactional content is deemed commercial if a recipient reading the subject line would expect an advertisement, or if the transactional content does not appear in whole or substantial part at the beginning of the body. A confirmation email that opens with a product carousel and puts the order number underneath has become a commercial message that is missing everything a commercial message requires.
Timing is a product question, not a marketing one
The replenishment trigger should fire at a fraction of the product’s actual consumption cycle, and you know that cycle better than any template does. A 30-day supply of something warrants a nudge around day 24, not day 30, because the reorder needs to arrive before the current one runs out. A durable good has no replenishment cycle at all, and sending one anyway signals that nobody thought about it.
The review request has the same logic and is more often got wrong. Asking on delivery day produces a review of your packaging. Asking after a fortnight of use produces a review of the product, which is the one that converts future buyers. If your product has a learning curve, wait longer still.
Make the review request cost one click
Review request performance is dominated by friction, not by persuasion. A star rating embedded directly in the email, where clicking a star opens a prefilled review form, will outperform a paragraph asking politely and linking to a login page by a margin no amount of copywriting can close. Ask for one specific product rather than a general opinion of your store, and do not offer an incentive contingent on a positive review, which most marketplaces prohibit and which poisons the reviews you already have.
Test the flow against a holdout, not against nothing
The revenue your platform attributes to a post-purchase flow includes every customer who was going to reorder anyway and happened to click your email on the way. To find the real effect, withhold a randomly selected 5 to 10% of eligible customers from the commercial portion of the flow for a quarter and compare revenue per customer. It is the only test that distinguishes a flow that generates orders from one that merely gets credit for them.
Key takeaways
- Triggered flows earn revenue per recipient an order of magnitude above broadcasts. Klaviyo measured $3.65 average on abandoned cart flows.
- Confirmations and shipping notices are transactional only while the order details stay at the top of the body, per 16 CFR 316.3.
- Send the transactional half of the flow from your transactional identity and the commercial half from your marketing identity.
- Fire replenishment before the product runs out, not when it does, and request reviews after real use rather than on delivery.
- A 5 to 10% holdout is the only honest measure of what the flow actually causes.
Related reading
Sources
- Abandoned Cart Benchmark Report, Klaviyo (2024)
- 16 CFR 316.3, Primary purpose, eCFR, U.S. Federal Trade Commission
- Email sender guidelines, Google Workspace Admin Help
- Automation
- E-commerce
- Retention

Written by
Valter Brandt
Chief Marketing Officer
Valter Brandt is the Chief Marketing Officer of ThisCom, working with clients across the United States and internationally. He has led marketing strategy through the major shifts in social advertising, mobile, content marketing, programmatic media, and marketing automation.
All articles by Valter Brandt →


