Key takeaways<\/h2>\n\n\n\n\n- Post-purchase messages are outbound<\/strong>, so they need an approved template and opt-in – the customer buying is not the same as the customer messaging.<\/li>\n\n\n
- A dispatch or delivery notice is a utility template. A discount offer is marketing, and is priced higher.<\/li>\n\n\n
- Every template should invite a reply, because a reply reopens the window and makes everything after it free-form.<\/li>\n\n\n
- Three messages across the first month beats a nine-step sequence.<\/li>\n\n\n
- The review request works far better after delivery is confirmed than after dispatch.<\/li>\n\n\n
- If the customer replies, stop the sequence and answer them – a running sequence over a live conversation reads as a machine.<\/li>\n\n<\/ul>\n\n<\/div>\n\n\n\n
Which Category Your Message Falls Into<\/h2>\n\n\n\n
This decides approval and cost, and it is not a labelling choice – the content determines it.<\/p>\n\n\n\n
Utility: About Their Order<\/h3>\n\n\n\n
Dispatch confirmations, delivery notices, a problem with the order, a scheduling request. These are informational and tied to a transaction the customer made, and they are the cheaper category.<\/p>\n\n\n\n
They are also the ones customers actually want. A dispatch notice is near the top of messages people are happy to receive, which is why it is the right first template to build.<\/p>\n\n\n\n
Marketing: About Something New<\/h3>\n\n\n\n
A discount, a related product, a restock announcement, a loyalty offer. Anything whose purpose is another sale is marketing, however it is phrased, and labelling it as utility to save money gets templates rejected.<\/p>\n\n\n\n
Marketing templates work, but they need to earn their place. One good one beats a monthly cadence.<\/p>\n\n\n\n
The Test<\/h3>\n\n\n\n
Would this message exist if the customer had not just bought something specific? If yes, it is marketing. If it only makes sense because of their order, it is utility.<\/p>\n\n\n\n
Mixed messages take the higher category<\/h4>\n\n\n\n
A dispatch notice with a discount code attached is marketing. If you want the utility price, keep the offer out of it.<\/p>\n\n\n\n
Opt-in has to match<\/h4>\n\n\n\n
Agreeing to order updates is not agreeing to promotions. Collect and record them separately.<\/p>\n\n\n\n
The Three Messages Worth Sending<\/h2>\n\n\n\n
More than this and open rates fall while complaints rise. These three carry almost all the value.<\/p>\n\n\n\n
Dispatch, With a Question<\/h3>\n\n\n\n
Sent when the order ships. Carries the tracking and – this is the part most stores miss – ends with something the customer can reply to. “Any questions about it, just reply here.”<\/p>\n\n\n\n
A reply reopens the window, and from then on the whole conversation is free-form and draws on the free service allowance rather than templates. That single sentence changes the economics of everything that follows, for the reasons set out in the cost arithmetic<\/a>.<\/p>\n\n\n\nArrival Check, a Few Days Later<\/h3>\n\n\n\n
After delivery is confirmed, not after dispatch. A short message asking whether everything arrived properly catches problems before they become returns or reviews, and it is genuinely useful rather than promotional.<\/p>\n\n\n\n
If something is wrong, this is where it surfaces – route those straight into the returns flow<\/a>.<\/p>\n\n\n\nOne Reason to Come Back<\/h3>\n\n\n\n
Weeks later, once, and only if they opted in to marketing. Tie it to what they bought – the refill, the matching item, the thing people buy second. A generic discount to everybody performs worse than a specific suggestion to the right people.<\/p>\n\n\n\n
With the store connected, this can carry the actual product rather than a link to the shop – the catalogue mechanics are in Shopify DM automation<\/a>, and choosing what to suggest is the job of a product recommendation chatbot<\/a>.<\/p>\n\n\n\n![\"How](\"https:\/\/dmpro.brandid.app\/blog\/wp-content\/uploads\/2026\/10\/post-purchase-dm-flow-categories.jpg\")
It sets approval, price, and which opt-in you need<\/figcaption><\/figure>\n\n\n\nThe Sequence, With Categories<\/h2>\n\n\n\n
Three messages, two categories, one rule about replies.<\/p>\n\n\n\nWhen<\/th> Message<\/th> Category<\/th><\/tr><\/thead> On dispatch<\/td> Tracking plus an invitation to reply<\/td> Utility<\/td><\/tr> 2-3 days after delivery<\/td> Did it arrive properly?<\/td> Utility<\/td><\/tr> 3-6 weeks later<\/td> The specific thing they would buy next<\/td> Marketing<\/td><\/tr> Any time they reply<\/td> Whatever they need, free-form<\/td> No template needed<\/td><\/tr> After a complaint<\/td> Nothing automated<\/td> Person, immediately<\/td><\/tr><\/tbody><\/table>Row four is the goal. Every template above it is partly trying to produce it.<\/figcaption><\/figure>\n\n\n\nReview requests belong after row two, never before it – asking for a review from someone whose parcel has not arrived is the classic own goal.<\/p>\n\n\n\n![\"The](\"https:\/\/dmpro.brandid.app\/blog\/wp-content\/uploads\/2026\/10\/post-purchase-dm-flow-sequence.jpg\")
Email sequences are free to ignore. These are charged and arrive next to family chats<\/figcaption><\/figure>\n\n\n\nSetting It Up<\/h2>\n\n\n\n
The sequencing is easy. The opt-in bookkeeping is the actual work.<\/p>\n\n\n\n
\n\n- Collect two separate opt-ins at checkout.<\/strong> Order updates and marketing. Record both with wording and date. Conflating them is the commonest compliance failure in this whole area.<\/li>\n\n\n
- Build the dispatch template first.<\/strong> Utility, with tracking and a closing invitation to reply.<\/li>\n\n\n
- Add the arrival check, keyed to delivery.<\/strong> Trigger on confirmed delivery, not on dispatch plus a guess.<\/li>\n\n\n
- Wire replies into your normal flows.<\/strong> A reply should land in the same place as any other inbound message, not in a sequence-only branch.<\/li>\n\n\n
- Stop the sequence on any reply.<\/strong> A template arriving mid-conversation is the single thing that makes this feel automated.<\/li>\n\n\n
- Add the return message last, and only for marketing opt-ins.<\/strong> Specific to what they bought. One, not a programme.<\/li>\n\n<\/ol>\n\n\n\n
What Not to Do<\/h2>\n\n\n\n
Three failures that are common enough to name.<\/p>\n\n\n\n
The Nine-Step Sequence<\/h3>\n\n\n\n
Borrowed from email, where the cost of being ignored is near zero. On messaging it is not: each one is charged, and the complaint threshold is far lower because this is the same app their family uses.<\/p>\n\n\n\n
The Review Request Before Delivery<\/h3>\n\n\n\n
Triggered on dispatch rather than delivery, it reaches people who have not received the thing. It produces exactly the review you did not want.<\/p>\n\n\n\n
Marketing Dressed as Utility<\/h3>\n\n\n\n
A discount code in a dispatch notice is marketing with a tracking number attached. It gets templates rejected and, at scale, gets numbers reported.<\/p>\n\n\n\n
The Reply Is the Whole Point<\/h2>\n\n\n\n
Everything in the sequence is partly an attempt to produce one, and it is worth being explicit about why.<\/p>\n\n\n\n
A Reply Changes the Economics<\/h3>\n\n\n\n
Before a reply you are sending templates: approved in advance, categorised, charged per message, and limited to what you got permission for. After one, you are in a conversation – free-form, inside the free monthly service allowance, and able to send anything the channel supports including real products.<\/p>\n\n\n\n
So a template that earns a reply has not just communicated; it has moved the customer into a cheaper and far more capable mode. That is why every one should end in something answerable.<\/p>\n\n\n\n
What to Do With the Conversation You Just Opened<\/h3>\n\n\n\n
Most stores waste it. The customer replies \u201cthanks\u201d and the thread dies, when it is the best moment of the whole relationship to ask one useful question – how they are getting on, whether the size was right, what nearly stopped them buying.<\/p>\n\n\n\n
The answers are worth more than the sale. They are also the cheapest customer research available, and they arrive in the same place your DM conversation history already lives.<\/p>\n\n\n\n
Do Not Let It Become a Programme<\/h3>\n\n\n\n
The failure mode of post-purchase messaging is enthusiasm. Three messages works; six annoys; a monthly cadence to everyone who ever bought is how a number ends up restricted. The constraint is not the rules here – it is that this is the same app people use for their family, and the tolerance is correspondingly low.<\/p>\n\n\n\n
If you want more frequent contact with buyers, the channel for that is email, and the choice between them is covered in choosing the right messaging channel.<\/p>\n\n\n\n
\n\n\n\nThe Second Order Costs Less Than the First<\/h2>\n\n\n\n
A customer who just bought is the most likely person to buy again, and the window for a useful message is short. Most stores use it for nothing at all.<\/p>\n\n\n\n
Which Category Your Message Falls Into<\/h2>\n\n\n\n
This decides approval and cost, and it is not a labelling choice – the content determines it.<\/p>\n\n\n\n
Utility: About Their Order<\/h3>\n\n\n\n
Dispatch confirmations, delivery notices, a problem with the order, a scheduling request. These are informational and tied to a transaction the customer made, and they are the cheaper category.<\/p>\n\n\n\n
They are also the ones customers actually want. A dispatch notice is near the top of messages people are happy to receive, which is why it is the right first template to build.<\/p>\n\n\n\n
Marketing: About Something New<\/h3>\n\n\n\n
A discount, a related product, a restock announcement, a loyalty offer. Anything whose purpose is another sale is marketing, however it is phrased, and labelling it as utility to save money gets templates rejected.<\/p>\n\n\n\n
Marketing templates work, but they need to earn their place. One good one beats a monthly cadence.<\/p>\n\n\n\n
The Test<\/h3>\n\n\n\n
Would this message exist if the customer had not just bought something specific? If yes, it is marketing. If it only makes sense because of their order, it is utility.<\/p>\n\n\n\n
Mixed messages take the higher category<\/h4>\n\n\n\n
A dispatch notice with a discount code attached is marketing. If you want the utility price, keep the offer out of it.<\/p>\n\n\n\n
Opt-in has to match<\/h4>\n\n\n\n
Agreeing to order updates is not agreeing to promotions. Collect and record them separately.<\/p>\n\n\n\n
The Three Messages Worth Sending<\/h2>\n\n\n\n
More than this and open rates fall while complaints rise. These three carry almost all the value.<\/p>\n\n\n\n
Dispatch, With a Question<\/h3>\n\n\n\n
Sent when the order ships. Carries the tracking and – this is the part most stores miss – ends with something the customer can reply to. “Any questions about it, just reply here.”<\/p>\n\n\n\n
A reply reopens the window, and from then on the whole conversation is free-form and draws on the free service allowance rather than templates. That single sentence changes the economics of everything that follows, for the reasons set out in the cost arithmetic<\/a>.<\/p>\n\n\n\n After delivery is confirmed, not after dispatch. A short message asking whether everything arrived properly catches problems before they become returns or reviews, and it is genuinely useful rather than promotional.<\/p>\n\n\n\n If something is wrong, this is where it surfaces – route those straight into the returns flow<\/a>.<\/p>\n\n\n\n Weeks later, once, and only if they opted in to marketing. Tie it to what they bought – the refill, the matching item, the thing people buy second. A generic discount to everybody performs worse than a specific suggestion to the right people.<\/p>\n\n\n\n With the store connected, this can carry the actual product rather than a link to the shop – the catalogue mechanics are in Shopify DM automation<\/a>, and choosing what to suggest is the job of a product recommendation chatbot<\/a>.<\/p>\n\n\n\n Three messages, two categories, one rule about replies.<\/p>\n\n\n\n Review requests belong after row two, never before it – asking for a review from someone whose parcel has not arrived is the classic own goal.<\/p>\n\n\n\n The sequencing is easy. The opt-in bookkeeping is the actual work.<\/p>\n\n\n\n Three failures that are common enough to name.<\/p>\n\n\n\n Borrowed from email, where the cost of being ignored is near zero. On messaging it is not: each one is charged, and the complaint threshold is far lower because this is the same app their family uses.<\/p>\n\n\n\n Triggered on dispatch rather than delivery, it reaches people who have not received the thing. It produces exactly the review you did not want.<\/p>\n\n\n\n A discount code in a dispatch notice is marketing with a tracking number attached. It gets templates rejected and, at scale, gets numbers reported.<\/p>\n\n\n\n Everything in the sequence is partly an attempt to produce one, and it is worth being explicit about why.<\/p>\n\n\n\n Before a reply you are sending templates: approved in advance, categorised, charged per message, and limited to what you got permission for. After one, you are in a conversation – free-form, inside the free monthly service allowance, and able to send anything the channel supports including real products.<\/p>\n\n\n\n So a template that earns a reply has not just communicated; it has moved the customer into a cheaper and far more capable mode. That is why every one should end in something answerable.<\/p>\n\n\n\n Most stores waste it. The customer replies \u201cthanks\u201d and the thread dies, when it is the best moment of the whole relationship to ask one useful question – how they are getting on, whether the size was right, what nearly stopped them buying.<\/p>\n\n\n\n The answers are worth more than the sale. They are also the cheapest customer research available, and they arrive in the same place your DM conversation history already lives.<\/p>\n\n\n\n The failure mode of post-purchase messaging is enthusiasm. Three messages works; six annoys; a monthly cadence to everyone who ever bought is how a number ends up restricted. The constraint is not the rules here – it is that this is the same app people use for their family, and the tolerance is correspondingly low.<\/p>\n\n\n\n If you want more frequent contact with buyers, the channel for that is email, and the choice between them is covered in choosing the right messaging channel.<\/p>\n\n\n\n A customer who just bought is the most likely person to buy again, and the window for a useful message is short. Most stores use it for nothing at all.<\/p>\n\n\n\nArrival Check, a Few Days Later<\/h3>\n\n\n\n
One Reason to Come Back<\/h3>\n\n\n\n
The Sequence, With Categories<\/h2>\n\n\n\n
When<\/th> Message<\/th> Category<\/th><\/tr><\/thead> On dispatch<\/td> Tracking plus an invitation to reply<\/td> Utility<\/td><\/tr> 2-3 days after delivery<\/td> Did it arrive properly?<\/td> Utility<\/td><\/tr> 3-6 weeks later<\/td> The specific thing they would buy next<\/td> Marketing<\/td><\/tr> Any time they reply<\/td> Whatever they need, free-form<\/td> No template needed<\/td><\/tr> After a complaint<\/td> Nothing automated<\/td> Person, immediately<\/td><\/tr><\/tbody><\/table> Setting It Up<\/h2>\n\n\n\n
\n\n
What Not to Do<\/h2>\n\n\n\n
The Nine-Step Sequence<\/h3>\n\n\n\n
The Review Request Before Delivery<\/h3>\n\n\n\n
Marketing Dressed as Utility<\/h3>\n\n\n\n
The Reply Is the Whole Point<\/h2>\n\n\n\n
A Reply Changes the Economics<\/h3>\n\n\n\n
What to Do With the Conversation You Just Opened<\/h3>\n\n\n\n
Do Not Let It Become a Programme<\/h3>\n\n\n\n
\n\n\n\nThe Second Order Costs Less Than the First<\/h2>\n\n\n\n