How Can I Automate Follow-Ups After a Customer Conversation?
Automating follow-ups after a customer conversation means designing a rule-based sequence that checks in, re-engages, or confirms a resolution after the conversation itself has ended — and on WhatsApp specifically, it means working around one hard technical constraint that most guides skip: the 24-hour window Meta enforces on free-form messages.
The WhatsApp 24-hour window: the constraint that shapes everything else
Per Meta’s own WhatsApp Business Platform documentation, when a customer messages or calls you, a 24-hour timer called a customer service window starts. Inside that window, you can send any free-form service message — text, images, buttons, whatever the conversation needs, with no pre-approval required. If the customer messages again, the timer resets to a fresh 24 hours.
Once the window closes, the rules change completely: you can only send pre-approved template messages, and Meta’s template approval process can take up to 24 hours on its own. This single rule is why “just send a follow-up message” is not as simple as it sounds on WhatsApp specifically, and it’s the reason a follow-up strategy built for email doesn’t transfer directly to WhatsApp without redesign.
The practical design implication: a same-day “did this resolve your issue?” check-in sent within 24 hours of the last customer message can be free-form. A follow-up sent after the window has lapsed — a 3-day post-purchase satisfaction check, for example — has to go out as a pre-approved template. If the customer replies to that template, the window reopens and the conversation goes back to free-form. This is why the timing of a follow-up isn’t just a UX decision on WhatsApp — it determines which message type you’re technically allowed to send.
What good automated follow-up logic looks like
GoHighLevel’s Auto Follow-Up feature is a useful reference point for the shape this logic should take, regardless of which platform you actually use. It fires on three trigger conditions — the contact stopped responding, the contact was marked busy, or the contact explicitly asked for a follow-up — and runs up to five sequenced messages with configurable delays in minutes, hours or days, respecting the business’s active hours and the contact’s timezone.
The detail worth copying regardless of platform is the suppression logic: follow-ups stop automatically the moment the system detects disinterest, disqualification, anger, or a natural close to the conversation. A follow-up sequence that keeps messaging someone who has clearly said no is worse than no follow-up at all — it turns a minor unresolved issue into a reason to distrust the brand.
Six follow-up scenarios worth automating
| Scenario | What the follow-up should do |
|---|---|
| Post-delivery confirmation | Confirm the order arrived and ask if anything’s wrong before the return window closes. |
| Post-purchase satisfaction / review request | Check in a few days after delivery, then ask for a review only once satisfaction is confirmed. |
| Abandoned conversation | Re-engage a customer who went quiet mid-chat, once, with a low-pressure nudge rather than repeated messages. |
| “Will check and get back to you” | Close the loop an agent opened but didn’t finish — this is the follow-up customers notice most when it’s missing. |
| Post-return/refund status | Confirm once a refund or replacement has actually processed, not just that the request was received. |
| Unconverted sales inquiry | A light re-engagement for someone who asked questions but didn’t buy — not a hard sell, a genuine “still interested?” check. |
Two of these deserve more detail because they’re the easiest to get wrong. The “will check and get back to you” case is a promise your business made, not a marketing touchpoint — it should be tracked as an open item with an owner, not left to whoever remembers. And the post-purchase satisfaction check works best as a gate, not a single message: ask if everything’s fine first, and only ask for a review once the answer is yes. Asking for a review before confirming satisfaction gets you reviews from people who haven’t actually decided how they feel yet.
How to build this on AeroChat
Be precise about what exists today rather than assuming a platform has a dedicated “follow-up automation” feature just because the individual pieces do. AeroChat’s documented capabilities relevant here are WhatsApp broadcast campaigns (Growth plan and above), reusable message templates, and conversation tagging. These are building blocks you combine into a follow-up workflow, not a single feature that runs a full sequence for you out of the box.
A practical version of the six-scenario table above, built from those pieces: tag a conversation with the scenario it belongs to as soon as it’s identified — “post-purchase,” “unresolved,” “will-check-and-return” — rather than trying to reconstruct that later from raw chat history. Then use conversation tags to segment who gets which broadcast, and schedule the broadcast for the appropriate delay per scenario — same-day for an abandoned conversation, a few days out for a post-purchase check. If any of those follow-ups need to cross the WhatsApp 24-hour window, confirm your message templates are Meta-approved templates specifically, not just internally saved reply snippets — an unapproved template will fail to send, not just look slightly different.
Measuring whether follow-ups are actually helping
A follow-up sequence that nobody checks is a guess dressed up as a system. Track, at minimum: response rate (did the customer reply at all), resolution rate for open-loop follow-ups specifically (did “will check and get back to you” actually get closed), and opt-out or complaint rate for the sequence overall. A follow-up type with a high opt-out rate is telling you something concrete — wrong timing, wrong tone, or a scenario that didn’t need automating in the first place.
Frequently asked questions
Can I send a WhatsApp follow-up any time I want?
Only within 24 hours of the customer’s last message, using a free-form message. Outside that window, it has to be a pre-approved template message, and template approval can take up to 24 hours, so plan ahead rather than trying to send one the moment you need it.
How many follow-up attempts is reasonable?
There’s no single official rule, but the pattern worth following is: stop the moment the customer responds negatively, shows disinterest, or the situation resolves itself — a fixed number of attempts matters less than knowing when to stop.
Should every conversation get a follow-up?
No. A quick, fully resolved “what’s your return policy” question usually doesn’t need one. Follow-ups earn their place on conversations with an open loop — an unresolved issue, a purchase decision still pending, or a delivery still in transit.
Does AeroChat send follow-ups automatically out of the box?
Not as a single pre-built feature. AeroChat provides the components — tagging, templates, and WhatsApp broadcast — that you configure into a follow-up workflow matching the scenarios that actually matter for your store, rather than a generic sequence that fires the same way for every conversation.
