A customer asks the price in chat, you answer within minutes, and then nothing — no order, no reply. Chasing threads like this by memory is why leads go cold: with dozens of chats running at once, nobody remembers who to message again or when. A three-touch, label-tracked follow-up fixes that without turning into spam.
Why “just follow up” breaks down at volume
One seller can hold five open price questions in their head. Nobody holds fifty — and fifty is a normal afternoon during a promotion or a live-selling session, when price questions from LINE, Facebook and Instagram DM all land at once, on top of the orders already in progress. Without a system, the ones that get a second message are whoever the agent happens to remember, not whoever is most likely to still be interested. New chats keep arriving and push older, unanswered ones further down the screen, so the questions most likely to close — the ones from a day or two ago that just need one more nudge — are exactly the ones that get buried. The fix isn’t a better memory; it’s moving “who’s owed a follow-up” out of anyone’s head and onto the conversation itself.
A three-touch cadence, timed to when interest actually cools
Three follow-ups, each doing a different job, work better than one generic nudge repeated three times:
- Touch 1, within 2 hours (same day): answer the question they didn’t ask yet — confirm stock, state the shipping fee, or give the size they’d need. Silence right after a price question is usually a stalled decision, not a lost one, and the fastest way to unstick it is to remove whatever they were probably weighing up before they went quiet.
- Touch 2, next business day: one specific, new detail, not a repeat of touch 1. “The color you asked about has one piece left” works harder than “still interested?” — it gives the customer a reason to answer rather than a reason to feel chased.
- Touch 3, 2–3 days later: a short, final check, no new pitch. No reply after this touch means stop — see below.
Spacing the three touches this way matters as much as the wording: same-day contact catches the decision while it’s still warm, the next-day message reaches someone who has had time to check a budget or ask a family member, and the final check two or three days out catches anyone who was simply away from their phone.
Track who’s next with a label, not memory
This only works if “awaiting reply” is visible across every channel at once, which is what labels and per-inbox SLAs in a unified inbox are for: label a conversation the moment a price question goes unanswered, and let the inbox sort by how overdue it is instead of scrolling app by app to check. Canned responses keep the touch-2 and touch-3 wording consistent even when a different agent picks up the thread later in the day, so the customer isn’t the one who notices the handoff.
Answer instantly when they come back
Because every contact keeps one conversation history regardless of channel, a reply three days later doesn’t have to start with “sorry, what did you ask about again?” — the agent, or an AI agent trained on the shop’s own knowledge base, sees the original question and can answer stock, sizing or shipping without re-asking. If what comes back is something only a person should decide — a discount, a swap, a second failed payment — that’s the moment to hand the cart back to a human instead of letting the bot guess. Once they do pay, the same checklist that applies to any transfer still applies: verify the slip before you pack the order.
When to stop
Three unanswered touches is the limit. Label the conversation “no response” and leave it — a fourth message reads as pressure, not service, and tends to cost more goodwill than the order is worth. If the price changes or the item comes back in stock later, that’s a new reason to reach out, not a continuation of the old sequence, and it’s worth starting the three-touch cadence fresh rather than picking up where it left off.
