A 30-day playbook to halve your order-status messages

Updated 2026-09-24 · 4 min read · By the Convo360 team

Week one: stop the silence

Before automating anything, look at when customers write. Almost always it is during a gap: after payment and before dispatch, or after dispatch and before the first courier scan. Those gaps are where the anxiety lives.

Send three Utility templates: order confirmed, shipped with a tracking link, and out for delivery. Nothing promotional. If you already send these by email only, assume most customers have not read them.

Brands that do only this step usually see a fifteen to twenty-five per cent fall in inbound status messages, with no bot involved at all.

Pick the sending moment carefully. A shipped message that goes out when the label is printed, hours before the courier actually collects, teaches customers that your updates are unreliable. Fire it on the first courier scan instead, even if that means the message arrives later.

Keep each one short. Three lines and a link. Long templates get skimmed, and a skimmed update produces the same question it was meant to prevent.

Week two: make the promise honest

Check what your product pages, checkout and confirmation message each promise. It is common to find three different delivery windows, and customers hold you to the fastest one they saw.

Pick one window per pin-code band, make it slightly pessimistic, and use it everywhere. A parcel arriving on day four against a promise of five produces a compliment. The same parcel against a promise of three produces a complaint.

• Fix the product page first, since that is where expectation is set.
• Match the confirmation message to it word for word.
• Say something explicit about weekends and festival days.

Say the awkward part out loud in the copy. If you do not dispatch on Sundays, write that. If a pin code adds two days, write that. Indian customers are entirely reasonable about a five-day delivery that was described as five days, and entirely unreasonable about a four-day delivery that was sold as two.

Get the returns window on the same page while you are there. A surprising share of "where is my order" messages are really "can I still return this if it arrives after the wedding".

Week three: let the bot answer the rest

Now automate. In Convo360, connect Shopify under Integrations and leave order answers switched on, so the bot replies with the real status and tracking link pulled from your store. The full setup is in building a Shopify order-status bot on WhatsApp.

Two rules make the difference between a bot customers trust and one they work around:

• It must never guess. If the store has no date, the bot does not offer one.
• It must hand over cleanly. "Let me get someone to look at this" is a better answer than a cheerful non-answer.

Do not turn everything on at once. Run the automated answer on one channel for a week, read the transcripts, then extend it. The first week almost always surfaces something specific to your catalogue, such as pre-orders or bundled items that look odd when described by a machine.

Tell your team what the bot will and will not do before it goes live. Agents who do not know the rule will answer over the top of it, which produces two replies to the same question and a confused customer.

Week four: give the real problems somewhere to go

Once status questions are handled, what remains is the work that actually needs judgement: a parcel the courier has lost, a wrong item, a refund after dispatch. Those must not sit in the same undifferentiated chat list.

Switch on the Support Desk so each of these becomes a ticket with a priority, an owner and a reply deadline counted in your working hours. Tag them as you go, because the tags are what tell you next month which courier or which SKU is generating the complaints.

Set reply deadlines that match your actual staffing rather than your ambition. A two-hour first-response target that is missed daily teaches everyone to ignore the alerts. Four hours, consistently met, is worth far more than two hours that are aspirational.

What to measure, and what to ignore

Ignore total message volume. It will barely move, because customers still write in and still get answers.

If the last row is rising, your bot is closing conversations it has not actually resolved, which is worse than not automating at all. Reopen rate is the honest check on every support automation.

One more thing to watch in month two: which upstream problem your tags are pointing at. Support automation makes the support queue calmer, but the tags are the first clean dataset most brands ever get about their own fulfilment. If a quarter of escalations carry one courier's name, the next improvement is not in support at all.

Step by step

1. Send three Utility templatesOrder confirmed, shipped with tracking, out for delivery. Keep all three free of promotional language so they stay cheap and deliverable.
2. Fix the delivery promiseOne window per pin-code band, slightly pessimistic, used identically on the product page, at checkout and in the confirmation.
3. Connect Shopify and switch on order answersIntegrations → Shopify → connect, then leave 'Let the AI answer where is my order?' ticked.
4. Switch on the Support DeskSo that anything the bot hands over becomes a ticket with an owner and a reply deadline, not a message in a list.
5. Tag every escalated ticketCourier, damaged, wrong item, refund. A month of tags tells you which upstream problem to fix next.
6. Review the four metricsShare reaching a human, first response time on non-order messages, order questions self-served, and reopen rate.

Example conversation

Frequently asked questions

Related reading

Available across India — see WhatsApp & AI calling automation by state and city.