One API, Every Channel: Automating Two-Way Messaging with Make and Sinch

Your customer books a dental appointment online. Four hours before they're due in, a WhatsApp message lands on their phone: "Hi Sarah, reminder about your 2pm tomorrow. Reply YES to confirm or NO to cancel."

Sarah replies "NO", she forgot she has a work call.

In the old world, that reply sits unread in a dashboard until someone processes it manually. The slot goes to waste. In the new world, the inbound "NO" is a trigger. Make catches it, updates the booking record, marks the slot available, and sends Sarah a rescheduling link.

That's the pattern this post is about: closing the loop.

Why outbound-only messaging leaves value on the table

Most messaging automation stops at the send. New order → SMS confirmation. New lead → WhatsApp welcome. Useful, but half the value is in what happens when the customer replies. The highest-ROI flows are transactional and bidirectional:

• Appointment reminders where the customer confirms or reschedules

• Delivery exceptions where the customer picks a redelivery slot

• Payment failures where the customer responds to retry

• Support triage where an inbound keyword routes to the right team

Building this used to require a backend engineer, webhook handlers, and conditional routing logic. The Sinch Conversation API abstracts all of that into a single integration and Make turns the logic into a visual canvas you can iterate on in minutes.

What the Sinch Conversation API does for your stack

The Sinch Conversation API is a single endpoint in front of SMS, WhatsApp, RCS, and more. Instead of maintaining separate integrations per channel, different auth flows, different payload shapes, you connect once and route intelligently.

In a Make workflow, this means:

• Send SMS via one standardized module, regardless of carrier or country

• Send WhatsApp template messages with rich formatting, buttons, quick replies, media

• Receive inbound messages as webhooks that Make watches and acts on

• Automatic channel fallback if WhatsApp fails, Sinch falls back to SMS

You're building logic. Channel complexity is handled below the abstraction layer.

The two-way appointment reminder: a complete walkthrough

Step 1: Set up your trigger Create a new Make scenario.  

Add a Google Sheets – Watch Rows module (or your booking system, Calendly, Acuity, etc.) and filter for appointments 24 hours out with status "Pending Confirmation."

Step 2: Send the reminder via Sinch

Add the Sinch Build, Send WhatsApp Template Message module. Authenticate with your Project ID, Access Key ID, and Access Key Secret from the Sinch Customer Dashboard (Conversation API → Settings). Map the fields: customer phone number, your pre-approved template, and dynamic parameters (name, datetime, location). The message lands like this:

"Hi [name], your appointment at [location] is confirmed for [datetime]. Reply YES to confirm or NO to cancel."

Step 3: Add a Webhooks to catch the inbound reply

Add a Custom Webhook as the trigger for a second scenario. In your Sinch dashboard, register this URL as the endpoint for MESSAGE_INBOUND events. When the customer replies, Make receives the sender's number and message text as a JSON payload.

Step 4: Parse and branch Add a Router module with two paths:

• YES → Update the booking row to "Confirmed" in Google Sheets. Send a confirmation SMS via Sinch:

"You're all set! See you at 2pm."

• NO → Cancel the slot and mark it available. Send a rescheduling link via Sinch. Post a Slack notification to your team:

"Appointment cancelled by Sarah M. The slot is now open."

The whole build takes about 20 minutes. What once required a Node.js server, a webhook handler, and a database migration is now one visual scenario.

More transactional patterns to build next

The same trigger → send → receive → branch pattern applies across many workflows:

• Order exception handling: Shipment delayed? Send a WhatsApp message via Sinch: "Reply 1 to hold, 2 for a refund, 3 to reroute." The reply updates the fulfilment record automatically.

• Payment retry flows: Stripe payment fails → Sinch sends an SMS with a secure payment link → customer replies "DONE" → Make checks Stripe and confirms or re-escalates.

• AI-powered inbound triage: Inbound WhatsApp message → OpenAI classifies intent and urgency → high-priority replies escalate to Slack, routine ones trigger an automated response and CRM update.

• RCS rich notifications: Package out for delivery → branded RCS card with map, ETA, and a one-tap "Leave at door" button. Users without RCS get an automatic SMS fallback from Sinch.

Getting started

Most teams have at least one messaging flow that's still half-manual, half-hoped-for. Make and Sinch fix that without a sprint cycle. Pick your highest-volume pain point, build the scenario, and you'll have time left over to build the next one.