Notified if a guest does not check in?

Hi ​@bremyan 

To ensure you don’t send cleaners to an empty home, I recommend a three-layered approach:

• Automated "First Morning" Message: Set up a message for the morning after check-in (e.g., "Hi %guest_first_name%, hope you’ve settled in! Let us know if you need anything."). A guest's response is the quickest confirmation of occupancy.

• Smart Lock Notifications: If your lock is integrated with Hospitable, enable notifications for when the guest code is first used. You can receive these via email or push notification per user.

• External Cameras: For the most reliable visual confirmation, use Ring or similar cameras to monitor arrivals and departures via their native apps.

Combining automated messaging with smart lock alerts is usually the most effective way to guarantee you know whether a guest has arrived before the cleaning crew is scheduled.

I hope this helps!

​@Mahmodul the request is a little more subtle… While it’s easy to determine that something DID happen, it’s a little harder to determine that something DIDN’T happen. Of course there is no built-in way to be notified that “your guest never checked in.” You would need to have some external solution “listening” for the “smart lock code first used” message, and if that message is received, it marks the guest as “checked-in,” but if that message never gets received by the check-out date, it would send some sort of alert that the stay never occurred.  The MCP doesn’t expose any kind of smart device capabilities yet, so that’s ruled out as a solution. Maybe webhooks are a way, or some solution that involves monitoring an email inbox and/or forwarding the email that Hospitable sends when the smart lock code is first used. Whatever the solution is, it’s not out-of-the-box. 

It’s possible to have Claude build this for you via MCP/Hospitable and use seam.co to send a webhook. I know for sure this will work with a Schlage Encode Lock, you’d have to do some additional research if you have a different lock.

Here's how it would work conceptually:

You'd need a separate scheduled job — Cloudflare Workers supports cron triggers, so you could have Claude build you a Cloudflare worker which will run in the cloud indefinitely and it’s free.

The worker would run a check every day at Xam or Ypm. The job would look at that day's reservations in Hospitable, check if there's a guest checking in, then check Seam's event history to see if the guest’s code was used that day by 3pm (or whatever time you choose).

If not, it sends you a notification — either a text via Twilio, (which you have to pay for), an email via Resend or similar service,  or a Push notification to your phone via Pushover or a similar service (Pushover is a one-time $5 app fee, then free). In theory, Pushover is probably the simplest — one API call sends a push notification directly to your phone. No email spam filters, no SMS costs. I haven’t used Pushover though so I can comment specifically on how easy it is to set that up. I assume Claude can walk you through it. 

If the code has been used, then the worker does nothing. You only get the notification if the code hasn’t been used. 

You could then manually contact the cleaning team or use Twilo to send the cleaner a text.

Automatic no-show detection is tricky because, by definition, you’re trying to detect something that didn’t happen. So you always need some kind of physical signal as a starting point.

As mentioned above, that usually comes from:

• smart lock usage (code never used)
• cameras (no arrival detected)
• noise sensors (no activity) (Minut, noiseAware...)
• or even WiFi/network data (no device connected, but that requires more advanced setups)

Once you have that physical input, you can build logic on top of it.

From there, it’s possible to create a simple automation chain using tools like n8n, Zapier, or webhooks:

• check reservations for the day
• verify if a “check-in event” was triggered (e.g. lock opened)
• if not → send a notification (email, SMS, push)

But the key point is: everything starts with reliable physical confirmation. Without that, there’s no solid way to detect a no-show automatically.