Automation: Draft the Client Incident Email the Moment a Site Goes Down

What This Builds

A Zap that watches your uptime monitor and, the moment a maintenance client's site goes down, drafts a calm, client-facing status email and drops it in your inbox as a draft, not a sent message. You're already troubleshooting when a site goes down; this automation means you're not also stopping to write a status update from scratch while the clock runs. You open the draft, adjust it for what you actually know, and send it yourself.

Prerequisites

• A Zapier account that supports multi-step Zaps and the AI by Zapier action (Professional, $29.99/month)
• An uptime monitor already running on each maintenance client's site, connected to Zapier through its native integration or a webhook
• Your email account connected to Zapier as a draft-creation step
• Total ongoing cost: $29.99/month for the Zapier plan this Zap needs, on top of whatever your uptime monitor already costs

The Concept

This works like a night-shift assistant who notices the alarm going off and starts writing the "we're on it" email while you go check what's actually wrong. The assistant doesn't decide what to tell the client and doesn't press send. It hands you a draft built from the one fact it knows for certain (the site is down, and when it went down), so the only thing standing between the alert and a client email is your review, not a blank page.

Build It Step by Step

Part 1: Confirm your uptime monitor's Zapier connection

Most uptime monitors offer either a native Zapier trigger or a webhook you can point at a Zapier "Catch Hook" trigger. A native trigger is usually simpler to set up and typically fires on both down and back-up events, so you'll filter for the down event specifically in the next part. If your monitor only offers a webhook, you can still build this Zap. It just takes an extra step to parse the incoming payload into individual fields (site name, status, timestamp).

Part 2: Build the Zap

1. Set the trigger to your uptime monitor's down-alert event (via its native Zapier integration) or a Catch Hook trigger if you're using a raw webhook
2. Add a Filter by Zapier step immediately after, set to continue only when the alert type is "down" and not "up," so this Zap doesn't fire a client email every time a site recovers on its own
3. Add an AI by Zapier action step with a prompt like:

Draft a short, calm status email to a small business client whose website just went down. Use the following facts only:
- Site name: {{site_name}}
- Time the outage was detected: {{alert_time}}
Write 3 to 4 sentences: acknowledge the issue, state that it's being actively investigated, and give a realistic sense of when to expect an update (within the next hour, or by end of day, depending on typical response time). Do not speculate about the cause. Do not mention internal tools, hosting providers by name, or anything about billing. Sign off with a professional but warm tone.

1. Add a final step to create a draft in your email account (not a send action), addressed to that client's contact, with the AI output as the body and a subject line like "Quick update on [site name]"
2. Add a separate, second notification step that alerts only you, not the client, through whatever channel you actually check first when working (text, Slack, a push notification). This is the step that gets your attention fast; the drafted client email can wait the few minutes it takes you to start diagnosing

Part 3: Test and set the review habit

Trigger a test alert (most uptime monitors have a "test this integration" or "send test alert" option) and confirm the draft lands correctly formatted, with no leftover template syntax or missing fields. Before you rely on this in a real outage, decide on a personal rule: you always read the draft in full before sending, every time, no exceptions, even when you're mid-troubleshoot and tempted to just hit send.

Real Example: A Site Goes Down at 2pm

Setup: A bakery client's site is monitored by your uptime tool, wired into this Zap.

Input: The monitor detects the site is unreachable at 2:14pm and fires the down alert.

Output: Within a minute, you get a push notification on your phone that the bakery site is down, and a draft email is waiting in your account: "Hi [contact name], we noticed your website became unreachable at 2:14pm today and are actively looking into it. We expect to have an update within the hour. Thanks for your patience, [your name]." You open it, confirm nothing in it is wrong, and send it while you start checking the server logs.

Time saved: The drafting step, which used to mean stopping mid-diagnosis to write a coherent update, is gone. The client hears from you faster without slowing down the actual fix.

What to Do When It Breaks

• The alert never fires and you find out from the client instead of the monitor → this is the failure that matters most here, because it means the whole safety net was down without you knowing. Test the full chain (monitor to Zap to draft) on a real schedule, not just once at setup, by triggering a monthly test alert and confirming the draft appears. Also turn on Zapier's own "notify me when a Zap fails" alert as a second layer.
• The draft goes out with a guessed cause or an internal detail in it → tighten the AI prompt to explicitly forbid speculation and internal references, as shown above, and always read before sending regardless of how the prompt is written. The prompt is a guardrail, not a guarantee.
• The Zap fires on every recovery too, not just outages → check the Filter step; make sure it's matching the down-status value exactly as your monitor sends it, since some monitors use "down" and others use a numeric status code.
• Multiple clients go down at once and drafts pile up → that's expected behavior, not a bug. Triage by whichever client has the most time-sensitive business impact (an e-commerce site mid-sale versus a static brochure site) rather than by draft order.

Variations

• Simpler version: Skip the AI drafting step and keep a single fill-in-the-blank incident email template saved in your notes app, triggered by nothing more than the same push notification. Slower to personalize, but nothing to review for accuracy beyond the blanks you fill in yourself.
• Extended version: Add a follow-up Zap that fires when the monitor reports the site back up, drafting a short resolution email the same way, so clients get a clean close to the incident without you having to remember to send one.

What to Do Next

• This week: Wire up the Zap for your two or three highest-priority retainer clients first
• This month: Extend it to your full monitored client list and run the monthly test-alert habit at least once to confirm it's still firing
• Advanced: Log every real incident and its resolution time into a sheet automatically, so you have real data for renewal conversations about your maintenance retainer's value

Advanced guide for freelance web developer professionals. These techniques use more sophisticated AI features that may require paid subscriptions. Automation and monitoring platform interfaces change often. If a menu name in this guide doesn't match what you see, look for the same feature under a nearby label.

Keep a human in the loop. This automation only ever produces a draft. Never let this Zap send a client email automatically, no matter how confident the AI output looks. Read every incident email before it goes out, and never let one mention credentials, internal tool names, or blame directed at a host or a previous developer, even if that's genuinely what caused the outage. A client-facing incident note states what's happening and when to expect an update, nothing more.