What are watch routines?
Watch routines are small scheduled agents whose only job is to check that things are as they should be, and to tell a person in chat when they're not. They don't try to fix everything. They notice early. We run them on the owner's own businesses alongside our ops agent and ads agent.
These run on our own businesses, not as client results.What do the ads watch routines check?
- Spend that's unusually high or low compared with normal for that day and time
- Campaigns that stopped delivering unexpectedly
- Disapproved ads or limited assets
- Conversion tracking that suddenly reports zero, often a sign of broken tracking rather than no customers
- Budgets close to running out
What do the live price checks do?
Our taxi website shows route prices on many pages. A watch routine visits a sample of live pages on a schedule and compares the prices displayed with the intended price data. If a page shows a wrong or missing price, which can happen after a template change, a cache issue or a data update, it raises an alert with the page link and the expected value. Wrong prices cost trust (and sometimes money) quickly, so this is one of the most useful checks we run.
What do site health checks cover?
- Key pages loading with the right status code and in reasonable time
- Booking and contact forms and WhatsApp links working
- Robots.txt and sitemaps not accidentally blocking search engines
- No sudden "noindex" on important pages
- TLS certificate expiry dates
How do alerts work?
Alerts go to chat, where the owner already is, not to an inbox nobody reads. Each alert says what was checked, what was expected, what was found, and a link to look. Routines are tuned to avoid noise: a single slow response doesn't page anyone, but repeated failures do. When nothing is wrong, a short daily "all clear" summary confirms the routines ran.
What is the watch routines' stop rule?
- Alone: run checks, compare against expected values, send alerts and summaries.
- With approval: any fix, which goes to the relevant agent or person, such as the ops agent for site issues or the owner for ads changes.
- Hard stops: pausing campaigns, changing prices or editing pages. Watch routines only observe and report.
- Handover: the alert itself, with evidence and a link.
Why keep watching separate from fixing?
Separating observation from action keeps things safe and simple. A routine that only reads can't break anything, so it can run often and widely. Fixes then go through agents or people that have the right permissions and stop rules. It's a pattern we recommend for any business starting with automation.
Can you set up watch routines for my business?
Yes. Watch routines are often a great first pilot: low risk, quick to build and immediately useful. Typical starters are checking that your website forms and WhatsApp links work, that prices on key pages are correct, and that ad spend is normal. See choosing a first pilot.
What should you prepare before talking to us?
You don't need a technical brief. A useful first conversation needs only a few things, and gathering them is valuable even if you never work with us:
- The process in plain words: what triggers it, who does what, and where it usually goes wrong.
- Rough volumes: how many messages, orders, tickets or invoices per week. A one-week tally is enough.
- Your tools: WhatsApp app or API, CRM, sheets, accounting or helpdesk software, including versions or plans where you know them.
- Three real examples: typical messages or cases, plus one awkward one, with personal details removed.
- What must never happen: the mistakes that would hurt most. These become the first lines of your stop rule.
- An owner: the person who will review the workflow once it runs.
Send these on WhatsApp or by email, or run the pilot scorecard first.
How does a pilot with us work?
Every engagement follows the same four steps. Consult: you walk us through the process, the tools and where it breaks. Pick the pilot: we choose one workflow and write down its inputs, outputs, tools, stop rule and success measure. Ship: we build it into the tools you already use and test it on real cases. Review: your team checks what the agent did and what it escalated, then decides whether to tune it, expand it or stop. Read the full pilot process.