Agents can now wake up at 7 a.m., check something and report back. What that looks like in a café, a clinic or a club, and where to be careful.
Most people meet AI as a chat window: you ask, it answers, you close the tab. The quieter change this year is AI that doesn’t wait to be asked. It runs on a timetable, does a defined job and leaves the result where your team will see it.
Anthropic’s developer platform added scheduled runs for its managed agents in June. In October it added a beta feature called dynamic workflows, where an agent can plan and carry out a multi-step job. Coding tools went the same way, with recurring tasks that run in the cloud rather than on someone’s laptop. Other providers offer similar features.
What a scheduled agent can do
Think of it as a very literal assistant with a checklist and a start time. Some examples that fit small businesses:
- A café: every evening, compare tomorrow’s pre-orders with stock levels and flag anything likely to run out.
- An immigration consultancy: every morning, check each open file for documents that expire in the next 60 days and draft reminder emails for a person to send.
- A youth sports club: every Monday, list registrations missing a waiver or payment and send the coordinator one tidy summary.
- A clinic: every afternoon, check tomorrow’s bookings for missing intake forms and queue a friendly reminder.
None of these are glamorous. All of them are the kind of task that gets skipped on a busy day.
Where to be careful
An agent that runs without someone watching needs tighter rules than one in a chat window. We build them with a few habits:
- Read first, act later. Start with agents that report and draft. Let them send or change things only once the reports have been right for weeks.
- Narrow access. Give the agent only the data and tools the task needs, not a login to everything.
- A paper trail. Every run should leave a record of what it checked, what it found and what it did.
- A person on the hook. Someone on the team owns each agent and reads its output.
The goal isn’t an AI that runs your business. It is a reliable helper that never forgets the boring check.
Is it worth it?
Ask how often the task is skipped or done late, and what that costs. A missed expiry date on a client’s permit or an unsigned waiver at a tournament costs far more than the agent. A report that nobody reads costs time and gives nothing back.
We usually start with one scheduled job, run it for a month in “draft only” mode and then decide together whether it has earned more responsibility.
Sources
From the DYOL blog. Photo by Winston Chen on Unsplash.
Keep reading
All posts
What’s new in AI this autumn, in plain language
Cheaper fast models, agents that run on a schedule and new rules on labelling AI. A short tour of what changed and what it means for a small team.

AI agents, in plain English
Everyone is talking about agents. What the word actually means, what an agent can safely do for a small business today, and what it can’t.

How to ask AI for what you actually want
Better results from AI rarely need clever tricks. Five habits that help any business owner get useful drafts, summaries and answers the first time.
A free 20-minute conversation, then an honest assessment. You’ll know what it costs and how long it takes, in writing, at a fixed price.
Or email lloyd.delacruz@outlook.com. We reply within 24 hours.
