Every Monday I take one popular belief about AI agents and test it against my own fleet. This week’s is the one on almost every landing page: “Your AI agent works 24/7.”
I run an AI agent 24/7 for this business. At least, that’s how I used to describe it. This morning I measured it properly. Over the last 30 days my agent was actually working for 59 hours. That’s 8.2% of the month. The other 91.8% it was asleep, and the business still published, posted, answered and followed up every single day.
The receipt: 402 runs, 59 hours, 8.2% awake
I timed every job from September 26 to October 25 using its log file: when it started and when it last wrote anything.
- 402 runs across 20 different jobs. Blog posts, social scheduling, inbox triage, outreach, a nightly health check.
- The median run took 6.6 minutes. 90% finished inside 18 minutes. The longest took 37.
- A typical day used about 88 minutes of agent time. The busiest day used 4.4 hours.
- Every night there’s a quiet stretch of up to 7 hours where no agent runs at all.
- My inbox agent checks email twice a day. That’s 60 check-ins in a month, 4.5 hours in total, about 3 minutes each. When I counted last month’s emails, about 2% actually needed a reply. A round-the-clock watcher would have spent its hours waiting.
So what does run 24/7? A 429-line Python scheduler. It has been up for almost two and a half days straight and has used 19 seconds of CPU and 13 MB of memory in that time. Its whole job is to look at the clock, start the right agent at the right minute, and go back to waiting. That’s the “24/7” part. It’s an alarm clock, not a worker.
Why the myth costs you money
“AI agent 24/7” sells the idea of a tireless employee who never clocks out. That idea leads people to build the wrong thing:
⚡ GET THE AI EDGE
Weekly AI tips that actually save you time and money. No fluff, no hype — just what works.
- They pay for idle time. An always-on agent burns tokens and server hours while nothing is happening. A scheduled one only costs money while it works. (I broke down what running an agent actually costs a while back.)
- They can’t tell when it’s done. A job with a start and an end can report “finished, here’s what I did”. A loop that never ends just drifts. Short runs are also why I can put a hard timeout on every agent.
- They skip the real design question: when does your business actually need something checked?
There is one honest exception: things that arrive on their own clock, like inbound phone calls or a live chat. Even there the agent isn’t “working” between calls. It’s triggered by an event. My AI receptionist cost breakdown covers that case.
Your action step this week
Before you buy or build an “always-on” agent, write down three times a day when something in your business genuinely needs attention. For most small businesses it looks like this: first thing in the morning, after lunch, and end of day. That list is your agent’s schedule. Start with one job at one of those times and let it finish. Then add the next one.
If you want to see how the scheduling layer works under the hood, my Claude Code loop guide walks through the cron loop behind 1,800+ unattended runs.
The verdict: myth busted. A good AI agent doesn’t work 24/7. It shows up on time, does the job in minutes, and leaves a note. Want help mapping your business onto a schedule like this? Book a strategy session and we’ll find your three check-in times together.
See you next Monday.
Jon
📥 FREE: THE AI PLAYBOOK
The exact tools and workflows I use to run a one-person agency. 25 years of marketing experience distilled into an actionable guide. Yours free.
AI & Digital Marketing Strategist
With over 25 years in digital marketing, e-commerce, sales, and international business, Jon Jones is a seasoned entrepreneur dedicated to empowering small businesses through innovative AI solutions. His journey began in investment banking on Wall Street, guiding companies through capital acquisition and complex markets, and has expanded to include expertise in web development, brand design, product development, and courses and training.
Fluent in Mandarin and with a rich foundation in both finance and tech, Jon launched JonJones.ai to provide accessible, transformative tools for small businesses to streamline operations, generate leads, and build sustainable growth. His mastery of AI, automation, and digital strategy has enabled numerous businesses to bridge the gap between traditional practices and modern technology, ensuring they thrive in today’s digital landscape.
