Why per-seat pricing hides the thing you need to know
A seat licence answers "how many people may open the dashboard". Nobody is asking that. The question is what happens to the bill when an agent that runs hourly starts taking twice as many steps, or when a customer-facing task suddenly gets used.
Per-seat pricing cannot answer it, so vendors add usage caps underneath, and the caps become the real price, discovered when you hit one. Usage pricing is less comfortable to put on a pricing page and it is the only shape that tells you what an automation costs before you switch it on.
The parts of a run
| What | Charged |
|---|---|
| Adding an agent from the library, duplicating, editing, changing a schedule | Free |
| Creating an agent from a written description | The AI’s cost to build it ($0.05–0.50 depending on complexity) |
| A run | The AI’s work, plus a small processing charge |
| Processing per run | A fraction of a cent, even for a heavy run |
| Web search | 0.01 credits per query |
| X (Twitter) reading / posting | 0.01 / 0.03 credits |
| Every other connection to your apps | No fee |
| Spending cap per run | $2.00 by default; the run stops there |
Two things in that table matter more than the numbers.
Editing is free. If iterating on an agent costs money, you stop iterating, and you keep a worse agent because improving it has a price. Creating an agent from a written description is what costs, because it uses the AI.
Most connections to your other apps are free. Reading your Shopify orders or adding a row to your own spreadsheet has no fee. The exceptions are outside services that charge per request, such as X, which carry a small per-request fee, stated plainly rather than buried.
The number to actually compute
Take the agent you are considering and work out one run. Roughly:
The AI’s work + a processing fee + any paid tools it uses
Then multiply by frequency. An hourly agent runs 720 times a month; a daily one runs 30. That factor of 24 dominates every other decision you will make, and it is entirely under your control, most "hourly" automations are answering a question that changes daily.
- Hourly 720
- Every 6 hours 120
- Daily 30
- Weekly 4
A simple reporting agent costs a small fraction of a cent in processing plus the price of the AI’s work. An agent that generates an image costs what the image costs, which is the dominant term the moment one appears. AI prices are on the rates page, per model, because the AI model you choose changes the answer more than anything else here.
Three habits that keep the cost where you want it
The per-call prices are small, so the bill is shaped by how you set an agent up. Three habits do most of the work:
Match the schedule to the need. Ask how fresh the answer has to be, then set the schedule to that. Hourly suits a question that changes hourly; most reports are just as useful daily. It is worth a quick look at the schedule when an agent moves from testing to live.
Let the spending cap do its job. If a connection to another app is down, a run may keep trying. The per-run cap, $2.00 by default, stops it there, so a hiccup stays small.
Try paid tools on a short list first. Web search and image generation are charged each time they run. Start with a few items and read the step-by-step log of the run: it shows exactly how many times each tool was used before you schedule it.
The takeaway
An agent costs a fraction of a cent to a few cents per run, so the schedule is the dial you control. Moving a task from hourly to daily makes it 24× cheaper to run.
The most useful habit is not choosing a cheaper model. It is asking, for each agent, how fresh an answer you need, and setting the schedule to match. You get answers exactly as often as they are useful, and the cost follows.