know.2nth.aiCEO Briefing
AI & automation · The build decision

When is an agent actually the right tool?

“Let’s build an AI agent” is the default ask in 2026. But most of what gets sold as an agent should be something simpler, cheaper, and more predictable — a fixed workflow, not an autonomous one. Knowing the difference is the call that decides your cost, your risk, and whether the thing actually ships. This is the plain-language version of Anthropic’s engineering guidance.

Source · Anthropic, Building Effective Agents 5-min read For decision-makers
01 · The distinction nobody draws

A workflow follows a recipe. An agent improvises.

Both use AI. The difference is who decides the steps — and it changes everything about cost and control.

Workflow

You wrote the recipe

The steps are fixed in advance. AI does the thinking inside each step, but the path from input to output is one you designed and can predict.

Like a kitchen following a set recipe — same dish, every time.

Agent

The AI decides the steps

You give a goal; the AI plans its own path, picks its own tools, checks its work, and keeps going until it judges the job done. Powerful — and less predictable.

Like a chef told “make me dinner” — flexible, but you can’t predict the bill.

Why it matters to you: a workflow has a bounded, forecastable cost and behaves the same way every run, so it’s easy to trust and audit. An agent trades that predictability for flexibility — more AI calls, a variable bill, and the risk that one wrong early step throws off everything after it. Neither is “better.” They’re tools for different jobs, and the expensive mistake is using an agent where a workflow would do.

02 · The ladder

Climb only as high as the job needs.

The rule from Anthropic’s engineers is simple: use the least complex option that solves the problem. Each rung up buys flexibility and costs predictability. Start at the bottom.

1

A single good prompt

One well-crafted instruction, maybe with examples. Astonishingly often, this is the whole answer. Cheapest, fastest, most predictable.

2

A workflow (fixed steps)

Chain a few prompts, route by category, run checks between steps. Predictable cost, auditable path. Most business automations live here.

3

A coordinated workflow

A “manager” step that splits work across several AI calls and combines the results. More moving parts, still designed by you.

4

A true agent

Hand over the planning. Reserve for open-ended jobs where you genuinely can’t script the steps in advance — and wrap it in guardrails.

The one-line test

Can you write down the steps in advance? Then it’s a workflow — build it that way and enjoy the predictable cost. Only when the steps genuinely depend on what the AI discovers along the way do you need an agent.

03 · The decision

Four questions before you greenlight an agent.

Can the steps be scripted?

If you can draw the flow on a whiteboard, it’s a workflow. Agents are for when the path can’t be drawn in advance.

Is the cost worth the flexibility?

Agents make many more AI calls. If a workflow gets you 90% there at 10% of the cost, the workflow wins.

Can you check the work?

Agents are safe only where each step produces verifiable output — a test passes, a number reconciles — not where errors hide.

Is there a safety net?

Autonomy needs guardrails: limits, human checkpoints, and the ability to undo. No harness, no agent.

The honest default

If you’re unsure, you want a workflow. The most common and expensive failure in 2026 isn’t “our agent wasn’t smart enough” — it’s building an unpredictable agent for a job a fixed workflow would have done cheaply, reliably, and with an audit trail. Sophistication is not the goal. The job done reliably is the goal.

Keep reading

Related

The take

Don’t buy an agent because the word is in the air. Climb the ladder from the bottom, stop at the first rung that does the job, and reserve real autonomy — and its cost — for the problems that genuinely need it.