---
title: "Your 30-60-90 AI plan as a manager (without promising what you don't control)"
slug: plan-30-60-90-ia
lang: en
situation: liderar
format: plantilla
minutes: 20
updated: 2026-10-19
---

# Your 30-60-90 AI plan as a manager (without promising what you don't control)

> For managers asked for "an AI plan" for their area who want to commit to deliverables, not to results that depend on others.

## When to use it

- Your director asked for an AI plan for the quarter and you do not want to show up with a list of tools.
- You just took over an area and want to sort out what is already in use before proposing anything new.
- There is pressure to show progress and you need dates that actually depend on you.

## The sheet

Area: [ ] · Responsible manager: [name] · Start: [date] · Final review: [day 90 date]

| Stage | Objective | Deliverable | Metric | Who reviews it |
|---|---|---|---|---|
| Days 1–30: understand | Know what is used today, with which data, and what the chosen process costs. | Inventory of tools in use, a baseline for one process and minimum usage rules in writing. | Inventory complete (yes/no); baseline measured with a source. | [director or committee] |
| Days 31–60: test | Test a single use case with a metric defined before it starts. | One scoped pilot: scope, participants, start and end date. | The baseline metric, measured again when the pilot closes. | [process owner] |
| Days 61–90: decide | Decide with evidence whether to continue, adjust or switch off. | One-page memo: result, real cost, what did not work and the decision. | Decision made and recorded on time (yes/no). | [director or committee] |

Illustrative example: "Days 31–60: pilot of draft replies in supplier support with two analysts. Metric: average response time, compared with the days 1–30 baseline."

What you promise and what you do not:

- **You promise:** the inventory, the baseline, the pilot, the measurement and the memo, on time.
- **You do not promise:** a savings percentage, adoption by other areas, or IT and procurement decisions that are not yours.

## How to fill it in 20 minutes

1. Put the three checkpoint dates in your director's calendar before filling in anything else.
2. In stage 1, pick a single process for the baseline. Measuring everything is the most common way of measuring nothing.
3. In stage 2, define the metric and the success criterion before the pilot starts, not after.
4. In stage 3, write down "switch off" as an acceptable outcome.
5. Name who reviews each stage. Without a reviewer, the date moves on its own.
6. The most common mistake: committing on day 1 to a day 90 result that depends on budget, IT or another area.

## The line that is almost always missing

"Switch off" as a valid day 90 decision. A plan where the pilot can only end in "scale up" is not a plan; it is a commitment made in advance. If you write from the start that the memo may recommend continuing, adjusting or switching off, the measurement becomes credible and your name is protected by the evidence, not by the promise.

## If this becomes a big decision

If the day 90 memo ends in an investment or a change of roles that carries your name, the free tools at vicsol.org/decision-clarity/ help you structure it. They do not replace a formal review.

---

Vic Sol · Decision Desk · vicsol.org/resources/ · Free to use with attribution.
This sheet structures a conversation; it is not legal, financial or audit advice.