AILOS← All modelers

AI Leadership Operations / 04

Pro

Where in this process does AI actually help?

Rank four process steps by net lift after rework risk — hedged by your situation. Insert at the top one or two.

About this tool

About this modeler

AI rarely belongs at every step of a workflow — usually one or two steps drive almost all of the lift. This modeler ranks up to four process steps by net AI lift after rework risk, so you insert AI where the math says it pays back and leave the rest alone. Use it when redesigning a process, scoping an AI build, or arguing against an over-engineered AI plan.

When to use it

  • Scoping an AI build before engineering starts
  • Redesigning a knowledge-work process
  • Cutting an over-engineered AI plan back to what pays

Related tools

Blast radius

Reversibility

Sensitivity

Novelty

Who owns it

Owner seniority

Your numbers

The process

Volume, cost, and what a mistake costs.

Step 1

Time, error rate, and how well AI fits this step.
How well AI fits this stepAI fit

Step 2

Time, error rate, and how well AI fits this step.
How well AI fits this stepAI fit

Step 3

Time, error rate, and how well AI fits this step.
How well AI fits this stepAI fit

Step 4

Time, error rate, and how well AI fits this step.
How well AI fits this stepAI fit

Pro modeler

Read the framing free. Run the numbers on Pro.

Inputs are locked on the Free plan. Upgrade to run every modeler with your own numbers.

Upgrade to Pro →

Sample numbers shown. Upgrade to Pro to edit and save your own.

Annual net lift from top insertions

$154,440

Insert AI at Step 4 + Step 1 only. That's about $154.4K a year of net lift after rework risk.

Insert at Step 4 + Step 1 · $154.4K/yr

Annual lift: cautious / top-2 / all positive steps

Pessimistic
$65,146
Realistic
$154,440
Optimistic
$196,872

Key numbers

Step 1 · high fit · net/wk$882
Step 2 · med fit · net/wk$816
Step 3 · low fit · net/wk-$0
Step 4 · high fit · net/wk$2,088
End-to-end · weekly cost$13,368
Process savings rate22.2%

What to do next

  1. 01

    Pilot AI at Step 4 (high fit) for 8 weeks.

    Owner · the process owner
  2. 02

    Then add AI at Step 1 (high fit) for 8 weeks.

    Owner · the process owner
  3. 03

    Hold the rest of the steps as-is until you've measured the first insertion.

    Owner · the process owner
  4. 04

    Re-rate fit + error rates after 8 weeks before expanding further.

    Owner · the process owner

Save this scenario

Create a free account to name, save, and revisit this run from your dashboard.

Save it →

Plain-English read

Baseline read

4-step process running 80 times/week. End-to-end loaded cost: $13.4K/wk.

Ranking each step by net weekly lift after rework risk (hedged 1.3× by your situation): Step 4 $2,088 · Step 1 $882 · Step 2 $816 · Step 3 -$0.

Insert AI at Step 4 + Step 1 only. Annual net lift: $154.4K (22.2% of end-to-end).

Rule: don't AI-ify all four. Insert, measure 8 weeks, then expand.

Recommended process

  1. 01

    Insert AI at Step 4 (high fit)

    Owner the process owner

    Net lift $2,088/wk after 1.3× rework hedge from situation.

  2. 02

    Then insert AI at Step 1 (high fit)

    Owner the process owner

    Net lift $882/wk after 1.3× rework hedge from situation.

  3. 03

    Hold the rest of the steps as-is for 8 weeks

    Owner the process owner

    AI-ifying everything multiplies failure modes. Insert, measure, then expand.

  4. 04

    Re-rate fit + error rates after 8 weeks

    Owner the process owner

    Steps that failed because of context (not the model) may move to high-fit later.

Share this decision

  • 80 runs/wk × 4 steps; end-to-end $13.4K/wk loaded.
  • Step ranks (net weekly lift): Step 4 $2,088 · Step 1 $882 · Step 2 $816 · Step 3 -$0.
  • Top insertion: Step 4 (high fit). Lift factor 55% of step time.
  • Annual net lift from top 1–2 insertions: $154.4K (22.2% of end-to-end).
  • Situation: blast: med · rev: costly · sens: routine · nov: trodden · owner: mid · reviewers: 1 (rework ×1.25).