Strategy with AI: Modelling Three Bets

Strategy with AI: Modelling Three Bets

Key answer

The three-bets method models a capital decision as three scenarios, bear, base, and bull, with a sensitivity view, so the board sees a range and the trigger for each. AI builds the scenarios, runs the sensitivities, and drafts the board memo in minutes; you own the assumptions, the provenance, and the call.

The three-bets method models a capital decision as three scenarios, bear, base, and bull, with a sensitivity view, so the board sees a range and the trigger for each. AI builds the scenarios, runs the sensitivities, and drafts the board memo in minutes; you own the assumptions, the provenance, and the call. The output is a decision with conditions, not a single number waiting to be wrong.

See the three bets diverge#

One starting point, three plausible paths. Isolate any one to read where it lands.

Three bets, one starting point

Y0Y1Y2Y3Y4Bull 181Base 141Bear 108

Bear, base, and bull projected from the same start. Tap a case to isolate it and read where it lands.

The value is not the lines; it is the gap between them. A wide spread tells the board the decision is sensitive and needs triggers; a narrow one says the bet is robust. The probabilistic engine behind a richer version of this is in Monte Carlo and DCF for FP&A.

Why scenario-led planning is the direction#

of organisations will replace bottom-up forecasting with AI by 2028, enabling continuous, scenario-led planning

50% of organisations will replace bottom-upforecasting with AI by 2028, enabling Gartner, Autonomous Finance

Gartner expects that by 2028, half of organisations will replace bottom-up forecasting with AI (a 2023 prediction), enabling continuous, scenario-led planning. Three bets is the executive face of that shift: a standing habit of modelling the range, not a once-a-year set-piece. The deeper FP&A version is AI scenario planning in finance.

The bet you pick matters as much as the method. BCG finds just 5% of firms are AI leaders, 35% are scaling, and 60% are laggards, with leaders delivering 1.7 times the revenue growth of laggards. McKinsey adds that 80% of firms set cost efficiency as the AI goal, yet only 39% report any enterprise EBIT impact: the biggest returns go to those who also bet on growth, not cost alone.

Where AI value actually lands

Leaders (1.7x revenue growth vs laggards)5%Scaling35%Laggards (minimal gains)60%

Most firms capture little; a small group of leaders takes outsized returns. Source: BCG, The Widening AI Value Gap (2025).

Run the three bets#

AI builds the cases; you own the assumptions and the call.

Run the three bets

1Frame the bet2Set driver ranges3Modelbear/base/bull4Sensitivity view5Decide withtriggers

AI builds the cases; you own the assumptions and the call.

Frame the bet, set driver ranges, model bear, base, and bull, run the sensitivity view, then decide with triggers. The discipline is to flex the drivers, not the outputs, so the scenarios are coherent rather than wishful.

Where AI helps, and where it does not#

Where AI helps, and where it does not

Build casesGenerate the three scenariosfast.SensitivitiesFlex the drivers that move theanswer.Board memoDraft the recommendation andrisks.ProvenanceCite the source behind everynumber.

Speed on the build and the memo, not the judgement.

AI builds the cases, runs sensitivities, drafts the memo, and keeps provenance. The assumptions and the decision stay yours. That division is the same one in the executive operating model: AI computes and drafts, humans decide and sign.

Build a three-bets pack on a real decision#

Practical GenAI for Business Leaders ships a 3-Scenario Strategy Pack in Session 5, on a real capital bet of yours, with the board memo to defend it. You leave able to present a range, not a guess.

Key takeaways

  • Model a capital bet as three scenarios, bear, base, and bull, with a sensitivity view.
  • Tie each scenario to the trigger that would make it real, so the board sees decisions.
  • AI builds the cases and drafts the memo; you own the assumptions and the call.
  • Keep provenance on every number so the strategy survives scrutiny.

Questions, answered

What is the three-bets method?
It is a way to present a capital decision as three scenarios, bear, base, and bull, built on a shared set of drivers, with a sensitivity view that shows which assumptions move the answer most. Each scenario is tied to a trigger, so the board sees a range and the conditions for each, not a single fragile point estimate.
How does AI help with strategy scenarios?
AI accelerates the build: it generates the three cases from your drivers, runs the sensitivities, and drafts the board memo with the recommendation and risks. What it does not do is set your assumptions or make the call. Used this way it turns a multi-day modelling exercise into hours, with you in control.
Is it safe to use AI for strategy decisions?
Yes, with provenance discipline. Every number in the model must cite its source, and the assumptions stay human and explicit. AI drafts and computes; the leader owns the inputs and the decision. That separation, plus a board memo that shows the workings, is what makes AI-assisted strategy defendable.
Where is AI value actually landing, cost or growth?
Both, but unevenly. About 80% of firms set cost efficiency as an AI goal, yet only 39% see any enterprise-level EBIT impact, and the biggest returns go to firms that also set growth and innovation objectives (McKinsey, 2025). BCG finds just 5% of firms are leaders, capturing 1.7 times the revenue growth of the 60% who lag, which is why the bet you choose matters as much as the modelling.
How is this different from a single forecast?
A single forecast hides the risk behind one number. The three-bets method makes the range explicit and attaches a trigger to each case, so the board can pre-agree what it will do if the bear or bull case starts to materialise. It turns a guess into a decision with conditions.
AE

Dr. Ahmed El-Shamy

Co-founder, CEO and Dean of Education, Digisoul

Dr. Ahmed El-Shamy is Co-founder, CEO and Dean of Education at Digisoul. He has more than a decade across AI, fraud risk, and FP&A, and teaches Practical GenAI in FP&A bilingually across MENA, the GCC, and Africa, governed by Digisoul's ISO/IEC 42001:2023-certified AI Management System. Read the leadership profile.

Sources

  1. Gartner · by 2028, 50% of organisations will replace bottom-up forecasting with AI (2023 prediction). https://www.gartner.com/en/newsroom/press-releases/2023-03-01-gartner-preditcts-three-ways-autonomous-technologies-will-impact-the-fpanda-and-controller-functions-in-
  2. BCG · The Widening AI Value Gap (5% leaders / 35% scaling / 60% laggards; leaders 1.7x revenue growth) (Sep 2025). https://www.bcg.com/press/30september2025-ai-leaders-outpace-laggards-revenue-growth-cost-savings
  3. McKinsey, The State of AI 2025: 80% set cost goals, only 39% see enterprise EBIT impact. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai
  4. Practical GenAI for Business Leaders (Session 5: 3-Scenario Strategy Pack). https://digisoul.io/ai4x/genai-for-business-leaders/

AI Agent · Built on Claude · Operated on Zoho One


What do you think?

From our blog

Articles & insights

  • Business Leaders
AI OKRs That Separate Adoption from Value
Set AI OKRs that separate adoption from value: usage proves people use the tool; value proves the business changed. Track both, fund against value.
Read more
  • Business Leaders
The Governed GenAI Operating Model for Executives
A governed GenAI operating model for executives: sense, decide, act with a human gate, plus a use-case portfolio, a RACI, go/no-go gates, and OKRs.
Read more
  • Business Leaders
Is Your Data AI-Ready? A Leader’s 5-Dimension Check
A leader's five-dimension readiness check for GenAI: data quality, access, tooling, skills, and governance. Score each Red, Amber, Green before you scale.
Read more