Decision Manufacturing Process
Break investment decisions into measurable stages and improve each stage
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 98%
This framework treats an investment organization as a manufacturer whose product is decisions. Instead of judging performance only by eventual returns, the team decomposes its work into distinct stages: sourcing ideas, making portfolio decisions, and continuously revaluing holdings. Each stage receives explicit inputs, assumptions, and measures so the team can identify where judgment added value and where it caused errors. Portfolio construction includes purchases, sales, position weights, and opportunity costs rather than merely deciding whether a company is attractive. Because prices and business evidence continually change, every holding re-enters the process for reassessment. Over time, recorded outcomes create a feedback loop that improves both quantitative analysis and qualitative judgment.
Origin
Daniel Kahneman told Chris Davis that an investment firm is in the manufacturing business because it makes decisions. He advised Davis to divide decision-making into measurable steps and study where each step adds or subtracts value.
Core principles
- 01Treat decisions as the primary output
- 02Separate sourcing, selection, and revaluation
- 03Make inputs and stages measurable
- 04Study where each stage adds or destroys value
- 05Update decisions as prices and evidence change
How to run it
- 1
Define the output
State the exact decision the process must produce, such as whether to buy, sell, hold, or resize a position.
Pro tip Frame the decision in terms of the underlying business rather than short-term price movement.
Watch out Do not confuse collecting information with making a decision.
- 2
Source candidates
Create a distinct idea-sourcing stage that surfaces businesses worthy of deeper investigation.
Pro tip Track which sources repeatedly produce useful candidates.
Watch out Do not let an exciting source bypass the remaining stages.
- 3
Analyze the investment
Estimate the business's economics, risks, prospective cash generation, and valuation using explicit inputs.
Pro tip Record assumptions so later evidence can be compared with the original thesis.
Watch out Avoid unsupported precision in uncertain forecasts.
- 4
Construct the portfolio
Compare the candidate with existing holdings and available alternatives. Decide purchases, sales, and relative weights through the lens of opportunity cost.
Pro tip Evaluate every allocation as a choice between competing uses of capital.
Watch out A good business is not automatically the best portfolio addition.
- 5
Revalue continuously
Revisit holdings as prices, company results, and other evidence change. Update the decision rather than defending the original conclusion.
Pro tip Use scheduled reviews alongside event-driven reassessments.
Watch out Do not treat the original purchase decision as permanent.
- 6
Audit the process
Compare outcomes with the recorded reasoning and identify which stages added or removed value. Feed transferable lessons back into the process.
Pro tip Distinguish process quality from an outcome produced by luck.
Watch out Do not rewrite the original rationale after seeing the result.
In the wild
A portfolio team records why it owns a bank, the expected credit performance, and its valuation range. After quarterly results reveal stronger earnings but deteriorating loan quality, the team updates both inputs. It then compares the revised return with other holdings before deciding whether to retain or reduce the position.
→ The decision reflects current evidence and opportunity cost rather than attachment to the original thesis.
Common mistakes
Measuring only returns
A profitable outcome can hide a poor process or lucky result. Review the inputs and stages that produced it.
Collapsing every stage into intuition
When sourcing, analysis, valuation, and sizing remain implicit, the team cannot identify where value was gained or lost.
Ignoring new inputs
A sound original decision can become wrong when prices or business evidence change.
Is it for you?
Best for
It is best for investors and teams making repeated, consequential decisions under uncertainty.
Not ideal for
It is not ideal for trivial or irreversible one-off choices that cannot generate useful feedback.
From the transcript
“you're in the manufacturing because I said, well, what do we make? And he said decisions?”
“I would break down the process of making a decision into as many steps and inputs that are measurable as possible”
“there's the sourcing of ideas. Then there's exactly what you're asking about, the sort of portfolio construction, the investment decisions”
From the episode
Christopher Davis Discusses Smart Value Investing