Home/Leadership/Decision-Making Frameworks

Leadership

advanced📖 7 min read

Decision-Making Frameworks

Also known as: Decision FrameworksDACI FrameworkRAPID FrameworkDisagree and CommitType 1 vs Type 2 Decisions

💡The Concept

Decision-making frameworks are structured approaches to making choices consistently and efficiently. Jeff Bezos's most influential insight: there are Type 1 decisions (irreversible, one-way doors — take your time) and Type 2 decisions (reversible, two-way doors — decide fast and iterate). Most companies treat ALL decisions like Type 1, leading to analysis paralysis. Amazon's research found that 90% of business decisions are Type 2, yet teams spend 70% of decision-making time on them. Using the right framework for the right decision type accelerates organizations by 40-60%.

⚠️The Trap

The consensus trap kills speed. Trying to get everyone to agree before acting leads to 'design by committee' — decisions are watered down to the least objectionable option, not the best one. Amazon's 'Disagree and Commit' principle: you can express disagreement, but once the decision is made, everyone commits fully. Another trap: decision fatigue. Leaders who make 100+ micro-decisions daily have 40% lower decision quality by end of day. Effective leaders build frameworks that push Type 2 decisions DOWN the org chart — decide once how decisions should be made, not making every decision yourself.

🎯The Action

Classify every decision as Type 1 or Type 2 before discussing it. For Type 2 decisions (reversible): set a 48-hour maximum decision time, appoint a single decision-maker (not a committee), and use the 70% information rule — if you have 70% of the data you'd like, decide now. For Type 1 decisions (irreversible): use the DACI framework — Driver (one person responsible), Approver (one person who can veto), Contributors (people who provide input), and Informed (people who need to know the outcome).

⚡Pro Tips

The '10/10/10' rule cuts through emotional decision-making: How will you feel about this decision in 10 minutes? 10 months? 10 years? Most decisions that feel high-stakes in 10 minutes are forgettable in 10 months. This instantly classifies most decisions as Type 2.

Amazon requires a 6-page memo instead of a PowerPoint for major decisions. Writing forces clarity — you can't hide fuzzy thinking behind bullet points. If the decision-maker can't write a clear 2-page rationale, the decision isn't ready.

Track your decision log: date, decision, reasoning, expected outcome, actual outcome. After 3 months, review. You'll find patterns — where you decide well, where you consistently misjudge, and where you over-deliberate on things that didn't matter.

🚫Common Myths

✗Myth: “Good leaders always make the right decision”

✓Reality: Good leaders make decisions QUICKLY and course-correct. Jeff Bezos says he's satisfied making decisions with 70% of the information he'd like — 'if you wait for 90%, you're too slow.' The cost of delay often exceeds the cost of a wrong but fast decision that can be reversed.

✗Myth: “More people involved means better decisions”

✓Reality: Research from Bain & Company shows that for each person beyond 7 in a decision-making group, effectiveness drops 10%. A decision meeting with 20 people is 130% less effective than one with 7. Keep decision meetings to 5-7 people maximum.

📊Real-World Case Studies

📦

Amazon

1997-Present

success

Jeff Bezos's Type 1 vs Type 2 decision framework, combined with the 6-page memo requirement, has become the most studied decision-making system in business. Bezos's 2015 shareholder letter introduced the concept publicly: 'Some decisions are consequential and irreversible — these are one-way doors. But most decisions are changeable and reversible — they're two-way doors.' Amazon's speed of innovation (launching 3,000+ features/year across AWS alone) is directly attributed to pushing Type 2 decisions to small, autonomous teams.

AWS Features Launched/Year

3,000+

Type 2 Decision Autonomy

Team-level (no exec needed)

Decision Memo Max Length

6 pages (no PowerPoints)

Revenue (2024)

$600B+

💡 Lesson: Amazon's decision framework creates asymmetric speed: fast on reversible decisions (90% of all decisions) and careful on irreversible ones (10%). This is why Amazon can innovate like a startup at $600B in revenue — their small teams make thousands of autonomous decisions daily.

Source →

📷

Kodak

1975-2012

failure

Kodak invented the digital camera in 1975 but treated the decision to transition from film as a Type 2 (reversible, take your time) when it was actually a Type 1 (irreversible industry shift). Their leadership commissioned study after study, formed committees, and ran pilot programs for 20 years while the market shifted beneath them. Each year of delay made the transition harder. By the time they committed to digital in 2003, they were 8 years behind Canon and Sony.

Digital Camera Invented

1975 (by Kodak engineer)

Time to Commit to Digital

28 years (2003)

Peak Revenue

$16B (1996)

Bankruptcy Filed

2012

💡 Lesson: Kodak's failure was a decision CLASSIFICATION error. They treated an irreversible market shift as a reversible experiment. When an industry is moving in one direction (film → digital), delay IS a decision — and it's the wrong one. The question wasn't IF they should go digital, but HOW FAST. Their committees and studies burned 20 years while competitors captured the market.

🎮Decision Scenario: The Strategy Debate

You're CEO of a B2B SaaS at $10M ARR. Growth has slowed from 120% YoY to 45% YoY. Your leadership team has 3 competing hypotheses for how to reignite growth, and they're deadlocked after 2 weeks of debate.

ARR

$10M

YoY Growth

45% (down from 120%)

Cash in Bank

$8M

Months of Debate

2 weeks and counting

Decision 1

Three hypotheses: (A) Your product is undifferentiated — invest $2M in building a moat (data advantage). (B) Your GTM is wrong — switch from self-serve to enterprise sales ($1.5M in new sales team). (C) Your market is saturated — expand to a new adjacent market ($3M to build v2 product). Each VP champions one option. You have data supporting all three but none conclusive.

Hold a 3-day offsite with the full leadership team to debate until consensus is reachedClick to reveal →

The offsite produces heated debate but no consensus — each VP doubles down on their position. By day 3, relationships are strained. You leave with a 'compromise' plan that half-funds all three initiatives — the worst outcome since each gets insufficient resources. Two more weeks pass. Total decision time: 5 weeks. Competitors are not waiting.

Decision Time: 5 weeks totalResource Allocation: Diluted across 3 half-baked initiativesLeadership Cohesion: Strained

Apply Bezos's 70% rule: you have enough data to decide. Classify this as a Type 1 decision, write a 2-page decision memo, pick enterprise sales (B) based on the data, and invoke Disagree and CommitClick to reveal →

Correct. You've analyzed the data: enterprise customers have 3x higher LTV and 50% lower churn than self-serve — the unit economics clearly favor going upmarket. You write a decision memo explaining the rationale, present it to the team, and say: 'I want your honest disagreement NOW. In 48 hours, we commit.' Two VPs disagree (thoughtfully). You listen, adjust one element, and commit. Total decision time: 2.5 weeks. Execution begins immediately.

Decision Time: 2.5 weeks (vs 5+ weeks consensus)Resource Allocation: 100% behind enterprise GTMLeadership Alignment: Disagree & Commit = full execution

Decision 2

Three months into the enterprise pivot, results are mixed. You've closed 3 enterprise deals ($400K total) but the sales cycle is 90 days (longer than expected). Meanwhile, your self-serve product is still growing 5% MoM on autopilot. The VP of Product says: 'We should go back to self-serve — enterprise isn't working fast enough.'

The VP is right — 3 months is enough data. Reverse the decision and go back to self-serve focusClick to reveal →

Enterprise sales cycles are typically 90-180 days. Judging a 90-day cycle strategy after 90 days means you've only seen one cycle complete. You abandon a strategy before the pipeline matures. Three months later, the 8 deals in your pipeline (worth $1.2M) would have closed — but you killed the effort. You've now flip-flopped twice, destroying team confidence in leadership decisions.

Team Confidence: Eroded from flip-floppingLost Pipeline: $1.2M abandoned

Set a clear evaluation checkpoint at 6 months with specific criteria: if enterprise pipeline isn't $2M+ at month 6, THEN revisit — but don't reverse a Type 1 decision based on early Type 2 dataClick to reveal →

Correct. You're applying decision rigor: the strategy was a Type 1 decision (major resource allocation shift). Reversing it after 3 months based on early results would be treating it like Type 2. You set measurable checkpoints ($2M pipeline at 6 months) so the team knows exactly when and how the strategy will be evaluated. At month 6, pipeline is $2.8M. By month 9, enterprise ARR is $1.5M. The patience paid off.

Strategy Evaluation: Checkpoint at 6 monthsEnterprise Pipeline (Month 6): $2.8MEnterprise ARR (Month 9): $1.5M

🧪

Scenario Challenge

Your startup needs to choose between two database architectures: PostgreSQL (proven, your team knows it well) or a newer NoSQL option (better at scale but team has no experience). The migration would take 2 weeks either way. Your current database handles today's load fine but you'll hit scaling issues in ~12 months. The CEO wants a decision in today's meeting with all 8 engineers present.

🧪 Challenge a Friend