What is the decision-making process? The decision-making process is a structured sequence of steps for moving from identifying a problem to choosing, acting on, and reviewing a course of action. The most common version has seven steps.
The canonical 7-step decision-making process is: identify the decision, gather relevant information, identify the alternatives, weigh the evidence (pros and cons of each), choose among alternatives, take action, and review the decision against the original need. Named alternatives include the DECIDE model (Kristina Guo, 2008: Define, Establish criteria, Consider alternatives, Identify the best, Develop a plan, Evaluate) for managers; the OODA loop (John Boyd: Observe, Orient, Decide, Act) run as a continuous loop for fast competitive situations, where Orient is the crucial step; RAPID (Bain and Company), which assigns decision roles — Recommend, Agree, Perform, Input, Decide — to fix accountability; and the Vroom-Yetton-Jago situational model, which tells a leader how much to involve the team. The most common failure is skipping the final review step. Argumentree supports the process with structured pro/con argument trees that make 'weigh the evidence' explicit, multi-dimensional rating that aggregates into consensus scores, AI extraction of arguments from transcripts and documents, role and department-based access that mirrors decision roles, and a full audit trail that closes the review loop.
The Decision-Making Process
From a problem to a choice you can defend — in seven steps. Plus the named models (DECIDE, OODA, RAPID) for when the basic process isn't enough.
Last updated: 2026-07-02
TL;DR
The decision-making process turns a vague problem into a defensible choice through a repeatable sequence. The classic version has seven steps; specialized models like DECIDE, the OODA loop, and RAPID adapt it for healthcare, fast-moving competition, and unclear accountability. The one step everyone skips — reviewing the outcome — is the one that makes you better next time.
The 7-Step Decision-Making Process
- 1
Identify the decision
Name the real choice you face — clearly enough that everyone agrees what's being decided.
- 2
Gather information
Collect the facts, data, and context relevant to the choice.
- 3
Identify the alternatives
List the realistic options, including doing nothing.
- 4
Weigh the evidence
Lay out the pros and cons of each option and evaluate them on the merits.
- 5
Choose among alternatives
Select the option the evidence best supports.
- 6
Take action
Implement the choice.
- 7
Review the decision
Check the result against the need from step one — and learn from the gap.
The 7 Steps, Worked: Build vs Buy
Abstract steps are easy to nod along to. Here's the same process run on a decision most teams actually face — should we build a feature in-house or buy an off-the-shelf tool?
1. Identify the decision
"Do we build our own analytics dashboard, or buy one?" — not "how do we do analytics," which is a bigger, different question.
2. Gather information
The engineering estimate to build, vendor pricing, our team's spare capacity, and how core analytics is to the product.
3. Identify alternatives
Build, buy, buy-then-customize — or the option everyone forgets: not now.
4. Weigh the evidence
Build = control and fit, but months of engineering. Buy = fast and cheap today, but a dependency and a recurring bill. Lay out the pros and cons of each.
5. Choose
Buy — because analytics isn't our differentiator and speed matters this quarter. Once the evidence is laid out, it points one way.
6. Take action
Sign the vendor, plan the integration, name an owner.
7. Review
Six months on: did it actually save the time we predicted? If the tool became a bottleneck, that's the lesson that sharpens the next build-vs-buy call.
Notice the two steps teams quietly skip: naming the real decision in step 1, and honestly reviewing it in step 7. Everything in between is the easy part.
Named Process Models
The 7 steps are the default. These models adapt the process for specific conditions:
DECIDE — Kristina Guo, 2008
Define the problem · Establish criteria · Consider alternatives · Identify the best · Develop & implement a plan · Evaluate and monitor. A memorable six-step model originally written for healthcare managers.
The OODA Loop — Col. John Boyd
Observe · Orient · Decide · Act — run as a continuous loop. Born in fighter-jet dogfighting, now used in business, cybersecurity, and litigation. Orient is the crucial step, and cycling faster than your opponent is the whole point.
RAPID — Bain & Company
Recommend · Agree · Perform · Input · Decide. Not steps — roles. RAPID fixes the "who actually decides?" problem by naming who recommends, who signs off, who executes, who's consulted, and the single person who decides.
Vroom-Yetton-Jago — situational model
A contingency model that tells a leader how much to involve the team — from deciding alone (autocratic) to reaching a collective decision together (group) — based on how much the decision needs quality, buy-in, and information.
How Argumentree Runs the Process
Most processes break at "weigh the evidence" and "review the decision" — the parts that need structure and memory. Argumentree handles exactly those, built on argument mapping:
"Weigh the evidence," made literal
Every option's pros and cons live in a structured pro/con tree — the messy step 4 becomes visible and shared.
Rate, score, converge
Participants rate arguments; ratings aggregate into net support scores, so "choose among alternatives" follows the evidence.
Decision roles, built in
Roles and departments control who contributes, moderates, and decides — RAPID-style accountability without the spreadsheet.
Closes the review loop
The audit trail preserves the reasoning so step 7 — reviewing the decision — actually happens.
Part of the broader practice of decision making; see also the decision-making models behind these processes and collaborative decision making for running the process as a group, where the final step of reaching agreement is consensus building.
Where the Process Fails
Solving the wrong problem
A weak step 1 poisons everything downstream — define the real decision first.
Skipping the review
Almost nobody runs step 7, so the same mistakes repeat.
Analysis paralysis
Looping 'gather information' forever until the window closes.
No clear decider
Without RAPID-style roles, decisions stall because everyone and no one owns them.
Frequently Asked Questions
What is the decision-making process?
The decision-making process is a structured sequence of steps for moving from identifying a problem to choosing, acting on, and reviewing a course of action. The most common version has seven steps: identify the decision, gather information, identify alternatives, weigh the evidence, choose, act, and review the outcome.
What are the 7 steps of the decision-making process?
(1) Identify the decision you need to make; (2) gather relevant information; (3) identify the alternatives; (4) weigh the evidence — the pros and cons of each option; (5) choose among the alternatives; (6) take action; and (7) review your decision against the need you identified in step one. Skipping step seven — the review — is the most common failure.
What is the DECIDE model?
DECIDE is a six-step model published by Kristina Guo in 2008 for healthcare managers: Define the problem, Establish the criteria, Consider the alternatives, Identify the best alternative, Develop and implement a plan, and Evaluate and monitor the solution. The acronym makes the steps easy to remember.
What is the OODA loop?
The OODA loop — Observe, Orient, Decide, Act — was created by US Air Force strategist John Boyd for fast, competitive, changing situations. It runs as a continuous loop rather than a straight line, and the key step is Orient: making sense of observations through your experience and mental models. Whoever cycles through the loop faster gains the advantage.
What is the RAPID decision-making model?
RAPID, from Bain & Company, isn't a sequence of steps — it assigns decision roles so accountability is clear: Recommend (drives the proposal), Agree (must sign off), Perform (executes), Input (consulted), and Decide (the single person who makes the final call). It solves the 'who actually decides?' problem that stalls organizational decisions.
References & Further Reading
Guo, K. L. (2008). DECIDE: A Decision-Making Model for More Effective Decision Making by Health Care Managers. The Health Care Manager, 27(2), 118-127.
The original source of the DECIDE model.
View source →Boyd, J. R. (1996). The Essence of Winning and Losing. Unpublished briefing slides.
The origin of the OODA loop, from the fighter pilot and strategist who created it.
Rogers, P., & Blenko, M. (2006). Who Has the D? How Clear Decision Roles Enhance Organizational Performance. Harvard Business Review, 84(1), 52-61.
Bain & Company's RAPID decision-roles framework.
View source →Vroom, V. H., & Yetton, P. W. (1973). Leadership and Decision-Making. University of Pittsburgh Press.
The original situational (Vroom-Yetton) contingency model.
Vroom, V. H., & Jago, A. G. (1988). The New Leadership: Managing Participation in Organizations. Prentice Hall.
The revised Vroom-Yetton-Jago model on how much to involve the team.
Run a process you can trust
Make every step visible — weigh the evidence as a group, assign who decides, and keep the record. Bring structure to your process with Argumentree.
Start Free