What-if analysis for a decision: testing assumptions one by one
The essentials in 30 seconds
- A what-if analysis tests the assumptions behind a decision by changing them one by one.
- In a simplified 9-dimension model, a supplier delay lowers a score from 0.660 to 0.596.
- The EQOS Asia scale places the STOP zone below 0.618 (EQOS Asia method page, as of October 2026).
- The EQOS Asia simulator offers 9 sliders from 0 to 100 for hands-on tests (EQOS Asia simulator page, as of October 2026).
- A written log of each scenario lets a committee repeat the analysis.
A what-if analysis asks one plain question about a decision: what happens to the answer if this assumption changes? This guide shows how to list the assumptions behind a decision, test them one by one, and group the results into named scenarios, as of October 2026. It builds on the sensitivity analysis of a decision score, which measures how far a single input can move a result, and it adds scenarios and a written record.
On this page
Compare the EQOS Asia offers and request a 30-minute pre-diagnostic.
What is a what-if analysis in a decision?
A what-if analysis changes one assumption in a decision and records what happens to the result. The assumption might be a price, a hiring date, a regulatory timeline, or a rating of team readiness. The rest of the model stays as it was, so the effect of the change is easy to read. The output is a short table of assumptions and results that a committee can discuss in a single meeting.
The value lies in the questions it raises before money is committed. A leader sees which beliefs carry the plan and which ones barely matter. Beliefs that carry the plan earn more evidence, a second source, or a milestone check. Beliefs that barely matter can stay as rough estimates.
How does a what-if analysis differ from a sensitivity test of a decision?
A sensitivity test moves each input across a plausible range and ranks the inputs by the size of the swing. A what-if analysis starts from a story: the supplier slips by a quarter, the new hire arrives early, the currency moves. The story may change several inputs at once, and it receives a name that a committee remembers. Both methods use the same equation, so their results sit on one scale.
The two work as a pair. The sensitivity test shows where to look, and the what-if analysis turns the findings into scenarios that people can picture. Run the sensitivity test first, pick the 3 inputs with the largest swings, and write one what-if scenario around each. The decision then carries a ranked list of inputs and a small set of stories.
Which assumptions behind a decision deserve a what-if test first?
List every assumption the decision relies on, then sort the list with two questions. How much would the result change if the assumption proved wrong? How uncertain is the assumption today? Assumptions that score high on both questions go to the top. A date promised by a partner, a rating based on a single interview, and a cost taken from one quote are typical candidates.
Write each assumption as a sentence with an owner and a source, for example: stability stays at 0.6 because the supplier contract runs for 2 more years. A sentence gives the test a clear target and shows who can update it. Naming one person per line keeps the list alive.
How do you turn what-if tests into named decision scenarios?
Group the tested assumptions into 3 scenarios: a base case, an adverse case, and an upside case. The base case holds the best current evidence. The adverse case moves the 2 or 3 most fragile assumptions in the unfavourable direction together, and the upside case moves them the other way. The EQOS Asia method uses the same 3 shapes for its trajectories: optimistic, median, and adverse (EQOS Asia method page, as of October 2026).
Give each scenario a name that describes the world, such as Supplier delay or Early hire, instead of a label like Case B. A named world is easier to discuss, and it points to the actions that fit that world. Keep each scenario to 2 or 3 changed assumptions, so a reader can trace every shift in the result back to its cause.
What does a worked what-if analysis of a decision look like?
Take a simplified model with 9 equal weights and a geometric mean. All values are illustrations, not EQOS Asia results. The base profile rates coherence and integrity at 0.8, resonance, care, and trust at 0.7, transparency, clarity, and stability at 0.6, and energy at 0.5. The geometric mean of these 9 ratings is 0.660.
Scenario 1, Supplier delay, cuts stability from 0.6 to 0.4 and energy from 0.5 to 0.3. The score falls to 0.596. Scenario 2, Stronger reporting, lifts transparency from 0.6 to 0.8 and energy from 0.5 to 0.7. The score rises to 0.707. The fall of 0.064 is larger than the rise of 0.047, because a product weighs low ratings heavily.
How does a what-if result relate to the decision line at 0.618?
The EQOS Asia scale places the STOP zone below 0.618 and the OK zone above it (EQOS Asia method page, as of October 2026). In the example, the base case at 0.660 sits in the OK zone. The Supplier delay case at 0.596 falls under the line, so in the simplified model a delay of that kind moves the verdict from OK to STOP.
That crossing is the most useful finding of the analysis. It tells the committee which story matters, and it shows how much room the base case leaves: 0.042 above the line. Scenarios that stay on the same side of 0.618 as the base case rarely change the decision. Scenarios that cross the line deserve a mitigation plan, an early-warning signal, or a staged commitment.
How can the EQOS Asia simulator run a what-if analysis on a decision?
The simulator offers 9 sliders from 0 (collapsed) to 100 (perfect). Set the sliders to your base case, then move the ones that carry an assumption and read the live result. The Reset button returns every slider to 50, and the Random button draws values across the 9 dimensions for a quick exploration (EQOS Asia simulator page, as of October 2026).
The prediction view compares 3 trajectories over 6, 12, and 24 months: Now, Without action, and Top 3 levers (EQOS Asia simulator page, as of October 2026). Those labels suit a what-if reading well, because each one is a named world with its own result. You can try a first score on the Λ simulator and write the settings of each run in your log.
How do you record what-if results so a decision committee can repeat them?
Keep one line per scenario: the name, the assumptions changed, the old and new values, the resulting score, and the date. Store the log next to the base case, so anyone can start from the same point. A deterministic engine returns the same score for the same inputs, which lets a reviewer check every line (EQOS Asia method page, as of October 2026).
Review the log at each milestone of the decision. When an assumption turns real, replace the scenario with the observed value and rerun the base case. The EQOS Asia method describes a monthly recompute through Sentinel and a 30/60/90 day lever plan. A 30-minute pre-diagnostic applies the equation to your own assumptions (EQOS Asia method and contact pages, as of October 2026).
See also in this guide
Frequently asked questions
What is a what-if analysis?
A what-if analysis changes one assumption, or a small group of assumptions, in a decision model and records the new result. The output is a table that pairs each assumption with its effect on the score. Leaders use it to see which beliefs carry the plan before they commit money, people, or time.
How many what-if scenarios does a decision need?
Three scenarios cover most decisions: a base case, an adverse case, and an upside case. Add a fourth when a specific event, such as a regulatory ruling, could change the whole picture. A small set lets a committee remember each scenario and discuss it properly in one meeting.
Can a what-if analysis run in a spreadsheet?
Yes. Place the base inputs in one column, copy the column for each scenario, change only the assumptions under test, and compute the score in a row below. Record the date and the source of every new value, and save the file with the decision papers.
Where does the EQOS Asia pre-diagnostic fit?
A free 30-minute video session, scheduled in Singapore time (SGT) and held under NDA, applies the EQOS Asia equation to a real decision. Bring your 3 most fragile assumptions and ask which ones move the score most (EQOS Asia contact page, as of October 2026).
Test your knowledge
1. How many scenarios does this guide recommend as a core set?
- Two
- Three
- Five
Show the answer
Three.
2. What is the base score in the simplified example?
- 0.707
- 0.596
- 0.660
Show the answer
0.660.
3. Which scenario falls below the 0.618 line?
- Stronger reporting
- Base case
- Supplier delay
Show the answer
Supplier delay.
4. What does the Reset button of the simulator do?
- Sets every slider to 50
- Draws random values
- Sets every slider to 0
Show the answer
Sets every slider to 50.
5. What does the log record for each scenario?
- Only the final score
- Name, changes, sources, score, and date
- Only the names of the owners
Show the answer
Name, changes, sources, score, and date.
About the author
Stéphane Jambu
President, EQOS Forecast Asia
Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile
Sources: EQOS Asia method page (eqos-asia.com/method.html), simulator page (eqos-asia.com/simulator.html), and contact page (eqos-asia.com/contact.html), consulted on 22 October 2026. Page updated on 22 October 2026.