Rohan Deshpande

Compare Two Options With a Weighted Decision Matrix

Score two options on weighted criteria with the arithmetic shown, then test which weight would flip the result and whether it matches your gut.

At a glance

Best for: Anyone stuck between two jobs, suppliers, tools or houses who wants their reasoning written down and tested, not just a verdict.
Tested on: Claude · Opus 5.5
You fill in: decisionoptionscriteriafacts
You get: Criteria check Salary should mean total pay. A: £62k + up to £6.2k bonus + £4.96k pension, about £73k. B: £58k + £2.32k pension, about £60k, with… (full result below)

Prompt

Help me decide between these options using a weighted decision matrix, but challenge my inputs rather than just doing the maths.

Decision: {{decision}}
Options: {{options}}
Criteria that matter to me, with rough weights if I have them: {{criteria}}
Facts I know about each option: {{facts}}

1. Check my criteria: flag any that overlap (double-counting) and suggest at most one missing criterion. Normalise the weights to sum to 100.

1. Score each option 1-5 on each criterion. Next to every score, give a one-line reason tied to my facts; write "guess" where you're not working from a fact.

1. Show the matrix: criterion, weight, score per option, weighted score per option and totals. Show the arithmetic.

1. Sensitivity: which single weight or score, if changed, would flip the result, and by how much would it need to change?

1. Gut check: if the totals disagree with what I seem to want, based on how I've described this, say so plainly.

Finish with a recommendation in two sentences and the one fact I should confirm before committing.

Fill in

{{decision}}: The decision in one linee.g. Which job offer to accept
{{options}}: The options, labellede.g. A) Senior Data Analyst at a bank. B) Analytics Engineer at a climate-tech startup
{{criteria}}: What matters, with rough importancee.g. salary (high), learning (high), security (medium), flexibility (medium), mission (low)
{{facts}}: What you actually know about each optione.g. A: £62k + 10% bonus, 50-min commute. B: £58k, remote, 20 months of runway

Example input

decision: Which job offer to accept.
options: A) Senior Data Analyst at a bank, hybrid (3 days a week in the Manchester office). B) Analytics Engineer at a 40-person climate-tech startup, fully remote.
criteria: salary (high), learning/growth (high), job security (medium), commute/flexibility (medium), mission (low, but I care).
facts: A: £62k + up to 10% bonus, 8% employer pension, 50-minute commute each way, stack is SQL and Power BI, team of 12 analysts, stable. B: £58k, no bonus, 0.1% share options, 4% employer pension, uses dbt, Snowflake and Python, I'd be 1 of 3 data people, raised a Series A 14 months ago, 20 months of runway. I keep reading about dbt and want to move towards engineering.

2.2k0

Generated Outputs (1)

3 hours ago

Claude

Opus 5.5

3 hours ago

Generated Output

1. Criteria check

• Salary should mean total pay. A: £62k + up to £6.2k bonus + £4.96k pension, about £73k. B: £58k + £2.32k pension, about £60k, with options valued at zero. That's roughly a £13k gap if A's bonus pays in full (a guess).

• There's no double-counting, though "learning" and "wanting to move towards engineering" are the same thing, so I've scored them once.

• Missing criterion: your manager and team quality. You haven't mentioned it, so I haven't scored it.

• Weights (high 3, medium 2, low 1, normalised): salary 27, learning 27, security 18, flexibility 18, mission 10.

2 and 3. Matrix

| Criterion | Wt | A | B | A x Wt | B x Wt |
|---|---|---|---|---|---|
| Salary | 27 | 5 (about £73k) | 3 (about £60k) | 135 | 81 |
| Learning | 27 | 2 (SQL and Power BI; guess you know these) | 5 (dbt, Snowflake, Python) | 54 | 135 |
| Security | 18 | 5 (established bank) | 2 (20 months of runway) | 90 | 36 |
| Flexibility | 18 | 2 (about 5 hours' commuting a week) | 5 (fully remote) | 36 | 90 |
| Mission | 10 | 2 (guess) | 5 (climate; you said you care) | 20 | 50 |
| Total | 100 | | | 335 (3.35) | 392 (3.92) |

4. Sensitivity

• Weights: moving about 10 points from learning to security flips it. At learning 17 and security 28, A scores 365 and B 362. So B wins unless job security matters more to you than learning.

• Scores: no single two-point change flips it. The closest is B's learning dropping to 3, which gives B 338 against A's 335, a near tie. That would happen if, as one of only three data people, you ended up doing reporting rather than engineering.

5. Gut check
The totals agree with you. "I keep reading about dbt and want to move towards engineering" is B's strongest point, and your weights reflect it. What they may understate is the £13k gap. If money is tight this year, raise the salary weight and re-run it.

Recommendation
Take B: it matches the direction you want your career to go, and the pay gap is the price of that move. Before you accept, ask B's founders how they plan to extend the 20 months of runway, whether through a new round or profitability, and when.

Why it works

Asking the model to challenge the inputs first (overlapping criteria, a missing one) fixes problems before the arithmetic makes them look objective. Marking every score as fact-based or a "guess" shows how much of the total rests on assumptions. The sensitivity step is the most useful part: knowing that the result flips only if security outweighs learning tells you exactly what to think about. The gut check respects that a matrix which contradicts what you plainly want is telling you something about your weights.

When not to use it

A matrix makes subjective scores look precise; don't treat a 3.9 versus 3.4 as settled. It handles two or three options well but gets unwieldy beyond five. For decisions with a single deal-breaker (a visa, a health need, a partner's job), deal with that first; no weighting captures it.

Comments (0)

No comments yet. Be the first to share your thoughts!