The Two-Offer Decision Matrix: Score the Option That's Actually Better
Stuck between two choices? Let Claude build a weighted decision matrix that scores each on what you genuinely value β and cuts through the headline price to the real winner.
Two job offers. Two apartments. Two cars, two loans, two quotes. You've been going back and forth for days, and the maddening part is that each one wins on something β this one's cheaper, that one's closer, this one feels right, that one's the "sensible" choice. Your brain keeps swapping which factor matters most depending on which option you're looking at, and you get nowhere.
The reason it's so hard is that you're trying to compare things that aren't directly comparable, weighted by preferences you haven't actually made explicit. So the decision gets made by whichever one you happened to think about last, or by the loudest single factor β usually price β even when price isn't really what matters most to you.
A weighted decision matrix fixes this, and it's the classic tool smart decision-makers use: list what matters, decide how much each factor weighs, score each option, do the maths. It's powerful and almost nobody bothers, because building one by hand is tedious. Claude builds it in seconds β and, crucially, pushes back on your weightings so you're not just dressing up a gut call in a spreadsheet. You decide what matters. It reveals which option actually delivers it.
What You Need
- A free or Pro account at claude.ai β free tier is fine.
- Two (or more) options you're deciding between, with the key facts about each.
- 15 minutes and a willingness to be honest about what you actually value.
- Recommended: a Claude Project if this ties to a bigger goal (career, money), so Claude can factor in your priorities.
Step 1: Define what actually matters (and how much)
The magic is in the weighting. Before scoring anything, get clear on your criteria β and let Claude challenge you.
You are a rigorous decision analyst. I'm choosing between two options and I want to build a weighted decision matrix so I decide on what I actually value, not just on price or gut feel. I'm deciding between: [OPTION A] and [OPTION B]. Here's the context and why this decision matters to me: [BRIEF CONTEXT]. Help me set up the matrix: 1. Suggest the 6β8 criteria that genuinely matter for a decision like this (include the obvious ones like cost, but also the easy-to-forget ones β time, stress, flexibility, long-term value, how it makes me feel, risk). 2. Ask me to weight each criterion (e.g. out of 100 total, or 1β5 importance), and push back if my weightings look inconsistent with what I've said matters to me. 3. Don't score anything yet β just lock in the criteria and my weightings first. Be willing to tell me if I'm underweighting something I'll regret ignoring, or overweighting price.
What just happened: you've just made explicit the thing that was swirling around unspoken β what actually matters to you in this decision, and how much. Having Claude challenge your weightings is the safeguard against the classic trap of building a "rational" matrix that's secretly rigged to confirm what you already wanted. The criteria are now honest.
Step 2: Score both options and reveal the winner
Now feed in the facts and let the matrix do its work.
Here are the details for each option against our criteria: OPTION A: [facts for each criterion β cost, location, terms, whatever applies] OPTION B: [facts for each criterion] Now: 1. Score each option on each criterion (say, 1β10), with a one-line justification for each score so I can sanity-check your reasoning. 2. Apply my weightings to produce a weighted total for each option. 3. Present it as a clean matrix table: criteria down the side, both options scored, weighted totals at the bottom. 4. Declare the winner on the numbers β and tell me by how much (a clear win, or basically a tie?). Be objective. If a score is a judgement call or based on an assumption, flag it so I can correct it.
What just happened: the fog is a table now. Every criterion is scored, weighted by your real priorities, and totalled β so you can see not just which option wins, but why, and by how much. And because Claude justified each score, you can spot and correct any assumption it got wrong, rather than trusting a black box.
Step 3: Pressure-test the result against your gut
The matrix is a tool, not a tyrant. The final, powerful move is to check the number against your instinct β because the gap between them is information.
Now pressure-test the result: 1. Run a quick sensitivity check: if I adjusted my top one or two weightings a bit, would the winner change? Tell me how robust the result is, or whether it's balanced on a knife-edge. 2. Here's my gut reaction to the winner: [HOW I ACTUALLY FEEL ABOUT THE RESULT β relieved? disappointed? surprised?]. If my gut disagrees with the matrix, help me work out why β am I missing a criterion I never named, or is my gut just resisting the sensible choice? 3. Give me your honest final read: does the matrix reflect a genuinely better decision, and what's the one thing I should double-check before I commit? The goal is a decision I can commit to without second-guessing, not just a number.
What just happened: this is the step that turns a spreadsheet into a real decision. If your gut is relieved, the matrix confirmed what you knew β commit. If your gut is disappointed by the "winning" option, that's a signal you've either left out a criterion that secretly matters or you're resisting the sensible call β and now you know which. Either way you commit with clarity instead of lingering doubt.
Bonus: The tie-breaker
When the matrix comes back nearly even, use this.
My two options scored almost the same on the matrix. Help me break the tie without overthinking it. Ask me: which one I'd be more upset to lose if it disappeared tomorrow, which one's downside I could live with more easily, and which I'd regret not choosing in a year. Based on my answers, give me a clear recommendation β when it's this close, the tie-breaker is usually about regret and risk, not points.
Payoff: when options score evenly, more analysis just spins in circles. Switching from "which scores higher" to "which would I regret less" is the psychologically smart way to break a genuine tie β and it usually surfaces a clear answer fast.
Bonus: The reusable decision template
Turn this one decision into a tool you keep.
Save me a reusable version of this decision matrix as a simple template I can use for future either/or choices. Give me: the criteria structure, the weighting method, the scoring scale, and the 3-step process (set criteria β score β pressure-test against gut). Write it so I can paste it back to you next time and just fill in a new decision. Keep it short enough to actually reuse.
Payoff: you'll face this kind of either/or choice dozens of times. Having a reusable framework means you never again spend three days going back and forth β you run the template and get to a confident answer in fifteen minutes.
Claude didn't decide for you β you set the criteria, you chose the weightings, and you had the final say when the number met your gut. All it did was take the tedious, error-prone maths of weighing incomparable things off your plate, and hold you honest about what you actually value. That's the whole idea: it structures the decision so your judgement can do its job clearly, instead of drowning in back-and-forth. The choice was always yours. Now you can make it and mean it.
Ready to go deeper?
Build your own AI assistant in one afternoon.
A step-by-step guide to building a custom AI assistant that actually fits your life and work β no coding, no overwhelm, just a system you can use from day one.
Get the guide β $47Daily Stack
The tools behind the work.
The tools that have earned a permanent spot in my workflow.
Claude
Anthropic's flagship product and the tool that almost every resource on this site is designed for.
ManyChat
The easiest way to build automated Instagram DM flows, comment replies and lead capture without coding.
Planoly
My go-to for planning, scheduling and visually mapping out social content across platforms.