From Subjective Data to Defensible Decisions.
Our platform transforms complex, qualitative data into clear, data-driven rankings. Prioritize product roadmaps, evaluate startup ideas, or select vendors with a transparent, scientifically robust process.
Your decisions stay private. No credit card required.
1. Add Your Items
Paste a list, import a CSV, or type items one by one. Add criteria to capture what truly matters.
2. Compare & Score
Answer one question at a time: A or B? Pairwise battles, triage passes, or weighted cardinal scoring — you only ever hold two things in your head, so decision fatigue never sets in.
3. Get Clear Rankings
Instantly see a data-driven, defensible ranking. Export to CSV or PDF, or share a public link.
This is not a vote.
It's structured personal judgement.
The biggest misunderstanding: people expect a majority-wins poll. What actually happens here is closer to science — a rigorous process that externalises your reasoning, not the crowd's preference. Traditional systems either discard how strongly you feel (plurality) or force your conviction into a preset linear scale (Borda count). Here, the distance between your ranked items is never predetermined — it emerges from how you actually compare them, one pair at a time.
Traditional voting systems
- Plurality (most common): only #1 counts — moderate and passionate support are identical
- Borda / ranked-choice: fixed linear decay applied the same to everyone — 1st always equals n points regardless of how strongly you feel
- How much you prefer A over B is either discarded or force-fitted to a preset formula
- Position #1 carries the same meaning for everyone in the pool
- Gut feel has no structural place in the result
- Groups: individual conviction patterns are averaged or summed away
What this platform does
- Score gaps emerge from how decisively you actually chose — not from a preset formula
- Consistently beating A by large margins produces a large gap; close calls cluster scores together
- The distance between your preferences is yours — nobody sets the scale for you
- Position #1 is meaningful only in your context and only to your weights
- Subjective criteria ("I feel like…") are first-class, weighted inputs
- MCDA criteria weights: set explicitly by you, your team, or a proven framework
On group ranking: when multiple people rank the same list, the platform pools everyone's pairwise comparisons into one Bradley-Terry calculation — so stronger, more consistent patterns of choosing naturally carry more signal than a single mild preference. Per-user voice weighting (e.g. by share percentage) is on the roadmap but not yet in v1.
Same 5 business ideas. Two founders. Completely different score distributions — both correct.
Alex
2 ideas dominate, rest fall off a cliff
57-point gap after #2. Alex has two clear winners and dismisses everything else.
Sam
5 ideas all feel genuinely competitive
Gradual slope across all five. Sam sees real potential everywhere and respects the tail.
What a traditional poll gives you — averaging Alex & Sam's scores:
How this platform resolves it — two people, two intact results:
✓ Alex's result — preserved exactly
Cliff intact. His decisive split between top-2 and the rest is honoured.
✓ Sam's result — preserved exactly
Gradual slope intact. Agency stays at #1 — where Sam put it.
But what if the group needs one answer?
Say Alex and Sam are co-founders and need to pick one idea to start. Traditional voting averages their scores (shown above in amber) — destroying both people's conviction patterns. The platform has a different mode: pool all pairwise comparisons from both users into one Bradley-Terry calculation. The result is not an average. It's a ranking where consistent, decisive choosing carries more weight than mild preference.
Traditional poll average
Sam's conviction erased, Alex's cliff gone
Pooled BT group ranking
Conviction strength from both shapes the result
Subjective criteria
"It excites me", "fits our culture", "feels risky" — these are real inputs. Assign them a weight alongside hard numbers and your ranking reflects the full picture, not just the measurable part.
Objective + subjective, together
Revenue potential and gut-feel alignment rarely sit in the same spreadsheet. Here they do. Weighted MCDA lets apples and oranges coexist in one defensible score.
Expert weight templates
Why start from scratch? Apply proven criteria frameworks — RICE, SWOT, KPI scoring (à la Daniel Priestley), and others — and make decisions the way experienced practitioners do, from day one.
Scientific, not democratic
The process is systematic and repeatable. The output is yours alone. Two people running the same list reach different #1s — and that's not a bug, it's the whole idea: your correct answer, rigorously derived.
Match the method to the decision
Some decisions just need a good night's sleep. Some need a whiteboard, a meeting, or a team vote. Some need to be staged — decide phase one, then phase two. This tool handles a specific kind: decisions complex enough that you can't hold all the tradeoffs in your head at once, personal enough that crowd opinion won't resolve them, and important enough to deserve a rigorous process. Pairwise comparison converts "rank these 20 items" — paralysing — into "A or B?" repeated until the ranking emerges from accumulated judgement. You can pause, step away, and the partial result holds.
AI can guess what you want.
It doesn't know your life.
Your entire journey — every relationship, failure, obsession, and hard-won lesson — is the complete context for any decision you face. No model has access to that. And for the decisions that truly matter, that context is everything. AI can surface patterns and probabilities. Only you can feel what clicks.
Your Experience Is the Model
Should you invite John A. or John B. to a small, once-in-a-lifetime event? Which business idea fits your strengths, not just the market? AI can pattern-match preferences — it cannot feel what resonates. For decisions rooted in identity, passion, and relationship, only you have the right data.
The Striking Comparison
Some 1:1 battles feel instant. Others stop you cold for an hour. That unexpected pause is the decision revealing hidden depth — an option matters far more (or less) to you than you realised. That moment of genuine hesitation is the insight. No AI prompt will surface it; only you living the comparison will.
AI Does the Research Grunt-Work
Where AI does help: factual groundwork. The platform uses GPT‑4o‑mini to instantly triage any item as Yes / No / Maybe, and GPT‑4o to estimate market sizes as objective criterion values. AI handles the research; you own the judgement.
Big decisions deserve dedicated time. The structure this tool provides is how you spend that time well — not faster, but clearer.
When Does This Actually Help?
Real decisions people use it for — or browse public rankings →
Decisions rooted in your values, relationships, and life priorities — where only your lived experience can provide the right answer.
Limited event invitations
Painful, personal, unavoidable.
20 seats, 60 people who matter. Ranking forces honest 1:1 comparison — the list you end up with is one you can actually stand behind.
A major purchase decision
When your criteria weights keep shifting.
Safety, range, boot space, price — which matters most to you, right now? Pairwise comparison of criteria settles the fuzzy weights before you even look at options.
Community or club priorities
Shared decision, many voices.
Where should the neighbourhood fund go? Which club initiative runs this year? Share a ranking list, collect input, and surface real consensus — not whoever spoke loudest.
Decisions with stakeholders, budgets, and accountability — where a transparent, auditable process beats gut-feel every time.
Which business idea to start now
Too many, can't do all at once.
Rank across market fit, personal passion, capital required, and your edge. As Naval Ravikant notes: big decisions deserve a disproportionate share of your time. This is how you spend it.
Product roadmap prioritisation
Defensible to stakeholders.
Run features through weighted criteria — customer impact, effort, strategic fit. The ranked output explains why something is Q3 vs Q1 without a politics battle in the room.
Vendor or candidate selection
Apples vs. oranges, made fair.
10 vendors, 5 job candidates, 3 agencies — each different enough that direct comparison feels impossible. Shared criteria and weighted scoring make the call traceable and fair.