Compare
WhittleOS vs DimeADozen
The closest positioning to ours in this category — both answer whether something is worth building. The difference is one deep report versus a live candidate set that is sourced, eliminated and ranked.
TL;DR
DimeADozen offers a "Give me an idea" entry point and validates one premise at a time with a 150+ page report and, by its own advertised numbers, typically 600+ citations. That is more references than our run produces, and this page says so in the table rather than around it. WhittleOS Discovery starts with a field: it sources candidates for your niche, ends most against 12 deal-breaker checks, scores what survives against your capital, time and channel, and ranks a shortlist. Want depth on one premise: them. Need to choose among live candidates first: us.
At a glance
Everything in the right-hand column is what their own site advertises, read on 2026-09-19. We have not bought a paid report, so there is no claim here about what one contains beyond what is advertised — including the citations row, which goes against us.
| WhittleOS | DimeADozen(as of 2026-09-19) | |
|---|---|---|
| Core job | Source candidates, end most of them, decide — bound to your founder profile | Generate ideas or validate one at a time; its own site says the validation product is "Built for one decision: should you build this?" |
| How many ideas enter the decision | A live set — Discovery sweeps a niche, eliminates candidates and ranks a shortlist | One idea per validation report; the homepage separately offers a "Give me an idea" entry point and an Ideate tab |
| Volume of citations | A published sample run links 32 of its 56 documented problems, each openable | More, and we will not pretend otherwise: it advertises typically 600+ citations and 100+ named sources in its full report |
| Shape of the deliverable | A verdict, the reasons, a recommended first move and the point at which to quit — short on purpose | A 150+ page report, advertised as depth you would otherwise pay a consultant for |
| Bound to the founder? | Yes — goal, capital, time budget, channel and support tolerance change the verdict | Not stated as an input to the report |
| Who sets the score | The Whittle Score: the model rates 7 areas, code computes the total from fixed weights, and a fixed matrix sets the verdict | A score is returned with the report; the method is not published in a form we could check |
| What happens after the verdict | A 17-mode catalog in total; focused follow-ups include pricing power, support burden, platform risk, pivot wedge, an async validation plan and an MVP scope | The report is the deliverable |
| Pricing model | Credits per analysis, free founder read to start | A free short score, then paid tiers advertised from $9, with a full report listed at $129 |
The row we lose
On volume of citations, they win and it is not close. Their full report advertises typically 600+ citations and 100+ named sources; a published sample run here links 32 of its 56 documented problems. If the number of references is your criterion, stop reading and go there.
What we would add is that the two counts measure different things. Ours is not a bibliography — it is the number of problems in the verdict you can click through to the page they were read on, and every problem we could NOT verify carries a label saying so and pulls the score down. A reference count tells you how much was read. A provenance label tells you which specific claims are load-bearing and which are not.
Detailed comparison
One premise or a field of candidates
This is the difference that survives every other difference. DimeADozen's homepage exposes both a "Give me an idea" entry point and an Ideate tab, while its validation report is framed around one decision and one premise. WhittleOS can check one idea too, but its flagship Discovery mode sweeps a niche you name (or no niche at all), sources live candidates, eliminates weak ones and ranks the few that clear the gate. The distinction is selection across a field versus depth on one premise, not who first authored the idea.
Length versus a decision
A 150-page report is a real deliverable and for some jobs it is the right one — if you need something to read end to end, or to hand to a partner who expects a market document, ours would disappoint you. Our bet runs the other way: a founder deciding whether to spend the next three months needs a verdict, the reasons behind it, a first move and the point at which to stop. Past a certain length a document stops being a decision and becomes a way of postponing one.
Bound to the founder
Their report is about the idea. Ours is about the idea in your hands: goal, capital, time budget, channel and support tolerance are inputs, so the same idea can come back BUILD for a funded founder with sales experience and KILL for someone with eight hours a week and no audience. Both verdicts are correct — an idea is not good or bad in the abstract, it is good or bad for someone.
Who sets the number
On our side the split is published: a model reads the sources and rates each area with its reasoning attached, and code computes the total from fixed weights. The deal-breaker checks are a fixed list, the verdict comes out of a fixed matrix. We cannot describe their method because it is not published in a form we could check, so this page does not characterise it — the row in the table says exactly that and nothing more.
What happens after the verdict
Their report is the deliverable. Here the verdict is the beginning of the follow-ups: pricing power, support burden, platform risk, a pivot wedge when an idea is nearly right, an async validation plan that says when to quit, and an MVP scope that refuses to run on an idea nothing has validated yet. That refusal is deliberate — scoping a build for an unvalidated idea is the most expensive kind of helpful.
Pricing model
(Model shape only — no price war.) Both are pay-per-use rather than subscriptions: they advertise a free short score with paid tiers from $9 and a full report at $129; we sell credit packs spent per analysis, with a free founder read to start. Neither of us is trying to be the cheap one, and the comparison that matters is what the spend buys.
Why there is no receipt here
We have not bought their paid report, so we have not seen one. That means this page cannot and does not say anything about how good their analysis is, how honest its verdicts are, or whether the citations hold up — only what the site advertises and what our own code does. If we ever run our idea through it, the raw output goes here whichever way it lands, the way we published the IdeaProof one.
Who DimeADozen is genuinely better for
- You want the largest pile of citations you can get on one idea — by its own advertised numbers, its report carries far more references than our run does.
- You want a long document: something to read end to end, share with a partner, or hand to someone who expects a market report.
- You want depth on one premise, may also value its separate idea-generation entry point, and do not need a live niche sweep to compare a field of candidates.
- You want a longer track record — it has been running since 2023, and we are younger than that by a lot.
FAQ
- You both say you answer 'should I build this' — what is actually different?
- How many ideas enter the validation decision, and what you are left holding. DimeADozen has both a 'Give me an idea' entry point and a validation flow; its report is framed around one premise at a time. WhittleOS Discovery sweeps a niche, ends candidates against a fixed checklist and ranks the few that survive against your constraints. If you want depth on one idea, that is them. If you need to choose among a live field before going deep, that is us.
- Does DimeADozen cite more sources than WhittleOS?
- By its own advertised numbers, yes. As of 2026-09-19 it states typically 600+ citations and 100+ named sources in a full report; a published sample run here links 32 of 56 problems. We are not going to argue our way out of that row. What we would say is that the counts measure different things: ours is the number of problems you can click through to the page they were read on, and every problem we could not verify carries a label saying so.
- Is a 150-page report better than a short verdict?
- For some jobs, genuinely yes — if you need a document to read through or to show someone, length is the deliverable and ours would disappoint you. Our bet is the opposite one: that a founder deciding whether to spend the next three months needs one verdict, the reasons behind it, a first move and the point at which to stop, and that a long report is a way of deferring that.
- Which one is more honest about a bad idea?
- We cannot answer that for them — we have not run our own idea through their paid report, and we do not publish negatives about a competitor we have not measured ourselves. What we can say about our side is structural: the checklist that ends an idea is a fixed list in code, the weights are fixed, the verdict comes from a fixed matrix, and a thin niche returns zero finalists rather than a consolation shortlist.
- Will the same idea get the same verdict twice here?
- Not word for word, and we will not claim it does. The model's ratings sample fresh on every run and can move an idea a few points. What is fixed is the standard around them: the checks, the weights, the threshold and the matrix are code, not a fresh opinion each time.
- Should I run both?
- If one idea has already emerged — whether you supplied it or another tool did — and the stakes are high, running both is not a silly use of money. Use theirs for the depth and the reference pile; use a run here to find out whether your own capital, time budget and channel survive contact with it.
Before you research one idea — compare the field.
Run a free founder read and see what a founder-bound, kill-biased read looks like. 2 free credits on signup, no card required.