PaperPMF publishes this comparison and is one of the products discussed.
Synthetic Users and PaperPMF both generate AI participants, but they behave like different research colleagues. One keeps asking questions and building a study. The other runs a fixed test and hands back a distribution.
Choose Synthetic Users when you want interviews, follow-ups, several solutions, summaries, and broader analysis. Choose PaperPMF when you want one product concept scored through a repeatable purchase-intent method. You may quite reasonably use both.
The short answer
Key takeaways
- Synthetic Users fits discovery: interviews, probing, context files, and iterative analysis.
- PaperPMF fits measurement-style triage: one prepared concept and one standardized intent distribution.
- Neither product recruits customers, so human or behavioral follow-up still matters.
Discovery tool or fixed instrument?
| Question | Synthetic Users | PaperPMF |
|---|---|---|
| Main job | Explore problems, solutions, and themes | Screen one product concept |
| Interaction | Interviews and follow-up questions | Fixed elicitation and scoring flow |
| Scale | Multiple studies, audiences, and generated users | 10-response preview or separate 100-response report |
| Output | Transcripts, summaries, graphs, and reports | Intent distribution, reactions, filters, and method details |
| Best moment | When you don't yet know the right question | When the question is already specific |
Sources: Synthetic Users: Core Concepts, Synthetic Users: Generating Reports and Insights, Synthetic Users: Study Summaries, PaperPMF methodology
What Synthetic Users actually covers
Its current docs describe workspaces, projects, audiences, generated users, multi-solution studies, interviews, dynamic follow-ups, summaries, knowledge graphs, reports, and attached files. So no, it isn't merely a handful of fake interviews in a pretty chat window.
That breadth is useful when you're learning how people might frame a problem, what objections deserve a human interview, or which themes should become survey questions. The output is still generated, even when it sounds uncannily plausible.
Sources: Synthetic Users: Core Concepts, Synthetic Users: Generating Reports and Insights, Synthetic Users: Study Summaries, Synthetic Users FAQ
What PaperPMF standardizes
PaperPMF narrows the task to one product and one audience hypothesis. It generates personas, asks for two purchase-likelihood reactions per respondent, and maps those reactions to five-point probability distributions through semantic similarity.
That consistency makes runs easier to inspect than a one-off chat. It doesn't make the personas representative, and it doesn't turn generated intent into sales.
Sources: Maier et al.: Semantic Similarity Rating preprint, PyMC Labs: Open-source Semantic Similarity Rating package, Nuremberg Institute for Market Decisions: Digital twins review, PaperPMF methodology
Use them together without making research soup
- Use exploratory interviews to surface language, anxieties, and missing context.
- Turn those themes into a factual product concept and a tighter audience brief.
- Run a fixed purchase-intent screen without changing the prompt halfway through.
- Take the important surprises to real customers or a live-market test.
The easy choice
Pick Synthetic Users if your next move is “ask why.” Pick PaperPMF if your next move is “score this exact concept the same way.” Pick human research if the answer must come from someone who can actually buy the thing.
Sources: PaperPMF methodology
Sources and verification
Product details are based on official documentation reviewed on August 4, 2026 unless noted. Features and pricing can change; verify them with the provider before making a purchase.
- Synthetic Users: Core Concepts — Vendor documentation.
- Synthetic Users: Generating Reports and Insights — Vendor documentation.
- Synthetic Users: Study Summaries — Vendor-reported product information.
- Synthetic Users FAQ — Vendor-reported grounding, methodology, and data information.
- Maier et al.: Semantic Similarity Rating preprint — arXiv preprint; not a general PaperPMF accuracy result.
- PyMC Labs: Open-source Semantic Similarity Rating package
- Nuremberg Institute for Market Decisions: Digital twins review
- PaperPMF methodology — How the diagnostic works and what its results can and cannot show.
