2026 campaign-assessment score
How atypica.ai scored
46 / 100
atypica supports qualitative audience exploration through synthetic personas. Its outputs serve hypothesis development rather than observed campaign measurement.
| Criterion and evidence | Points |
|---|---|
Campaign explanationScout interprets public content around a topic or audience; campaign investigation and multimodal post-level analysis are not established. | 12 / 30 |
Reaction analysisScout observes public content and builds motivation-based personas, with stated qualitative limits. Reaction to specific campaign posts is not established, and simulated responses are not observed reactions. | 8 / 20 |
Finished assessmentDocumented persona panels, group discussions and report sharing support reusable research outputs. Campaign-specific recommendations are not established. | 14 / 20 |
Creator baselineNo usable evidence establishes creator-history or campaign-performance benchmarks; Scout explicitly provides qualitative, non-statistical research. | 0 / 10 |
China and international coverageScout names Xiaohongshu, Douyin, TikTok, X and Instagram and explains public-access, demographic and qualitative limitations. | 8 / 10 |
Source trailInspectable personas and research inputs support exploration. Simulated conclusions are distinct from observations collected from real audiences. | 4 / 10 |
Evidence differs by product. No product was tested head to head. Scoring methodology · Publisher disclaimer
atypica explores how audiences might respond before a team has observed their response. It builds AI personas and runs simulated research. Scout brings public social observations into that process. The useful output is a set of hypotheses to develop; it is different from a measurement of what happened in a real campaign.
Who it fits
atypica is best suited to early research: interpreting messages, exploring objections and framing questions for human research. Its product combines study design, inspectable AI personas, simulation and validation. atypica official overview
Persona-based research produces model-generated explanations. Those outputs are research hypotheses, not statements collected from real participants.
Social observation is one input
Scout is the product’s social-media observation module. Its documentation describes reading public posts, forming audience hypotheses and returning to the source environment to challenge them. The output feeds AI personas. Crucially, atypica itself calls Scout qualitative, says it does not produce statistically significant numbers and advises treating conclusions as hypotheses requiring validation. Scout documentation and limitations.
Scout fits audience-language research and exploratory framing. Its qualitative observations and simulated answers are distinct from a representative survey or a measurement of campaign impact.
What changed in 2026
The public changelog records additions around persistent persona panels and simulated group discussion, followed by improvements to reporting and sharing. These developments support repeated exploratory work and communicating results. They are product updates, not external validation of predictive accuracy. atypica changelog.
The strongest use of synthetic research is to develop and organize hypotheses before collecting direct audience evidence. A simulated response alone does not establish demand, population-wide attitudes or sales causality.
Where unrandom fits
unrandom’s product workflow analyzes campaign content and audience responses, with historical comparisons where coverage permits. unrandom product.
The distinction is timing and evidence: atypica explores potential reactions before a launch; unrandom examines campaign content and audience responses after publication. We place atypica in the adjacent-research category because it serves the former job.