Research repository or causal test: choosing the right tool before a launch decision
On this page
A product or marketing lead can use existing research to define a launch question, then test the missing contrast. A repository and a choice study serve connected tasks: preserving what is known and generating new evidence about specified alternatives.
The decision: organize what you have, or test what you don't
If your team is sitting on a backlog of interviews, transcripts, support tickets, and sales-call notes, the job is retrieval: find the pattern, tag the theme, get it in front of product and design. Dovetail does exactly that: a searchable repository where real customer conversations get tagged, clustered, and linked back to the moment they came from.
Existing research can identify objections, attributes, segments, and hypotheses for a new price or message. A repository does not itself randomize that new contrast. Subconscious can add a controlled choice study with generated responses, whose effects apply within its design until relevant human transfer is checked.
What happens if you pick the wrong tool?
Two errors can leave a launch unsupported: ignoring useful prior customer evidence, or treating plausible generated responses as proof of market behavior. Retrieve the relevant research, state the missing contrast, and select a design and response source suited to that question.
What each tool actually does
| Repository (Dovetail) | Causal testing (Subconscious) | |
|---|---|---|
| Source of evidence | Real customer interviews, transcripts, tickets already collected | Controlled experiments on simulated markets |
| When it has value | During question definition, study design, and preservation of results | When a specified choice contrast needs new evidence |
| Core question answered | What evidence do we already have, and where did it come from? | How do randomized attributes affect generated choices in the study? |
| Output shape | Searchable, evidence-linked library | Estimated choice contrasts, with uncertainty where supported |
A repository can preserve both human research and new study results. A choice study adds a specified comparison; its evidence should stay linked to the assumptions and source material that shaped it.
Where the proof stands and where it doesn't
The public causal-fidelity paper compares estimated choice parameters with corresponding human studies. That is a specific validation endpoint, rather than a universal accuracy claim. A recruited human study can check the same question and alternatives, with population and instrument differences documented.
Stating the exact scope of an output keeps the claim checkable run by run. Confidence intervals, segment-level breakdowns, and decision memos are study-specific outputs, not universal guarantees on every run. The audience Subconscious can model at the population level is also distinct from a recruited real-human validation study, or from the volume of interviews any one repository holds, worth keeping separate when comparing the two tools.
How do you sequence using both tools?
For example, retrieve interviews describing objections to setup effort, retain their source clips and buyer context, and draft two claims addressing that objection. Specify the eligible audience and choice response before comparing the claims. Store the design, response source, run configuration, estimates, limitations, and any matched human follow-up back in the repository. Review study structure, applied examples, or scope the contrast.
