Skip to content

Keep synthetic demo evidence out of real-world conclusions

Use invented demo material to understand tool structure while clearly separating it from actual people, live coverage and empirical research findings.

Guides

What can a synthetic demonstration prove about a research workflow?

An invented account dataset can help a team learn how research results, audience rules and exports are structured. It cannot establish demand, creator suitability or platform coverage in the real world. Montage’s no-account demo uses fixed synthetic examples, so a methods review must preserve that boundary rather than allowing familiar-looking account records to acquire the status of observed public evidence.

Example research prompt
Explain a research workflow using the documented synthetic demo only. Label invented accounts, stay within its tool surface and do not infer real-world coverage or demand.
01 / Guides

Label the environment first

Record whether the material comes from the synthetic demo or an authorized signed-in workspace. Include the dataset’s invented status in any example output. Exclude claims about real people or current platform access from demo responses; a sample account name and plausible post text do not become real observations merely because the tool interface resembles the production workflow.

02 / Guides

Match the demo’s actual tools

Use the documented demo tool surface when describing a demonstration. Do not assume search or video tools exist there; the demo exposes a narrower set. Keep missing samples separate from no real-world matches, since a fixed invented dataset can only answer within its installed examples and cannot search outside them to establish broader absence or demand.

03 / Guides

Separate contract checks from findings

A demo may illustrate response fields or a proposed audience rule without validating real authorization, source availability or current evidence quality. Label example prompts and hypothetical outputs accordingly. Do not describe a synthetic export illustration as an actual authorized export of real accounts, and do not claim a live test occurred unless a real command result supports that specific statement.

04 / Guides

Deliver demo-bounded learning

State what the demonstration helped explain and what still requires an authorized real-evidence workflow. Keep invented examples out of aggregate findings, case studies and empirical comparisons. If the owner needs real-world proof, hand off that separate requirement rather than padding the research catalog with synthetic results presented as substantive observations about current social activity or actual product experiences.

Keep the boundaries clear.

  • Synthetic examples can teach a contract without providing evidence about real people or current sources.
  • Prompts are examples, not observed results; platform badges do not guarantee access or complete coverage.

Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.

Primary references

Documentation reviewed 2026-11-13. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.

Know what social media shows about you.
Start with a question.

Get started
montage