Mavera DocsOperator's Guide
Checking your Mavera sessionLoading workspace audiences and API keys…
Checking…Use or share this pageSend the page itself, print it, or open its complete context in your AI workspace.
Advanced: run Mavera like an operating system
Time-sliced audiences, branch tests, trackers, autopilot, persistent personas, and API-powered research operations.
Interactive product walkthrough
Build an advanced standing system
Create a bounded program with cadence, spend cap, and run inspection.
Simulation onlyNothing is saved or charged
1
Step 1 of 7Open Programs
Choose Programs from the sidebar.
Mavera workspace
Good afternoon
Choose a surface from the sidebar to begin.
Demo workspace0 actions completed
Recent workspace activityStatus
Audience comparisonReady to review
Customer interview evidenceGrounded
Campaign pre-flightDraft
- 1Open ProgramsChoose Programs from the sidebar.
- 2Start a programCreate a long-running research program.
- 3Set the objectiveWrite the standing question the program should keep answering.
- 4Choose cadenceSet when the work should run.
- 5Set a hard capChoose a bounded credit limit for each run.
- 6Launch locallyStart the simulated program.
- 7Inspect the first runOpen run history, evidence, and the next scheduled pass.
Advanced use is not more prompts. It is a repeatable research system with controls, provenance, and monitoring.
The advanced toolbox
| Capability | What it does | Best use |
|---|---|---|
| Audience compare | Compares two audiences on demographic, theme, and response overlap | Segment strategy and positioning conflicts |
| Persistent personas | Carries a persona through multiple conversations while retaining relevant prior context | Longitudinal interviews and diaries |
| Insight branch / flip test | Re-runs a conclusion under changed assumptions or an adversarial frame | Stress-testing an important recommendation |
| Trackers | Watches an indicator across runs and alerts on movement | Ongoing message, sentiment, or competitor monitoring |
| Autopilot / nightly | Uses a defined budget to research while you are away | Recurring research briefs, not one-off questions |
| Programs | Plans, executes, reads findings, and re-plans a larger job | GTM, category, launch, or competitor work |
| Dossiers + exports | Turns study material into a report and downloadable deliverables | Client handoff and evidence trails |
Prompt: historical segment comparison
Prompt3 fields left
Research Lens: Point in time + Current sources. Create two versions of [audience]: one in [historical period] and one today. Compare their decision criteria, language, media environment, and price sensitivity. Do not force a percentage where evidence cannot support it. Produce a table of what changed, what did not, and the implications for [decision].
Unfilled variables remain in brackets.
Prompt: adversarial conclusion test
Prompt1 field left
Take our current conclusion — "[conclusion]" — and run a flip test. Identify the minimum plausible assumption changes that would reverse it. Which assumption is most fragile? What real-world evidence should we collect before acting?
Unfilled variables remain in brackets.
Prompt: a standing research system
Prompt2 fields left
Design a monthly research operating system for [business]. Include: a fresh-source scan, a customer-voice ingest step, one tracker, one message test, a spend cap, an approval point before external claims, and a client-ready monthly summary. Save it as a Skill called "[name]".
Unfilled variables remain in brackets.
Autonomy is not permission to invent
Set a budget, a source rule, and an approval point. Autopilot should surface work for review, not publish claims or change customer-facing materials without you.