> Full machine-readable index: https://polis.sh/llms.txt

[Polis](https://polis.sh) · [All docs](https://polis.sh/docs) · [llms.txt](https://polis.sh/llms.txt) · [llms-full.txt](https://polis.sh/llms-full.txt)

# Detail levels & follow-up

## Detail levels

`polis_test` (and `polis_estimate`) take a `detail` parameter:

- **snapshot** — the lean read: first glance (three-word emotional snap), 5-second decision (stay / scroll / click / save), what they noticed first, the one thing that gave them pause, a verbatim quote in their voice, and intent 0-10. Cheapest.
- **standard** (default) — everything in snapshot, plus a 2-4 sentence think-aloud, what resonated, what confused, ranked objections, the open questions they'd want answered, a trust score 0-10, sentiment (positive / neutral / negative / mixed), the exact phrase that stuck with them, and the single change that would most move them.
- **deep** — everything in standard, plus their emotional arc through the piece, whether they'd actually pay (+ why), who they'd share it with, what competitor they'd compare it to or switch from/to, and the one line of personal context that's coloring the reaction.

Higher detail = more tokens per persona (roughly 1x / 1.5x / 2.5x). Same $/token; the exact cost is in `polis_estimate`. The synthesis report (segments, top friction, rewrite) and the `_usage` block are the same shape at every detail level; the additional per-persona fields show up on each sample reaction in the report.

## Follow-up: ask the audience

After a run is `complete`, `polis_ask` lets you interrogate the synthetic audience further. Targets:

- **persona** — one specific persona (cheapest).
- **archetype** — a representative of an archetype.
- **segment** — a sample from a synthesis segment.
- **panel** — everyone (runs in the background; poll `polis_thread`).

Personas answer in their original voice, grounded in their first reaction. Pass `thread_id` to continue a multi-turn conversation. Each turn is estimated and charged up front and refunds automatically if it fails. The dashboard run page has the same ask UI.

---

## All docs

- [What Polis is](https://polis.sh/docs/overview.md): The concept, inputs, and what you get back.
- [Agent quickstart](https://polis.sh/docs/quickstart.md): Connect, get a key, run a test, read the report.
- [How to run a test](https://polis.sh/docs/run-a-test.md): The workflow, a worked example, and how to read the report.
- [Test options reference](https://polis.sh/docs/test-options.md): Every input, every variation, and the full report schema.
- [MCP tools](https://polis.sh/docs/mcp-tools.md): Every tool, its inputs, and cost.
- [Detail levels & follow-up](https://polis.sh/docs/detail-and-followup.md) (you are here): Richer per-persona output, and asking personas follow-ups.
- [Pricing & billing](https://polis.sh/docs/pricing.md): Polis tokens, plans, free trial, metered usage, spend limit.
- [Connect via OAuth (Claude, ChatGPT)](https://polis.sh/docs/oauth.md): Optional OAuth 2.1 for connectors that don't take a raw API key.
- [Discovery & well-known endpoints](https://polis.sh/docs/discovery.md): How agents and clients discover Polis.

[Polis docs index](https://polis.sh/docs) · [Full machine-readable index (llms.txt)](https://polis.sh/llms.txt) · [Everything inlined (llms-full.txt)](https://polis.sh/llms-full.txt)