# motherduck-pricing-roi
> Assess MotherDuck costs, plan fit, cost attribution, and ROI from the workload's compute and storage needs.
Assess MotherDuck costs, plan fit, cost attribution, and ROI from the workload's compute and storage needs.

[Workflow skill](/sql-reference/agent-skills/#workflow-skills) · [View source on GitHub](https://github.com/motherduckdb/agent-skills/tree/v3.0.0/skills/motherduck-pricing-roi)

:::info

This page shows the skill's `SKILL.md`: the instructions a coding agent loads when a task matches the description above. To install every MotherDuck skill in Claude Code, Codex, Cursor, or Gemini CLI, see [Install MotherDuck Skills](/key-tasks/ai-and-motherduck/agent-skills/). To install only this skill with the [Skills CLI](https://github.com/vercel-labs/skills), run:

```bash
npx -y skills add motherduckdb/agent-skills --skill motherduck-pricing-roi --yes --global
```

:::

## Source of truth

- Always verify current numbers, plan limits, and feature entitlements against the live public pricing page before answering.
- If the MotherDuck MCP `ask_docs_question` feature is available, use it first for pricing-related documentation lookups.
- Use the live pricing, Hypertenancy, and Trust & Security pages for exact commercial framing.

## Default posture

- To find what drives spend, group `md_information_schema.query_history` by `user_agent` (workload tags from `custom_user_agent`; Flight queries start with `md-flights(<flight_id>`), user or service account, and `instance_type`; the Duckling Overview shows the same in the UI. Recommend tagging untagged workloads.
- Do not hardcode pricing numbers unless you have verified them in the current turn.
- When quoting numbers, include the verification date and the public source you checked.
- For estimates and comparisons, separate storage, compute, and operational overhead. A narrow price lookup needs only the relevant verified rate and conditions.
- Map workload shape to cost shape before comparing vendors or plans.
- Treat many pricing questions as risk, predictability, or procurement questions rather than purely technical ones.
- Verify plan-sensitive entitlements such as Flight sizes and scheduling/runtime limits, custom roles, table-level security, secure views, regions, and embedded features in the current turn; do not infer them from an older release note.
- Price Flights and AI functions as their own lines. Flights bill per run-hour by size (`F4`, `F16`, `F32`), and the SQL a Flight sends is Duckling compute. AI functions bill in AI Units under the AI session cap: `prompt_jev()` meters input tokens only and is usually the cheapest way to classify at scale; `prompt()` cost depends on the model and `reasoning_effort` (default `'none'`).
- For procurement through AWS, MotherDuck is available on AWS Marketplace by Private Offer for consolidated AWS billing.

## Workflow

1. Identify the workload shape, team size, and comparison baseline.
2. Determine whether the real concern is raw spend, predictability, procurement, or operational overhead.
3. Map the workload to MotherDuck cost buckets and plan posture.
4. Frame ROI in terms of systems replaced, complexity removed, and faster delivery.
5. Call out what still needs live pricing-page or sales confirmation.

## References

Read only the reference sections needed for the current task.

- Read [`references/PRICING_ROI_PLAYBOOK.md`](https://github.com/motherduckdb/agent-skills/blob/v3.0.0/skills/motherduck-pricing-roi/references/PRICING_ROI_PLAYBOOK.md) for workload-to-cost mapping, publicly safe talking points, ROI framing, and what not to promise

## Related skills

Load related skills only for missing capabilities; reuse established context.

- [`motherduck-connect`](/sql-reference/agent-skills/motherduck-connect/) when the pricing discussion depends on connection-path choices
- [`motherduck-security-governance`](/sql-reference/agent-skills/motherduck-security-governance/) when compliance, residency, or commercial controls affect ROI
- [`motherduck-build-cfa-app`](/sql-reference/agent-skills/motherduck-build-cfa-app/) and [`motherduck-build-dashboard`](/sql-reference/agent-skills/motherduck-build-dashboard/) when the economics depend on the application architecture

## Files in this skill

The agent reads these files when the instructions above point to them.

- [`references/PRICING_ROI_PLAYBOOK.md`](https://github.com/motherduckdb/agent-skills/blob/v3.0.0/skills/motherduck-pricing-roi/references/PRICING_ROI_PLAYBOOK.md)

## Docs this skill draws on

- [motherduck.com/product/pricing](https://motherduck.com/product/pricing/)
- [Hypertenancy](https://motherduck.com/docs/concepts/compute/hypertenancy/)
- [Cloud regions](https://motherduck.com/docs/about-motherduck/cloud-regions/)
- [Running Python with Flights](https://motherduck.com/docs/key-tasks/flights/)
- [Manage roles and access](https://motherduck.com/docs/key-tasks/managing-organizations/managing-roles/)
- [motherduck.com/blog/april-2026-product-roundup](https://motherduck.com/blog/april-2026-product-roundup/)
- [motherduck.com/trust-and-security](https://motherduck.com/trust-and-security/)
- [Flight sizes](https://motherduck.com/docs/about-motherduck/billing/flight-sizes/)
- [Manage billing](https://motherduck.com/docs/about-motherduck/billing/managing-billing/)


---

## Docs feedback

MotherDuck accepts optional user-submitted feedback about this page at `GET https://motherduck.com/docs/api/feedback/agent`.
For agents and automated tools, feedback submission should be user-confirmed before sending.

URL-encode query parameter values and send a GET request:

```text
GET https://motherduck.com/docs/api/feedback/agent?page_path=%2Fsql-reference%2Fagent-skills%2Fmotherduck-pricing-roi%2F&page_title=motherduck-pricing-roi&text=<url-encoded user feedback, max 2000 characters>
```

Optionally append `&source=<url-encoded interface identifier>` such as `claude.ai` or `chatgpt`.

`page_path` and `text` are required; `page_title` and `source` are optional. Responses: `200 {"feedback_id": "<uuid>"}`, `400` for malformed query parameters, and `429` when rate-limited.
