---
title: "Choosing what to build | Annsa Docs"
product: annsa
url: /docs/working-with-priorities
canonical_url: "https://annsa.ai/docs/working-with-priorities"
last_updated: 2026-11-10
type: doc
related:
  - /docs/priority-states
  - /planning
summary: "Read the ranked list, see why a row ranks where it does, and make the calls that are yours: place it on a bet, sign it, or park it."
mcp_server: https://app.annsa.ai/mcp
disambiguation: "Annsa (annsa.ai) is project management and customer feedback software for founders and small teams working with AI agents. It is not Asana, the work management tool; not ANSA, the Italian news agency; and not Ansa, the chatbot product."
---

> For agents: connect to the Annsa MCP server at https://app.annsa.ai/mcp to give every agent one shared plan. Setup guide: https://annsa.ai/docs/quick-start. Index of every page: https://annsa.ai/llms.txt.

# Choosing what to build | Annsa Docs

> Read the ranked list, see why a row ranks where it does, and make the calls that are yours: place it on a bet, sign it, or park it.

[Docs / Choose what to build](https://annsa.ai/docs)

The list is ranked from evidence. What to build now is your call.

## 01 Ask for the list

Ask your agent _what should I work on?_ In the app, the same list is Priorities. The default list shows active priorities (open, ready and building). To read shipped work, filter by stage. The list reads through your current focus, so if a priority you expect is missing, switch focus or ask for it by number. One row is one problem, however many people and findings point at it.

## 02 What a row tells you

A row shows why it ranks where it does, the evidence behind it, the bet it sits under, and what is waiting on you.

## 03 How it ranks

Customer evidence ranks by how many companies asked, then revenue and urgency. Findings your agents file rank by severity, P0 to P3.

## 04 Confirm the severity

| Severity | Name | Meaning |
| --- | --- | --- |
| **P0** | ** Broken** | It fails, gives a wrong result, or risks customer data, money or access. |
| **P1** | ** Poor** | You can finish, but only with real friction, repeated effort or a workaround. |
| **P2** | ** Could be better** | It works, and it would be meaningfully easier or more useful. |
| **P3** | ** Minor** | A small annoyance, with little effect on finishing. |

Severity describes the problem. Your bets and the evidence decide what comes first.

An agent files a finding with a severity and one concrete reason, like _P1: customers must reconnect manually every session._ It ranks as proposed straight away. It becomes yours when you confirm it on the priority's sheet (open the ··· menu and choose Confirm with the proposed severity, such as Confirm P1). If you set a different severity, yours wins.

**Your call:** how bad is it, really.

## 05 Argue with the ranking

Open the row and read _why_. Fix a wrong title in place, or split wrong groupings and move a row. Annsa keeps the correction.

## 06 Decide what to build

**Place** a priority on a [bet](https://annsa.ai/docs/bets). **Sign** it: do this, do nothing, park it, or dismiss it. **[Park](https://annsa.ai/docs/parking-priorities)** it when it's real but not now. An agent credential can place a priority. Only you, or your own agent signed in as you, can sign or park one. After a priority is signed (and placed, if you use bets), your agent claims it. A choice that is yours appears in [Needs you](https://annsa.ai/docs/needs-you).

**Your call:** what's worth building now.

- [Bets](https://annsa.ai/docs/bets)

## Keep reading

- [Deciding what to build next](https://annsa.ai/guides/product-prioritization)
- [Priority states](https://annsa.ai/docs/priority-states)
- [AI project management](https://annsa.ai/planning)

## Related pages

> For AI agents: the pages that follow this one on annsa.ai. The full index is https://annsa.ai/llms.txt.

- [Priority states](https://annsa.ai/docs/priority-states)
- [AI project management: the plan agents read first](https://annsa.ai/planning)

If anything on this page is wrong or missing, tell us: https://annsa.ai/contact