Skip to content

Welcome

Questo is a workspace for go-to-market teams where an AI agent does the work alongside you.

You give it context about your product and your market. You frame the work as missions and tasks. The agent researches, writes and produces deliverables. You review what comes back, approve what needs your judgement, and the things you correct become context the agent keeps.

The important part is that last sentence. Questo is not a chat window that forgets you every morning. What you curate compounds, so the tenth task is better briefed than the first.

Before the loop: map your go-to-market

Everything Questo knows about your go-to-market is the Questo System - an atlas of your business, held as structured objects rather than scattered across documents and somebody's memory. It covers what you know, what you are driving at, the work, the deliverables, and the signal coming back. You start it here, with the part you already know.

Fill in your knowledge base: what you sell, who you sell it to, how you talk about it, plus any files, links, skills and rules the agent should have. This is the highest-leverage thing you will do in Questo, and it is a one-time cost that pays off on every run afterwards. See Your knowledge base.

This is not a step in the loop below. It is the ground the loop runs on.

The loop

Questo runs your go-to-market as six stages, and every other page in this handbook zooms into one of them. Each stage has a job that stays yours.

  1. Focus - which outcomes you are driving. A mission is a steered effort that runs over weeks, with milestones to check it against and, when it needs one, a rough plan of what each week is for. Your job: manage missions. Start one, kill one that is not earning its place, turn up the intensity. See Missions.

  2. Plan - what today is for. Questo proposes one day plan per active mission: the tasks worth doing today, each with the reason it is there, drawn from the mission's milestones, cadence and roadmap rather than re-invented each morning. Your job: steer planning. Cut what is not worth it, reshape the plan by talking to the agent, or approve in a click.

  3. Execute - the work gets done. Work with the agent live in chat, where you can interrupt and redirect, or let it run in the background while you do something else. Both use the same agent and the same context. Your job: run ad-hoc work. Go in with your own prompt whenever you want something now. See Tasks and the agent.

  4. Distribute - it reaches the world. The agent creates one output per shareable deliverable. Outputs are editable documents that save as you type, so a draft you disagree with is one you fix in place rather than send back. Your job: judge outputs. Nothing ships until you say so. See Inbox and outputs.

  5. Monitor - what actually happened. Questo reads your connected tools and captures what bears on an active mission, then diagnoses each mission with a health verdict. Your job: settle questions. It asks what it cannot observe - how the call went, why the deal stalled - and on the outside world you are the only sensor it has.

  6. Iterate - the next turn is smarter. Tomorrow's plan is cut against today's reading, and what was learned is proposed back as a change to your knowledge base. Your job: approve actions. A plan to approve, a reply to send, a tool to connect - while knowledge edits apply directly and can be undone from History. Everything needing your judgement lands in one queue, so nothing consequential happens behind your back. See Routines.

Once a mission is running, Questo turns this loop daily on its own: it captures signal from your connected tools in the afternoon, diagnoses each mission the next morning, and proposes a plan for the day. You approve what is worth doing.

If you are the sort of person who wants the map before the tour, Core concepts defines every piece and how they connect.

Your first thirty minutes

  • Fill in the product, positioning and targeting tenets. Onboarding starts these three for you. Read them and fix anything that is not how you would say it.
  • Add two or three files you would hand a new hire on day one, and mark them as knowledge.
  • Add the URLs you check weekly: your site, your pricing page, two competitors.
  • Create one mission with a real target date, not a test one, and write what the first week is for.
  • Give the agent one small task inside that mission and watch it work in chat.
  • Open the inbox and resolve whatever it proposed.
  • Correct one thing the agent got wrong, then write that correction into a rule or a tenet.

If the first task comes back generic, that is a context problem rather than a model problem. Go back to step one.