Routines¶
Most of Questo runs when you ask it to. Routines are the parts that run without you.
A routine is recurring work Questo does for your workspace. Routines run in the background on a fixed schedule, and anything that needs your judgement lands in the inbox rather than being decided for you.
What you control¶
Routines live in Settings > Routines, grouped as Daily, Weekly and Custom (a routine that runs on several chosen weekdays). Each row has an on/off switch, a Documentation link to its section below, and a menu with Run now and View runs.
You cannot edit a routine's name, instructions or schedule. Routines only work if they behave predictably, so Questo keeps them fixed and improves them for everyone. If you want recurring work shaped your own way, run it as a mission: the routines then work that mission for you, day to day.
Run now is refused while a routine is switched off, while it is already running, or while the workspace is out of credits.
Routines run on your workspace's time zone, set in Settings > Organization. Changing the time zone moves every routine with it.
The routines at a glance¶
| Routine | When | What you get |
|---|---|---|
| Mission planning | Weekday mornings | A diagnosis and a proposed day plan for each active mission |
| Insight extraction: connected tools | Weekday afternoons | Insights from each tool Questo reads |
| Insight extraction: Questo | Weekday afternoons | Insights from Questo's own finished work |
| Insight request | Weekday afternoons, after extraction | A few questions only you can answer |
| Knowledge update | Monday, Wednesday and Friday evenings | Your tenets and records kept current |
| Weekly report | Sunday | A recap of the week, per mission |
| Roadmap review | Sunday, after the weekly report | Each mission's coming weeks kept true |
A workspace gets one insight extraction routine per connected tool that Questo reads, so a workspace with Slack Insights and Gmail connected has two.
Everything the routines produce adds up to how a mission iterates, because that is what they are for.
They run in order, not just on a clock¶
The chain matters more than the times. The insight request waits for insight extraction to finish, because its job is asking about what extraction could not find. The roadmap review waits for the weekly report, so it reads a closed week.
The times above are when a routine becomes eligible, not a promise about the minute. If an earlier routine fails, the next one still runs with a gap in what it knows, and the failure shows in View runs rather than silently freezing tomorrow's plan.
Turning them off¶
Turning a routine off is a normal thing to do.
If you do not run missions week to week, the weekly report is noise. If you find the daily questions intrusive, turn off the insight request and the rest carries on without it.
One is worth keeping if you keep any: mission planning, because its daily diagnosis is the only thing tracking whether your missions are working.
Before you turn everything off, try the cheaper fix. Most inbox noise comes from too many active missions, not too many routines. Pausing the missions you are not working this week quiets everything at once.
If nobody in your workspace has opened Questo for a few days, Token saving mode in Settings > Capabilities can pause scheduled work until someone comes back. See Settings.
What background runs cannot do¶
A routine runs with nobody at the screen, so it cannot approve its own proposals, resolve inbox cards or approve a plan. It does its work, and anything that needs a decision waits for you. The one exception is the day plan, which approves itself after six hours unless you press Hold. The full policy is in what the agent does without asking.
Mission planning¶
Why it exists. A mission runs for weeks. Without something looking at it every day, you are the one who has to notice that a milestone slipped, that three prospects raised the same objection, or that nothing has moved since Tuesday. Mission planning does that looking so you can spend your time deciding instead of diagnosing.
What you get. For each active mission, two things in one card:
- A mission update: a short diagnosis of what changed, a one-line status for each pending milestone, and a health verdict of On track, At risk, Off track or No trend. It reads the trend across recent days, not just today.
- A day plan: the handful of tasks worth doing today, each with the reason it is there and the insights behind it. It proposes only work the agent can do itself, such as research, drafting and analysis. Calls, meetings, sending and configuring other tools stay with you. The mission's intensity decides how wide and bold the plan is.
Along the way it may adjust the mission. It marks a milestone completed only when what it read shows it landed. It updates the playbook only when direction has durably changed, which most days it has not. It moves or cancels a scheduled task that today's signal contradicts, with a reason on the task you can undo from History. If a mission is past its target date, it asks on the card whether to extend it, close it or drop it, and never moves the date itself.
A mission with nothing worth doing that day gets no card. That is a result, not a failure.
When. Weekday mornings. It reads everything captured up to that morning, so Friday afternoon's insights reach Monday's plan. It also runs when you create a mission as active, and when you choose Update and plan mission from a mission's menu. It plans only active missions and is skipped when there are none.
What you can change. You steer it through the mission rather than the routine:
- Status decides which missions get planned at all. Pause the ones you are not working this week.
- Intensity sets how hard each mission is pushed.
- The playbook is what the plan starts from. A sharper playbook gives a sharper plan.
- Your answers to the insight request are facts the next diagnosis can use.
Insight extraction: connected tools¶
Why it exists. Most of what matters to a mission happens somewhere you were not looking: an objection in a sales call, a complaint in a support ticket, a deal that stalled in the CRM, a decision in a Slack thread you were not in. The daily plan is only as good as what it knows, and nobody has time to copy that signal into Questo by hand.
What you get. One extraction routine for each connected tool Questo reads, created and switched on when you connect the tool and removed when you disconnect it. Connections that only publish or only open pages you picked get none. See Integrations.
Each run reads its own tool since the last run and saves what bears on your active missions as insights. Every insight carries a quote or link from the source as evidence, the date it happened, and the missions it bears on. When a signal concerns a competitor or persona, the insight is linked to that record and marked as confirming or contradicting what the record says.
It writes insights and nothing else. It never edits missions, tasks, tenets or records, and it never asks you anything. Nothing arrives in your inbox. You meet its insights on the Insights page and in the next morning's plan.
When. Weekday afternoons.
What you can change.
- Connect the right tools. It only sees what the connection can read.
- Name your sources in each mission's strategy. The channels, labels or pipelines you list under Insight sources are read first, which is the single biggest improvement to what it finds.
- Switch off one tool's extraction if its signal is mostly noise, without disconnecting the tool.
Insight extraction: Questo¶
Why it exists. A research task finishes, an output is written, and the finding is trapped inside a document. Next week's plan has no idea it exists unless someone re-enters the point. This routine makes what the agent already learned reusable.
What you get. It reads the tasks finished since its last run and their outputs and files, and saves the facts that work established as insights, each linked to the output or file it came from. It keeps only facts the work traces to a source outside Questo, such as a page that was read, a number from a file you uploaded, or something a customer said. The agent's own conclusions never come back as evidence.
Nothing lands in your inbox, and it never edits your knowledge base. The knowledge update weighs these insights on its next run.
When. Weekday afternoons, alongside the other extraction routines.
What you can change. Nothing needs configuring and no integration is needed. What it captures follows the quality of the work: an output that shows where its facts came from produces insights, one that only offers opinions does not.
Insight request¶
Why it exists. Extraction sees what was written down. It cannot see that a deal signed over the phone, that you decided to drop a channel, or that a blocker cleared on a call. Rather than plan on a picture that is quietly wrong, Questo asks.
What you get. One inbox card covering your active missions, with up to three questions per mission, each with a sentence on why it is asking. It asks only what no connected tool can tell it, such as whether a milestone landed or a decision only you hold, and only when the answer would change what gets planned in the next few days. It does not re-ask what is already answered.
Answer the ones you can in a line and press Submit. Each answer becomes an insight on its mission, which is where the next plan looks. Ignore the rest; the card closes after a day. If nothing is worth asking, no card is posted. With Slack Inbox connected you can answer from Slack.
When. Weekday afternoons, once insight extraction has finished.
What you can change.
- Switch it off if the questions feel intrusive. The rest carries on, with less to go on.
- Fewer active missions means fewer questions, because the limit is per mission.
Knowledge update¶
Why it exists. Every run reads your knowledge base, so a stale tenet misleads all of them at once. The facts that would correct it arrive scattered across sales calls, competitor sites and forum threads. This routine is where they are weighed and written back, so the rest of Questo starts the day on current knowledge.
What you get. Each run does two things:
- It searches the web for what changed about your competitors, your market, your buyers and your active missions, and saves what it finds as insights with the source Web.
- It weighs every insight captured since its last run, from the web and from the extraction routines, and writes the well-supported facts into the tenet or competitor or persona record they belong to.
The bar depends on what is being changed. Facts about the world, such as a competitor's pricing, change when a reliable source shows them. Your own decisions, such as positioning, messaging or packaging, change only on evidence the decision is live, like your site or product showing it. A chat message arguing a decision should change is left for mission planning to act on. Edits are surgical: it changes the passage that is wrong rather than rewriting the document.
It is the only routine that edits tenets and records. The extraction routines never do. Edits apply directly and nothing lands in your inbox. Each tenet's and record's History shows what changed with Undo on every edit, and the Knowledge page's Updated clock shows a dot when something changed since you last looked.
Most runs, little has moved. No edits is a good result on a quiet day.
When. Monday, Wednesday and Friday evenings, after that afternoon's extraction and before the next morning's plan. Each run covers everything since the last one.
What you can change. You cannot edit the routine, but what it watches follows what you keep current: your competitor and persona records, your Market tenet, and each mission's Insight sources. If an edit is wrong, undo it from History.
Weekly report¶
Why it exists. Every other routine is about the next move, which makes it hard to see at the end of a week what the work added up to. The report puts it in one place.
What you get. One document recapping the past week per mission, in the past tense: what closed, the key insights and the health verdicts. Missions with nothing to report are left out. It proposes nothing.
It arrives as two things: a standalone output you can read, edit or share like any other, and a Report card in your inbox that points at it. The card closes after a day; the output stays. The X on the card discards the card, never the output.
When. Sunday, so it is waiting at the start of the week.
What you can change. Switch it off if you do not run missions week to week. Its length follows how much happened, so a quiet week produces a short report.
Roadmap review¶
Why it exists. The daily plan draws on each mission's roadmap. But roadmaps rot: a launch slips, a bet dies, and the weeks you planned no longer describe what you are doing. Something has to keep the next few weeks correct, so the daily plan can stay focused on today.
What you get. For each active mission, it looks at this week and the next few and writes a week only where that week would change the daily plan, such as a launch week, a holiday or a slip to absorb. It rewrites a week you planned only when something specific has made it wrong. A mission that runs on its cadence may never get a row, which is the correct outcome.
It touches nothing but the roadmap and posts no card. Open the mission's Playbook and click Updated to see what changed, alongside every other edit to the mission, and undo any of it.
When. Sunday, right after the weekly report, so it reads a closed week and finishes before Monday's plan.
What you can change.
- Keep status honest. A paused mission is not reviewed.
- Write weeks yourself whenever you know. The review leaves what you wrote alone unless reality has made it wrong.
- Set a target date on a campaign, so the roadmap stops at the end.
- Switch it off if you manage the roadmap by hand. The daily plan still repairs a week that broke.