Example
How to write a policy analysis brief without letting the last meeting set the answer
Amrita's problem is not a lack of effort. It is that Deliver a Friday brief that shows options with sources, marks missing costings, and recommends only what the evidence can carry has outgrown scattered reminders, tabs, and memory. This fictional playbook shows a concrete Board, the Cards inside it, one consequential Room, and the line Ava does not cross.
This fictional example was created with AI assistance from a structured editorial brief and reviewed by the You.one Editorial Team for usefulness, distinctness, and product truth.
Amrita is an explicitly fictional portrait, not a customer or testimonial.
The Friday decision brief Board gives one mission a visible field; the Choose how to treat option B Room keeps its evidence, conversation, state, and decision together.
Current Superboard provides the durable structure and explicit, bounded Ava paths described here; broader proactive or external work is not a current promise.
Amrita's pressure is ordinary—and still heavy
Amrita is a fictional 37-year-old analyst in a provincial ministry preparing a decision brief in Moncton, New Brunswick. Amrita has a Friday brief for a deputy. Three options are in a draft. A last-minute stakeholder meeting pushed option B. The costing note for option C is still with finance.
Amrita can write. If option B wins because it was spoken last, the brief becomes minutes with letterhead.
A draft doc, a meeting recap, and a finance email are three authorities.
Option C looks weaker only because its numbers have not come back. That absence is currently being read as a no.
A meeting is input; a blank costing is a hole
The mission is specific: Deliver a Friday brief that shows options with sources, marks missing costings, and recommends only what the evidence can carry
The consequential choice is not something a board or an AI should quietly make: Whether to recommend B with the meeting noted, hold a recommendation until finance replies, or present A and C as equal pending numbers
That distinction matters. Superboard can make the work, evidence, waiting, and decision visible. Amrita still owns the purpose, tradeoff, and final call.
Available today
Friday decision brief: one Board shape to adapt
A useful Board gives this mission one durable operating picture. Amrita does not have to convert life into project-management jargon or decide the perfect taxonomy first. Lists separate kinds of attention; Cards keep each meaningful item visible and movable.
The Cards are deliberately mixed. Some are tasks, some are decisions, some hold a person or promise, and some are reference points. Current Superboard supports that flexibility without flattening the mission into one long to-do list.
| List | What belongs here |
|---|---|
| Mission | Audience, date, and the no-meeting-as-verdict rule |
| Options | A, B, and C with what is known |
| Sources | Evidence attached to claims |
| Holes | Missing costings and unanswered questions |
| Waiting | Finance and stakeholder follow-ups |
Available today
The Cards make the operating picture concrete
A Board becomes useful when the Card titles sound like the actual situation. These are not generic placeholders; they show what Amrita would need to see to run this mission without carrying every open loop in memory.
| Card | List | Job |
|---|---|---|
| A brief, not minutes | Mission | Keep Friday’s audience beside the rule that the last meeting is input |
| Choose how to treat option B | Options | Compare recommend-B, hold-for-finance, and present-A-and-C-equal |
| Option C costing with finance | Holes | Name the missing numbers so absence is not a verdict |
| Stakeholder meeting recap | Sources | File the meeting as a source, not as the recommendation |
| Option A implementation note | Options | Keep an option that nobody spoke last still in view |
| Finance reply on C | Waiting | Hold the unanswered costing request |
Available today
Available today: Choose how to treat option B becomes a Room
Opening the Card gives the visible item durable depth. Stage can hold Recommendation brief: Place meeting input, missing costing, and recommend-B / hold / equal-pending options on one page. Chat keeps the request and response beside that artifact instead of in a detached thread. Pulse can show Waiting for Amrita to choose what Friday is allowed to recommend. Activity preserves the attributable Card record supported by the current product.
A useful Card Chat request would be: “Using only this Board, compare recommending B, holding for finance, and presenting A and C as equal pending numbers. Do not write the deputy note or choose for me.” When live AI is configured, current Ava can respond to an explicit Card mention using supported context and can make limited reversible product changes. She does not deeply reason over the whole Board by default, and she does not gain authority over the decision merely because the context is organized.
Amrita's call remains explicit: Amrita chooses the recommendation posture, writes the brief, and owns what she sends upward
| Surface | Job in this example |
|---|---|
| Stage | Recommendation brief: Place meeting input, missing costing, and recommend-B / hold / equal-pending options on one page |
| Chat | Keep Amrita's request and Ava's attributed response with the work |
| Pulse | Waiting for Amrita to choose what Friday is allowed to recommend |
| Activity | Preserve attributed Card changes and the supported record around the request |
What to ask Ava—and what not to assume
These are useful requests to adapt, not claims that current Ava automatically surveys the whole Board, prepares every comparison, or follows up on her own. The relevant facts must be present in supported Board/Card context, and the human still checks the result.
Amrita owns analysis, recommendations, and what she sends. Ava does not contact finance, write the brief as authority, or make a government decision.
The modest outcome is not that Ava lives Amrita's life. The modest outcome is a brief that shows its work: option C’s missing numbers are a hole, the meeting is a source, and Friday’s recommendation is chosen rather than inherited.
- Request idea: summarize options, sources, and holes already on the Board
- Request idea: compare recommendation postures using the missing costing Amrita captured
- Request idea: help keep a meeting recap in Sources rather than as a verdict
A draft document plus a meeting recap may still be enough
A draft document plus a meeting recap is enough when costings are in and options are stable
It starts to break when a last meeting and a blank finance cell both try to be the answer
Superboard earns a place only when the responsibility needs a durable picture around the list: the reason, artifact, conversation, current state, decision, and history.
A starter recipe to adapt, not obey
Use the names that already make sense in Amrita's world. The recipe is successful when the Board reduces remembering and exposes the real decision—not when every Card is perfectly categorized.
| Step | Action |
|---|---|
| 1. Name audience and date | The brief has a reader |
| 2. Card each option | Knowns and unknowns |
| 3. Attach sources | Meetings are input |
| 4. Surface holes | Missing numbers stay visible |
| 5. Choose recommendation posture | Open a Room before the recap writes the answer |
Direction
Direction, not a current promise
Later, Ava may help this Board notice when a recommendation Card still depends on a Waiting costing, still leaving every brief with Amrita.
The unified You.one and Superboard runtime, cross-product personalized Memory Spine, broad proactive coordination, realtime shared editing, and general external execution are not available today. Future actions would still require the applicable capability, connection, grant, and human authority.
What this realistic example does not claim
- Amrita is fictional and is not a customer, testimonial, research participant, or disguised real person.
- This is not government policy, legal advice, or a claim about any real ministry.
- It does not show Ava contacting officials or issuing a recommendation.
- It is not a measured policy outcome.