Lesson 12 of 25
The Council of Invisible Counselors
Instead of asking for an answer, ask several specified viewpoints to argue the question out and then reconcile them. A single answer smooths the disagreement away; a staged debate forces the objections into the open, which is usually the information you were missing. The counsellors do not have to be real people - "a CFO" and "a shop owner" are enough to produce genuinely different arguments.
The council structure in the original was written for this platform, so keep it and add the two things only Gemini brings: ground the perspectives that depend on verifiable fact in Search, in a clearly separate section, and save a panel you will reuse as a Gem. For the business decision, attaching your actual project and utilisation data turns a hypothetical CFO into one arguing from your numbers.
- Keep the debate ungrounded; put sourced facts in their own section.
- Attach your real utilisation or project data to the business panel.
- Save a reusable panel as a Gem.
- Do not search while generating creative concepts.
- Send the synthesis to a Doc the panel can comment in.
Instead of
Tell me the benefits of my company switching to a four-day work week.
Perplexity version
### THE DECISION ### My tech services company in Hyderabad - 60 people, project work for US and UK clients - is considering a mandatory four-day work week for all employees. Do not tell me the benefits. Stage a debate. ### THE PANEL ### Each voice is accountable for something and should argue like it. **1. Head of People** (accountable for attrition and open roles) - the case FOR: morale, retention, productivity per hour, and the hiring advantage in a market where everyone wants the same engineers. **2. Chief Financial Officer** (accountable for margin) - the case AGAINST: utilisation and billing, the cost of coverage, a fixed cost base against 20% fewer available hours, and the compression risk where five days of work becomes four days of exhaustion. **3. Delivery Head** (accountable for client satisfaction) - neither for nor against; states what would break. Client expectation of weekday response, incident cover, handover between people off on different days. **4. A senior engineer with two small children** - what it means day to day. If flexible hours would help more than a fixed fourth day off, say so. Each ends with "The best argument against my own position". ### USE MY DATA, NOT INVENTED NUMBERS ### I am attaching a Sheet with the last four quarters of utilisation by person, our project list with delivery dates, and the client time zones. The CFO and the Delivery Head must argue from these figures, citing the cells they are using. Do not invent a single number, and do not use Google Search to find industry averages for this part - my numbers are the ones that decide it. ### THEN, SEPARATELY - GROUNDED FACTS ### Under a heading "Researched, with sources" and outside the debate: use Google Search for what is actually documented about four-day-week trials - which organisations, what was measured, what the published results were, and specifically whether any Indian IT services firm has tried it. Cite sources with dates and separate independent evaluation from company press releases. Keep this out of the panellists' mouths entirely. ### SYNTHESIS ### Where they genuinely conflict; which objection is fatal if unaddressed; a pilot with specific stop numbers drawn from my attached data; and which panellist I would be wisest to overrule. ### OUTPUT ### Put the synthesis and the pilot design into a Google Doc called "Four-day week - panel review" so the actual leadership team can comment in the margins. Save the panel structure as a Gem called "Decision panel" - I want to run other decisions through the same four voices.
Instead of
What was the impact of India's economic liberalization in 1991?
Perplexity version
### WHAT I WANT ### The long-term impact of India's 1991 economic reforms, from four incompatible vantage points rather than as a verdict. ### THE PERSPECTIVES ### 1. **A major industrialist** who ran a licensed manufacturing business before 1991. The end of the License Raj, imported inputs, foreign capital, global competition - and the loss of a protected position. 2. **A small-scale shop owner in a tier-2 city.** Competing with large and foreign brands, changing supply chains, organised retail, credit. Where it went well; where it hollowed out a local economy. 3. **A social economist.** Inequality, urbanisation, the informal sector, divergence between states, public provision - separating what the reforms caused from what was happening anyway. 4. **A public sector bank employee.** The voice usually missing: inside an institution that went from policy instrument to competitor. These are composites. Do not invent quotes or attribute views to real named people. ### THE NUMBERS - THIS IS THE IMPORTANT PART ### Do not state growth rates, poverty figures or inequality measures from memory. Instead, use Google Search and put every figure in a separate section headed "Data, with sources", citing RBI, MoSPI, World Bank or NITI Aayog, each with its date and the series it comes from. Where two sources give different numbers for the same thing - and for 1990s poverty estimates they do - show both and name the methodological argument between them. The perspectives above should refer to this data section rather than carrying numbers inside the role-play, so I can always tell which parts are sourced and which are argument. ### THEN ### * **Where these accounts conflict** - the real disagreements, not a summary. Show how the industrialist and the economist can both be accurate and reach opposite verdicts. * **Where scholars dispute it** - whether growth accelerated in the 1980s rather than after 1991; how much poverty reduction is attributable to reform; whether inequality was a necessary cost. Cite the economists on each side, with real references. * **Who is missing** - which Indian experiences of the 1990s none of these four represents. ### OUTPUT ### A Google Doc called "1991 reforms - four views", with the data section at the end and a short reading list of authors and works you can actually source.
Instead of
Give me an idea for a superhero.
Perplexity version
### THE TASK ### A concept for a new Indian superhero, developed through a writers' room argument rather than a brainstorm. ### THE ROOM ### **Writer A - The Traditionalist.** Powers rooted in Indian mythology: the Astras, boons earned through tapas, power granted conditionally and withdrawable. Argues this gives the hero a moral architecture imported superhero logic lacks. **Writer B - The Futurist.** Powers from technology. Argues the mythological route is a nostalgia trap, and that a hero whose power is infrastructure - satellites, payment rails, the grid - is more original than another divine weapon. **Writer C - The Script Editor.** Asks only what the hero does in chapter one and who they fight. Interrupts whenever either writer is building a mythology instead of a story. They must genuinely disagree - A attacking B as rootless, B attacking A as a costume over a Marvel template - and at least one must change their mind about something specific. ### DO NOT SEARCH WHILE GENERATING ### Turn off the research instinct for the debate and the concept. If you ground this, the writers' room becomes a summary of comics that already exist, which is the one thing I do not want. ### THE SYNTHESIS ### One concept built from the argument - resolving the tension between granted power and built power, not splitting the difference. Give me: the hero, the source and the cost of the power, the antagonist, a specific named setting, and the opening scene. ### THEN, AFTER THE CONCEPT IS DONE ### Now search, under a heading "Prior art check", and tell me whether anything close to this already exists in Indian comics, film or streaming. Cite what you find with dates. Do not invent titles - if you cannot verify something exists, say so. ### INPUTS AND OUTPUT ### I am attaching reference images for the visual direction I have in mind; use them when describing the hero's look, and tell me where my references and the concept pull against each other. Put the final concept into a Google Doc called "Hero concept v1" with the prior art check as a separate section. Save the three-writer room as a Gem called "Writers' room" for future concepts.