A quick note

We're becoming PromptSharp Daily

The AI Rundown is joining the PromptSharp family and becoming PromptSharp Daily. Same free daily you already read — one sharp prompt from a rotating vertical plus the operator's honest AI read — now under one platform where every vertical lives together. Nothing you do changes: same inbox, same time, still free.

▸ USE AI TODAY · a workflow to steal

Rewrite anything for the exact person who has to read it

Paste your content and say: 'Rewrite this for [my CEO / a new client / a skeptical engineer]. Match what they care about and cut what they don't.'

Why it works: Same facts, different reader = different message. Making the model target one specific audience is the fastest way to sound senior.

⌘ TODAY'S PROMPT · copy & paste  ·  Communication

Explain something complex to a specific audience

You understand it; now you have to make someone else get it.

Explain [topic] to [audience, e.g., a non-technical executive] so they truly get it and care. Use a plain-language analogy, lead with why it matters to them, keep it under 200 words, and end with the one thing they need to remember. No jargon unless you define it.

Why: Being the person who can make hard things clear is a career superpower — this manufactures it on demand.

✦ GET BETTER AT AI

Start a fresh chat for a genuinely new task. Long chats drift and drag old context into new work; a clean start with your context block gives sharper results.

◦ Worth knowing in AI · 5 moves in full

◆ YOUR ROLE TODAY · pick your seat

One AI move + one copy-paste prompt built for your role. Read the one that's yours — 30 seconds each.

▣ FOR THE CMO  ·  marketing & growth

Your sales team hears the real reasons deals are won and lost, but that gold rarely reaches marketing. AI can turn scattered win/loss notes into concrete messaging fixes — the objections to pre-empt and the strengths to lead with.

⌘ Turn win/loss feedback into messaging fixes · copy & paste

You are a product-marketing analyst. From the win/loss notes below, extract: the top reasons we win (lead with these), the top reasons we lose (pre-empt these), the competitor most often mentioned and why buyers pick them, and 3 specific messaging changes to make on the site or in outreach. Quote the buyer's own words where possible; separate patterns from one-off anecdotes. [paste win/loss notes / deal feedback]

Why: The reasons you lose deals are your marketing to-do list, written by your buyers.

▤ FOR THE CFO  ·  finance & operations economics

AI is now a fast first-pass analyst for any spreadsheet you can describe. It won't replace your judgment, but it will build the model skeleton and stress-test your assumptions in minutes — freeing you to argue about the assumptions that matter.

⌘ Stress-test a plan's assumptions · copy & paste

You are a skeptical CFO. From the plan below, list every assumption it depends on, rank them by how much the outcome swings if they're wrong, and for the top 3 give a conservative / base / optimistic value plus the one question I should ask the owner to defend it. [paste plan / forecast summary]

Why: Moves the debate to the 3 assumptions that actually decide the number.

▧ FOR THE CSO  ·  strategy & planning

Every strategy doc hides unstated assumptions. AI is unusually good at making them explicit — which is exactly where strategies quietly fail.

⌘ Surface the hidden assumptions · copy & paste

Read the plan below and list every assumption it silently depends on — about the market, customers, our capabilities, and timing. For each, mark it as Safe / Shaky / Unknown and tell me which shaky one to go validate first before we commit. [paste plan]

Why: Strategies die on the assumption no one wrote down. This writes them down.

▨ FOR THE CTO  ·  product & technology

Technical documentation — the thing that never gets written — is now a byproduct, not a project. AI drafts it from the code or the change so the knowledge doesn't live only in one person's head.

⌘ Generate the docs you never have time to write · copy & paste

From the code or change below, draft: a short overview of what it does and why, how to run/use it, the key decisions and their rationale, and the top 3 gotchas for the next person. Write it for a competent engineer who has never seen this. [paste code / change]

Why: The bus-factor risk goes down every time this runs.

▩ FOR THE COO  ·  operations & execution

Staffing to the average leaves you underwater at the peak. AI can turn a workload pattern into a capacity plan — where you're short, where you're carrying slack, and the cheapest way to flex.

⌘ Build a capacity plan from workload · copy & paste

You are a capacity planner. From the workload and current staffing below, show: where demand exceeds capacity and by how much, where there's slack, the peak periods that break the plan, and 2-3 ways to flex (cross-training, temp, shifting work, automation) ranked by cost and speed. Mark the assumption the whole plan rests on. [paste workload pattern + current headcount]

Why: Planning to the average leaves you understaffed exactly when it matters most.

🔭 Where the signals converge

Where independent intelligence streams — macro podcasters, our email-intel desk, AI-lab feeds — line up on the SAME read, cross-checked against our own live trading algo. The one AI brief that shows you where unrelated operators agree.

AI as an INFLATIONARY / capex-supercycle force (not just deflationary productivity) [MEDIUM · email intel + podcasts agree]

AI capex & compute economics — hyperscaler buildout / 'the AI trade' [MEDIUM · email intel + podcasts agree]

The cross-source read: Cross-source tape today: 2 of 10 convergent themes lean defensive/structural-risk. Highest-conviction edge: 'Mega-cap tech distribution / late-cycle topping' (3-layer agreement, incl. our own algo model). Convergence = where independent operators agree; that's the higher-conviction read for brief/newsletter/algo than any single source.

Where they split: Direction tally across convergent themes: 2/10 lean defensive/structural-risk — the CONVERGENT sub-themes are more informative than aggregate inbox sentiment (which is typically split).

▶ Watch this week · stay an AI expert

No Priors

with Sarah Guo & Elad Gil · YouTube

The clearest read on where AI money and power are moving — VC-grade conversations with the founders building it. Best single show for operators who need the strategic picture, not the code.

One expert AI prompt a day from a rotating discipline — paste it into your own LLM and run it on real work.

MARKETING

The campaign brief builder

For: Marketers, founders, and anyone who owns a message, a funnel, or a launch

Turn a fuzzy 'we should promote this' into a one-page campaign brief your team (or your future self) can execute without meetings.

Copy & paste into your own LLM · archive →

Act as a senior campaign planner. I need a one-page brief for: [WHAT WE'RE PROMOTING]. Audience: [WHO]. Primary goal: [SIGNUPS / PIPELINE / AWARENESS — PICK ONE]. Budget and channels available: [LIST]. Timeline: [DATES].

Build the brief with these sections: 1) Objective — one measurable sentence. 2) Audience insight — the single tension or moment that makes them care now. 3) Core message — one sentence, plus the proof point behind it. 4) Channel plan — for each channel, the format, the hook, and what 'working' looks like in numbers I can actually measure. 5) Risks — the 2 most likely ways this campaign underperforms and the early signal for each.

Format: numbered sections, short lines, no filler. Guardrails: do not invent audience statistics, benchmarks, or performance numbers — if a number matters, write VERIFY next to it and tell me exactly what to look up externally. No confidential data goes into this prompt.

Why it works: The brief structure forces one goal and one message (where most campaigns die is trying to say three things), and the 'early signal' section builds the kill-or-double-down decision in before money is spent.

Useful? 👍 Yes👎 No

SALES

The objection-handling matrix

For: AEs, SDRs, founders who sell, and anyone who owns a quota or a pipeline

Build a reusable objection playbook for your product — honest responses, not verbal judo — so the whole team stops improvising.

Copy & paste into your own LLM · archive →

Act as a sales enablement lead. What we sell: [PRODUCT + PRICE POINT]. Who buys it: [BUYER]. The objections we actually hear, in the buyer's words: [LIST THEM — E.G. 'TOO EXPENSIVE', 'WE USE X', 'NOT A PRIORITY'].

For EACH objection build a matrix row: 1) What the buyer usually MEANS underneath the words (best 1–2 interpretations). 2) The clarifying question to ask FIRST, before answering anything. 3) An honest response for each interpretation — grounded in something true about the product I described, never a claim I'd have to walk back. 4) The proof asset that would help (case study, demo moment, reference) — as a placeholder if we don't have it yet. 5) A 'when to walk away' note: the version of this objection that means they're genuinely not a fit.

Format: one table, objections as rows. Guardrails: do not invent case studies, customer names, ROI numbers, or competitor claims — mark missing proof as NEEDED ASSET and note it must be verified before anyone uses it. Nothing confidential in this prompt.

Why it works: The clarify-first structure stops reps from answering the wrong objection, and the walk-away column builds qualification discipline into the playbook — which paradoxically closes more by chasing less.

Useful? 👍 Yes👎 No

CONSULTING & STRATEGY

The hypothesis-driven workplan

For: Consultants, strategy and ops leads, chiefs of staff, and anyone who structures messy problems for a living

Plan an engagement or internal study backwards from the answer you must produce — the consulting move that saves weeks of undirected analysis.

Copy & paste into your own LLM · archive →

Act as an engagement manager. The question I must answer: [THE QUESTION]. Who's asking and what decision it feeds: [AUDIENCE + DECISION]. Deadline: [DATE]. Resources: [WHO/WHAT I HAVE].

Build the workplan: 1) State the 2–3 competing hypotheses that could answer the question — including the uncomfortable one nobody wants to be true. 2) For each hypothesis: the analyses that would prove or kill it, the data each analysis needs, where that data likely lives, and the fastest acceptable version (proxy or sample) if the ideal data isn't available. 3) A day-by-day plan to the deadline with a midpoint checkpoint where we kill dead hypotheses and reallocate. 4) The 'ghost deck' — the 5 slide headlines I expect to write, marked clearly as PLACEHOLDERS TO PRESSURE-TEST, not conclusions.

Format: hypotheses table, then the calendar. Guardrails: hypotheses are for testing, not asserting — do not present any as likely true before the data is in; every data assumption gets a VERIFY SOURCE tag. Keep client-confidential specifics out of this prompt.

Why it works: Working backwards from competing hypotheses focuses every analyst-hour on discriminating evidence, and the explicit kill-checkpoint stops the classic failure of analyzing everything and concluding late.

Useful? 👍 Yes👎 No

FINANCE

The board-pack readout

For: FP&A, finance leads, founders who own the numbers, and operators who answer to them

Prep the finance section of a board or leadership meeting: the narrative, the two charts that matter, and the questions you'll be asked.

Copy & paste into your own LLM · archive →

Act as a CFO's chief of staff. The metrics I'll present: [PASTE — REVENUE, MARGIN, CASH, RUNWAY, KEY KPIS, WITH LAST PERIOD'S VALUES]. The audience: [BOARD / EXEC TEAM / INVESTOR UPDATE]. What changed since last time: [BRIEF CONTEXT].

Build my prep: 1) The 3-paragraph narrative — trajectory, the one thing to celebrate honestly, the one thing to own before they raise it. Numbers only from my data. 2) Which TWO exhibits carry the story and why (describe them; don't fabricate data points). 3) The 6 questions this audience is most likely to ask given these specific numbers — hard ones included — each with a truthful draft answer built from my data, and where my data cannot answer, the honest line: what we know, what we'll find out, by when. 4) The pre-read one-pager version.

Guardrails: no invented comparisons, benchmarks, or forward numbers — if a question needs data I didn't provide, the answer says so rather than bluffing; flag anything I should verify with accounting before presenting. Share only what you'd put in the actual pack — nothing beyond-confidential in the prompt.

Why it works: Boards trust presenters who name the bad news first and say 'we'll know by Friday' instead of improvising — rehearsing the hard questions against your real numbers is the cheapest confidence money can't buy.

Want the full desk edition? Finance pros get a deeper daily version — role-based prompts for banking, trading, and the buy side. → AI Finance Brief

Useful? 👍 Yes👎 No

PRODUCT

The epic splitter

For: PMs, founders, and anyone who decides what gets built next

Split a big epic into stories that each ship value and carry real acceptance criteria — the difference between a plan and a wish.

Copy & paste into your own LLM · archive →

Act as an agile delivery coach. The epic: [DESCRIBE THE BIG THING]. The user it serves: [WHO]. Constraints: [TEAM SIZE, STACK, DEADLINE IF ANY].

Split it: 1) Identify the 'walking skeleton' — the thinnest end-to-end slice that a real user could actually use, however ugly. That's story #1. 2) Split the rest into stories where each: delivers observable user value on its own, is completable by the team in a few days, and doesn't depend on unfinished sibling stories. Flag any story that fails these tests and re-split it. 3) For each story: title in 'user can X' form, 3–6 acceptance criteria in given/when/then form, and its main technical risk. 4) Sequence them: skeleton first, then risk-reducing stories before polish. 5) List what I've probably forgotten (error states, permissions, migration, observability) as candidate stories.

Guardrails: do not invent effort estimates or velocity numbers — sizing is the team's job; mark stories whose feasibility needs an engineering spike as SPIKE FIRST — VERIFY. No proprietary code or confidential specs needed in this prompt.

Why it works: The walking-skeleton-first sequence front-loads integration risk (where epics actually die), and the 'user can X' + given/when/then discipline makes scope visible enough to negotiate before the sprint, not during it.

Useful? 👍 Yes👎 No

Which vertical would you want as a FULL daily brief?

One tap — it decides which deep-dive brief we build next.

What would you like to see more of?

Know someone who'd get better at AI with this?

The AI Rundown is free — every story, every play, the full archive and the audio edition, always. The best way to support it: forward it to one person who'd actually use it.

Always free · one email a day · unsubscribe anytime