Curated AI prompts for business, creative, and technical use cases.
Works well with any current-generation reasoning model — GPT-5.4, Claude Sonnet 5, and Gemini 3 Pro all handle multi-step analogical mapping accurately. Weaker or older models tend to stop at a single anchor analogy and skip the mapping table or the "where it breaks" section unless explicitly prompted to continue through the full structure.learningA product manager with eight years running retail store operations is joining a database architecture review next week and needs to actually understand how sharding works — not just nod along — because she's grown up in grocery logistics, not distributed systems, and generic explainer articles keep losing her at the third paragraph.
A structured prompt that maps every working mechanism of an unfamiliar concept onto direct equivalents in a domain you already understand deeply, with an explicit list of where the analogy breaks down.
prompt-engineeringlearning
Works well with GPT-5.4 and Claude Sonnet 5 — both handle platform-specific character constraints and distinct psychological framing reliably. Gemini 3 Pro produces strong copy but occasionally runs over stated character limits, so budget a quick manual trim pass if you use it.marketingYou're the performance marketing manager at a 15-person direct-to-consumer skincare startup. Your CEO wants a paid campaign live on Meta and Google Search by tomorrow morning for a new $42 vitamin C serum, with a $5,000 weekly test budget approved. Marketing is a two-person team, nobody has bandwidth tonight to brainstorm five genuinely distinct ad angles from scratch, and the launch date isn't moving.
A structured prompt that takes a single product brief and returns five genuinely different ad variants across psychological angles, each paired with a testable A/B hypothesis instead of just five versions of the same idea.
prompt-engineeringadvertising
Works well with any current-generation model — GPT-5.4, Claude Sonnet 5, and Gemini 3 Pro all handle spreadsheet formula logic and edge-case reasoning accurately. Smaller or older models occasionally suggest deprecated functions (like legacy array-entry CSE formulas) or mix Excel and Google Sheets syntax, so it's worth double-checking the output against the tool you specified.Data AnalysisYou're a sales operations analyst at a mid-size SaaS company, working in a Google Sheets pipeline tracker with roughly 3,000 rows. Finance needs the weighted average discount percentage across all deals marked "Closed Won" in Q3, weighted by deal value, and a few rows have a blank discount field because a rep forgot to fill it in. The number is due in the quarterly business review deck in two hours, and you need a formula that handles the blanks correctly instead of silently including them as zero and skewing the average down.
A structured prompt that takes a calculation described in ordinary language and returns a copy-paste-ready Excel or Google Sheets formula, with a step-by-step breakdown of how it works and what could break it.
prompt-engineeringdata analysis
Claude Opus 5 (also works well with GPT-5.4 and Gemini 3 Pro)CareerYou have a final-round interview in three days for a senior product manager role, and you know behavioral questions are coming, but your work history spans five different projects and you're not sure which stories map to which likely questions — or where your prep has gaps.
A three-pass prompt that predicts the behavioral questions a job will ask, maps them to your real work history, and writes STAR-format answers — while flagging the gaps where you don't yet have a strong story.
prompt-engineeringCareer
Claude Opus 5 (also works well with GPT-5.4 and Gemini 3 Pro)Developer ToolsYou just merged a pull request that adds a new POST /api/v1/refunds endpoint to your team's billing service. The endpoint works, tests pass, but nobody wrote documentation — and your team lead just asked you to get docs into the developer portal before external partners can start calling it next week. You have the code in front of you and forty-five minutes before your next meeting.
A two-pass prompt that reads your endpoint code, extracts the actual contract, and writes reference documentation that flags ambiguity instead of guessing — built for the gap between “the code works” and “the docs are live.”
prompt-engineeringdeveloper tools
Claude Opus 5 (also works well with GPT-5.4 and Gemini 3 Pro)Writing & CommunicationYou have a 47-page customer research report due to the CEO before Monday's board meeting, and you know from experience that a report this long simply won't get read before the meeting starts. You need the actual findings — the numbers, the causal story, the recommendation — condensed into something a time-pressed executive will read in the elevator on the way to the boardroom.
A structured prompt that compresses long reports into a single-page executive brief with traceable findings, clear implications, and specific recommended actions — built to survive a five-minute skim from someone who won't read past page two.
prompt-engineeringproductivity
Claude Opus 5 (also works well with GPT-5.4 and Gemini 3 Pro)finance-negotiationYou've received a written job offer with a base salary $8,000 below what levels.fyi and three Glassdoor data points show is market rate for the role. You have one week before you need to confirm your start date, and your current employer hasn't offered to match. You know you should negotiate, but you're not sure what number to counter with, how to justify it without sounding entitled, or what to say if the recruiter pushes back with "this is our best offer."
A structured prompt that transforms your compensation research and competing offers into a complete negotiation script, with anchor numbers, objection responses, and a walk-away threshold you actually calculated instead of guessed.
prompt-engineeringCareer
Nano Banana Pro (also works with Midjourney v7 and DALL-E 3)creativeA homeowner is getting renovation quotes and wants to visualize a specific style direction before committing to a contractor, without paying a designer for a mood board.
A structured image-generation prompt that renders a realistic after-photo of a room renovation while preserving fixed architectural elements and matching a real budget tier.
image-generationprompt-engineering
Claude Opus 5 (also works well with GPT-5.4 and Gemini 3 Pro)Startups & BusinessA product marketing lead is refreshing messaging ahead of a Q3 rebrand and needs to know, with evidence, which competitive claims actually hold up before locking in a new tagline.
A prompt that pressure-tests your product positioning against real competitors and flags any claim you cannot actually defend.
prompt-engineeringmarketing
Works well with GPT-5, Claude Sonnet 5, or Gemini 3 Pro — this is primarily a pattern-recognition and arithmetic task, so mid-tier reasoning models handle it reliably. Larger models add value mainly in phrasing the bias explanation clearly.productivityA freelance web developer has blown three consecutive project deadlines, each time because "one more round of client revisions" ran long, and has a new project proposal due tomorrow where the client is specifically asking for a firm time-and-cost estimate broken down by task.
An AI prompt that mines your own history of estimated-versus-actual task times to calculate a category-specific bias correction, so your next quote is calibrated instead of guessed.
prompt-engineeringproductivity
Works best with Claude Opus or GPT-5 for nuanced persona consistency across multiple rounds. Gemini 3 Pro is a solid alternative; smaller models tend to let personas agree too easily instead of staying skeptical.marketingA B2B SaaS marketing lead has a positioning statement drafted for next month's product launch and an exec review meeting in two days, but has a nagging feeling the current wording is generic enough that it would not survive a real skeptical buyer conversation — and there is no time to run a proper customer discovery session before the deadline.
A three-round AI role-play that puts your positioning statement in front of a skeptical champion, economic buyer, and CFO — before a real prospect ever sees it.
prompt-engineeringmessaging
Works best with Claude Sonnet 5 or GPT-5.4 — both reliably apply named statistical concepts (Simpson's paradox, survivorship bias) to a specific dataset rather than listing them generically. Gemini 3 Pro is a solid alternative when the pasted dashboard data is very long.Data AnalysisYou're a growth PM about to tell the exec team that a redesigned onboarding flow increased 30-day retention by 12%, based on a dashboard comparing users who completed the new flow against users who went through the old one last quarter — and you have a strategy meeting in an hour where this number will justify rolling the flow out to 100% of new signups.
A prompt that reviews any metric or dashboard for the statistical illusions that make bad decisions look data-driven — survivorship bias, Simpson's paradox, cherry-picked date ranges — before you walk into the room and stake your credibility on the number.