--- id: copilot-advanced-prompts-by-goal lang: en track: copilot title: "Advanced prompts by goal" seoTitle: "Advanced Copilot prompts by goal" description: "Ready-to-copy Copilot prompt patterns for writing, analysis, research and coding, plus naming your scope so Copilot draws on the right sources for the job." part: prompting partTitle: "Prompting" partOrder: 3 order: 4 level: Intermediate durationMinutes: 6 outcome: "Reach for the right prompt pattern for writing, analysis, research, or coding, and know when to name your scope." summary: "Different jobs want different prompt patterns: ready-to-copy recipes for writing, analysis, research, and coding, plus naming your scope, the one extra thing Copilot needs." draft: false updated: "2026-10-14" furtherReading: - title: "Get better Copilot responses with great prompting" url: "https://support.microsoft.com/en-us/microsoft-365-copilot/get-better-copilot-responses-with-great-prompting" note: "Microsoft's tips for getting more useful answers from Copilot." - title: "Prompt Gallery" url: "https://adoption.microsoft.com/en-us/copilot/prompt-gallery/" note: "Ready-made prompts from Microsoft to copy and adapt." - title: "What information does Copilot use to answer my prompt?" url: "https://support.microsoft.com/en-us/microsoft-365-copilot/what-information-does-copilot-use-to-answer-my-prompt" note: "Microsoft explains where Copilot gets the information behind its answers." - title: "Get started writing prompts in Microsoft Copilot" url: "https://support.microsoft.com/en-us/microsoft-365-copilot/get-started-writing-prompts-in-microsoft-365-copilot" note: "A beginner's starting point for prompting in Copilot." keywords: - advanced Copilot prompts - Copilot prompts for writing - Copilot prompts for analysis - Copilot research prompts - Copilot coding prompts - naming your scope quiz: intro: "Five quick questions. Ace them all and the next lesson unlocks." passScore: 5 questions: - q: "For a writing task, what lifts an answer most reliably?" options: - "A long list of adjectives describing the tone" - "Telling Copilot to \"sound professional\", with no further detail" - "A sample of the tone you want, plus naming the reader" - "Leaving the reader out so the text works for anyone" answer: 2 explain: "Giving a sample of the tone you want and naming the reader, rather than describing the tone in adjectives. \"Match this\" beats five adjectives." - q: "When you hand Copilot data or text to analyse, what should you always add?" options: - "The question you actually want answered, and what to look for" - "Nothing, \"analyse this\" is enough" - "A request for one confident answer, without caveats" - "As much extra data as you can find" answer: 0 explain: "The question you actually want answered, and what to look for, not just the raw data on its own. Copilot can't read your mind about what you're looking for." - q: "What's the single most important habit for research prompts?" options: - "Ask for one tidy number so it's easy to repeat" - "Trust the answer if it sounds confident" - "Keep the prompt as short as possible" - "Ask for sources and check the important claims yourself" answer: 3 explain: "Ask for sources and check the important claims yourself. Confident and correct are not the same thing." - q: "Why does a Copilot prompt sometimes need to name its scope in a way a Claude prompt doesn't?" options: - "Because Copilot starts from a blank slate" - "Because Copilot is already grounded in your mail, files, and chats, so it can pull from anywhere" - "Because Copilot can only search one file at a time" - "It never does; scope only matters in Claude" answer: 1 explain: "Copilot starts already grounded in your mail, files, and chats, so a prompt without a named file, mailbox, or time range can pull from anywhere and return a different answer each time." - q: "What do all four prompt patterns share?" options: - "Nothing, they're four different skills" - "They all lean on a sample input and output" - "A clear goal, the right context, and the shape of the answer" answer: 2 explain: "They give Copilot a clear goal, the right context, and the shape of the answer: the same recipe wearing four hats." pass: badge: "Perfect score" title: "You've finished the Prompting part." body: "Full marks. You have a pattern for writing, analysis, research, and coding, and you know when to name your scope. Next up: Part 4, Working smart, starting with context and grounding." retry: "Check the marked answers and give it another go. You've got this." --- > **Level:** Intermediate · **Reading time:** ~6 min > > **What you'll be able to do:** reach for the right prompt pattern for writing, analysis, research, or coding, and know when to name your scope. By now you know the recipe (role, task, context, examples, format) and you know how to steer a reply. This lesson is about noticing that different kinds of work want slightly different prompts. The recipe never changes underneath. But a prompt for writing an email leans on tone and examples, while a prompt for analysing a spreadsheet leans on the exact question you're asking. Here are four patterns for the four jobs you'll hit most: writing, analysis, research, and coding. Each comes with a worked example built around Copilot. Steal them, adapt them, keep the ones that fit your week. ## Writing: drafts, edits, and tone Writing tasks live and die on two things: who's reading, and what it should sound like. So lead with those. The single biggest upgrade is giving a sample of the voice you want instead of describing it. "Match this" beats five adjectives. A pattern that holds up: *You're writing as me: [who you are, in a line]. The reader is [who], and they care about [what]. Task: [the one thing to write]. Tone: like this sample [paste something you've written]. Not stiff, not salesy. Format: [length and shape]. End with [the ask or next step].* Worked example, a Teams post about a product update: *You're writing as me, a founder who hates corporate cheerleading. The reader is a busy operations manager scrolling on their phone. Task: a short post announcing we cut onboarding from two weeks to two days. Tone: like this sample [paste a post you liked of your own]. Dry, human, no "thrilled to announce." Format: under 100 words, one clear point, a line break before the last line so it lands. No hashtags.* You'll get something in your voice, not the default corporate hum. If the first pass is off, steer it with one specific note, that's the [last lesson](/learn/copilot/prompting/iterating). ## Analysis: turning data and text into findings The mistake here is asking Copilot to "analyse this" and hoping it reads your mind about what you're looking for. It can't. Analysis prompts need a real question and a clear idea of what a useful answer looks like. The pattern: *Here's [the data or text, attached or in the sheet]. Question: [the specific thing you want to know]. Look for: [what matters: trends, outliers, the why behind a number]. Give me: [the shape: top three findings, a short table, one recommendation]. Flag anything you're unsure about rather than guessing.* Worked example, a month of support tickets in Excel: *Here's last month's support tickets [the open sheet]. Question: what are people actually struggling with, and is it getting worse or better than the month before? Look for: the top themes by volume, and any theme that's growing fast even if it's still small. Give me: the top five issues as a short table (issue, rough count, trend), then one line on the thing I should fix first. Say so if the data's too thin to call.* That last instruction matters. Telling Copilot to flag its own uncertainty is how you stop a confident-sounding number you can't trust. Which brings us to research. ## Research: gathering and checking sources Research is the job where a wrong answer does the most damage, because it sounds just as sure as a right one. Copilot's default web grounding makes this easy to reach for, which is exactly why the whole pattern is built around one habit: ask for sources, and check the ones that matter. The pattern: *Research [the question]. I want: [what you're after, and how deep]. For each key claim, give me the source and a link. Separate what you're confident about from what's shaky. If sources disagree, tell me.* Worked example, checking a market before a pitch: *Research the market for meal-kit delivery in Belgium. I want: rough market size, the three biggest players, and one trend that's changed in the last year. For each number, give the source and a link and the date it's from. Flag anything you can't back up. If the figures you find disagree, show me the range, don't average them into one tidy fake number.* Then do the part no prompt does for you: open the links on the claims you're about to repeat out loud. Copilot is a fast researcher, not a fact-checker. Confident and correct are different things, and the gap between them is your job to close. ## Naming your scope Here's the one place a Copilot prompt needs something a Claude prompt doesn't. Claude starts from a blank slate, so a vague prompt is vague but at least predictable. Copilot starts already grounded, in your mail, files, and Teams chats. That's a head start, but it also means a prompt without a named scope can pull from anywhere, and the same question can return a different answer depending on what happened to be recently active. The fix is simple: for anything beyond a quick, everyday question, say which file, mailbox, channel, or time range you mean. "Summarise our Q2 numbers" can mean five different documents. "Summarise the Q2 numbers in the Finance Review Teams channel, posted this month" can only mean one. The narrower you point it, the less Copilot has to guess, and the more repeatable the answer is if a colleague asks again tomorrow. ## Coding: explaining, writing, and fixing You don't need to be a developer to use this. Plenty of business work touches a formula, a script, or a snippet someone left behind. Coding prompts want three things: the language, the goal in plain words, and an example of the input and the output you expect. The pattern: *[Language or tool, for example: a formula for Excel.] Goal: [what it should do, in plain words]. Input looks like: [a sample]. Output should look like: [a sample]. Explain what it does in one line so I can check it.* Worked example, a spreadsheet formula: *An Excel formula. Goal: in column D, flag any order over 1000 euros that's still unpaid. Input: column B is the amount, column C is the status ("paid" or "unpaid"). Output: the word "CHASE" if it's over 1000 and unpaid, otherwise blank. Give me the formula and one line on how it works.* Because you gave a sample in and a sample out, you can check the result against your own sheet in seconds. Asking it to explain the formula in a line means you're never pasting in something you don't understand. If it breaks, tell Copilot what you expected and what you got, and it'll fix it. ## The thread running through all four Notice these aren't four different skills. They're one recipe wearing four hats: a clear goal, the right context, and the shape of the answer you want. Writing leans on tone and examples. Analysis and research lean on the exact question and on honesty about what's uncertain. Coding leans on sample input and output. Same bones, different emphasis, plus one extra habit, naming your scope, worth carrying into all four.
One recipe, four jobsA central recipe: role, task, context, examples, format, connected to four job cards: writing, analysis, research, coding. Same bones, different emphasis per job.One recipe, four jobsRole · Task · Context · Examples · FormatWWritingTone andexamples.AAnalysisThe exactquestion.RResearchSources,checked.CCodingSample in,sample out.Same recipe wearing four hats, plus naming your scope, for Copilot specifically.
## Try it Pick the goal closest to your actual work right now: writing, analysis, research, or coding. Take the matching pattern above and fill it in with something real from your week, naming a specific file, mailbox, or channel if the task calls for it. Run it. If the first answer is close but not there, don't start over. Steer it with one specific follow-up. You've now used every tool in the Prompting part on real work, which was the whole point. ## Key terms [Prompt](/glossary#prompt), [Prompt engineering](/glossary#prompt-engineering), [Context](/glossary#context), [Grounding](/glossary#grounding), [Copilot](/glossary#copilot)