Visualize key data 3 wildly different ways.
How often are you looking at a spreadsheet, a chart, a mess of data, and have no idea what it’s telling you and why it matters? (And how often are you trying to convey something complicated to a sea of blank faces?)
Try it now
Copy the prompt into your AI tool of choice, then answer its questions as they come.
Paste it into any AI chat to start. For most people, this is all you need.
Everything on this page in one paste: the prompt plus the full plan, every step and what to check. Use it for longer moves, to plan ahead, to share with a colleague, or if your AI can't open links.
How it goes
What the AI will do
- Read your data and help find the story the numbers support.
- Create three finished visuals using contrasting approaches suited to the story and where people will see it.
- Check the visuals against the data and refine your favorite into a usable file or artifact where the tools support it.
What to have ready
- Bring the data in any usable form—spreadsheet, screenshot, PDF, link, pasted numbers, or rough account. The AI will ask for missing figures or unreadable details.
- If you have a presentation you like, share it in the chat. An example is optional.
Where you come in
- Fill in missing context about the audience and viewing format, then confirm the evidence-supported takeaway and tone before design starts.
- Choose your favorite visual and say what you want to change.
What you'll get
Three distinct rendered visual approaches and a refined favorite that conveys the evidence-supported story at a glance, in an agreed format where supported.
You're done when
You have seen three distinct visual approaches and refined your favorite into something that conveys the intended, evidence-supported story at a glance.
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result. ### Start with the activity and the user's context Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear. Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them. ### Check the capabilities needed for this work Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit. When useful access is missing: - Explain what you cannot reach and how that material would help. - Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot. - If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response. - Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available. ### Ask focused questions and respect the answers Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful. Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements. ### Keep evidence and independent contributions distinct Distinguish evidence, inference, and missing information. Check your work against the source material. Use independent agents when your system actually supports delegation and independent investigation or review would improve the work: - Give each agent a bounded brief and the context appropriate to its role. - Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations. - Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately. ### Produce, check, and revise the result Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them. Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result. Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized. Help me turn these numbers into three striking, truthful visual stories. ### 1. Get the material If I have not supplied the data, invite whatever I have: files, links, screenshots, pasted figures, or a spoken or typed account. It can be a mess. Use what you can actually access; offer an alternative my organization permits if a format or tool is unavailable. Extract the figures, units, dates, and relevant comparisons. Ask about consequential ambiguity. If a screenshot is unreadable or a description lacks the numbers, ask for the missing detail rather than inventing it. ### 2. Find the story Before designing, extract the audience, viewing format, intended takeaway, emotional weight, and any reference presentation or preferences I have already supplied. Interview me only about what is missing or consequentially unclear, one question at a time. If the audience or viewing format is missing, ask who will see this and where. If the takeaway is missing or unclear, ask: “After someone sees this and describes it to someone else, what is the short sentence you hope they say?” Help me find the big-picture point if I do not yet know it. #### Confirm the takeaway and tone Use any emotional weight I have already explained. If it is missing or unclear, ask whether this is good news, bad news, catastrophic, extraordinary, or reassuringly routine. Establish what makes it so—the target, expectation, previous period, or other comparison. Check the intended message against the actual data. If they disagree, explain that and help me find the story the evidence supports. Briefly state the evidence-supported takeaway and tone for my confirmation before creating anything. Wait for my confirmation or correction unless I have already explicitly confirmed that same takeaway and tone; supplied context alone is not that confirmation. #### Use an optional reference If I have not supplied a reference or said I have none, ask: “Is there a data presentation you like? You can share a screenshot, upload an example, or paste a link here in the chat.” If I share one, check that you can read it. Use any preferences I have already described and ask what appeals to me only if that is unclear. If the format is inaccessible, invite an accessible version or a description. If I have no example, keep going. ### 3. Create three versions Choose three contrasting approaches from the list below that fit our story, audience, data, and viewing format. Let any example I shared inform your choices. - Edward Tufte — quiet precision: Small, aligned comparisons that make a departure from the pattern jump out. <https://www.edwardtufte.com/notebook/sparkline-theory-and-practice-edward-tufte/> - Hans Rosling / Gapminder — make change move: Animated journeys or successive frames that let you watch the story unfold. <https://www.gapminder.org/tools/assets/guide.pdf> - Giorgia Lupi — make the data human: Distinctive marks and individual detail that give a large dataset texture and personality. <https://giorgialupi.com/data-humanism-my-manifesto-for-a-new-data-wold> - David McCandless — make scale click: Bold visual comparisons that make an otherwise abstract number feel comprehensible. <https://informationisbeautiful.net/2010/the-billion-dollar-o-gram-2009/> - W. E. B. Du Bois and the Atlanta team — a bold visual argument: Striking color, geometry, and direct comparisons that give the evidence presence. <https://www.searchablemuseum.com/w-e-b-du-bois-at-the-1900-paris-exposition/> - Nadieh Bremer — reveal the connections: An explorable web that makes relationships, concentrations, and dependencies visible. <https://www.visualcinnamon.com/portfolio/alis-os/> - Shirley Wu — make the pattern unexpected: Flowers, trees, or another visual metaphor whose shapes and colors encode real values. <https://www.shirleywu.studio/notebook/2025-04-flowers> #### Create each finished visual Where sub-agents are supported and permitted, assign one of your three selected approaches to each sub-agent. Give each the same verified data, takeaway, tone, audience, and viewing format, plus the example I shared and what I liked about it, if any. Tell each agent to follow its assigned approach and create a finished visual. Choose animation or interaction when it serves the story and works in the intended viewing format. If sub-agents are unavailable, create the three versions yourself, following one selected approach for each. #### Verify and present the visuals Keep every encoded quantity faithful to the data. Verify calculations, scales, proportions, labels, and any animation against the source before presenting. Keep references to sources and material limitations available with the output. If the tools cannot render a visual, explain the limitation and help me use an approved option; a design description is not a finished visual. Show the three finished visuals. Make the central point visible through composition, emphasis, color, readable titles, labels, and necessary legends. ### 4. Let me choose Ask which version I like best and what I want to change. Iterate on that version while preserving the data. Provide a usable file or artifact in the agreed format where supported. These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
View full AI instructions
Visualize key data 3 wildly different ways.
What the AI will do
- Read your data and help find the story the numbers support.
- Create three finished visuals using contrasting approaches suited to the story and where people will see it.
- Check the visuals against the data and refine your favorite into a usable file or artifact where the tools support it.
What you can provide
- Bring the data in any usable form—spreadsheet, screenshot, PDF, link, pasted numbers, or rough account. The AI will ask for missing figures or unreadable details.
- If you have a presentation you like, share it in the chat. An example is optional.
Where you'll take part
- Fill in missing context about the audience and viewing format, then confirm the evidence-supported takeaway and tone before design starts.
- Choose your favorite visual and say what you want to change.
What you'll get
Three distinct rendered visual approaches and a refined favorite that conveys the evidence-supported story at a glance, in an agreed format where supported.
Activity context
How often are you looking at a spreadsheet, a chart, a mess of data, and have no idea what it’s telling you and why it matters? (And how often are you trying to convey something complicated to a sea of blank faces?)
Enter some data and we’ll come up with a few ways to present it so it tells the right story clearly and compellingly.
Visual results vary by model and by the tools your organization enables. A disappointing first attempt may be a reason to try another approved model. Ask for actual rendered visuals, then judge what you can see.
For the user
Put the data in any way you can. A spreadsheet, a PDF, a slide, a screenshot, a website, a pile of documents, or some numbers pasted into the chat. Type or talk through what you know. Add a voice note or connect a folder if your app supports it. It does not have to be neat. Use the methods your organization permits; the AI will help you work around an unavailable format or ask for anything it cannot read.
The AI will help establish what matters, building on what you already know. After someone sees this and tells a colleague about it, what’s the short sentence you hope they say? “X is going better than expected.” “We weren’t worried about this. Now it looks like a very real issue.” Is this good news, bad news, an absolute catastrophe, an extraordinary opportunity—or things ticking along as they should?
You do not need to know the answer before starting. The AI will help you find the story in the numbers. If you have seen a presentation you love, give the AI a screenshot, link, or example directly in the chat. If you haven’t, there are plenty of directions to explore.
You’ll confirm the takeaway and tone before the AI starts designing. It will create three very different versions, choosing three of the approaches below to suit the story and any example you shared in the chat. Where supported, it will assign one approach to each of three sub-agents, each creating a finished visual; otherwise the AI will create the three versions itself.
You’ll pick the one that makes you see it and tell the AI what you like and what is still missing: “That one—but make the scale of the problem impossible to miss,” or “Keep the playfulness; make the comparison easier to see.” It will help you refine that version until the visual carries the story and give you a version you can use. If you arrived in your AI chat through this page’s URL, the AI will guide you through the activity. To start directly, copy this prompt.
Guidance for the AI
These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.
Start with the activity and the user's context
Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.
Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.
Check the capabilities needed for this work
Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.
When useful access is missing:
- Explain what you cannot reach and how that material would help.
- Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
- If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
- Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.
Ask focused questions and respect the answers
Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.
Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.
Keep evidence and independent contributions distinct
Distinguish evidence, inference, and missing information. Check your work against the source material.
Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:
- Give each agent a bounded brief and the context appropriate to its role.
- Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
- Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.
Produce, check, and revise the result
Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.
Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.
Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.
Edward Tufte — quiet precision
Small, aligned comparisons that make a departure from the pattern jump out.
https://www.edwardtufte.com/notebook/sparkline-theory-and-practice-edward-tufte/
Hans Rosling / Gapminder — make change move
Animated journeys or successive frames that let you watch the story unfold.
https://www.gapminder.org/tools/assets/guide.pdf
Giorgia Lupi — make the data human
Distinctive marks and individual detail that give a large dataset texture and personality.
https://giorgialupi.com/data-humanism-my-manifesto-for-a-new-data-wold
David McCandless — make scale click
Bold visual comparisons that make an otherwise abstract number feel comprehensible.
https://informationisbeautiful.net/2010/the-billion-dollar-o-gram-2009/
W. E. B. Du Bois and the Atlanta team — a bold visual argument
Striking color, geometry, and direct comparisons that give the evidence presence.
https://www.searchablemuseum.com/w-e-b-du-bois-at-the-1900-paris-exposition/
Nadieh Bremer — reveal the connections
An explorable web that makes relationships, concentrations, and dependencies visible.
https://www.visualcinnamon.com/portfolio/alis-os/
Shirley Wu — make the pattern unexpected
Flowers, trees, or another visual metaphor whose shapes and colors encode real values.
https://www.shirleywu.studio/notebook/2025-04-flowers
Your AI prompt
Help me turn these numbers into three striking, truthful visual stories.
1. Get the material
If I have not supplied the data, invite whatever I have: files, links, screenshots, pasted figures, or a spoken or typed account. It can be a mess. Use what you can actually access; offer an alternative my organization permits if a format or tool is unavailable. Extract the figures, units, dates, and relevant comparisons. Ask about consequential ambiguity. If a screenshot is unreadable or a description lacks the numbers, ask for the missing detail rather than inventing it.
2. Find the story
Before designing, extract the audience, viewing format, intended takeaway, emotional weight, and any reference presentation or preferences I have already supplied. Interview me only about what is missing or consequentially unclear, one question at a time. If the audience or viewing format is missing, ask who will see this and where. If the takeaway is missing or unclear, ask: “After someone sees this and describes it to someone else, what is the short sentence you hope they say?” Help me find the big-picture point if I do not yet know it.
Confirm the takeaway and tone
Use any emotional weight I have already explained. If it is missing or unclear, ask whether this is good news, bad news, catastrophic, extraordinary, or reassuringly routine. Establish what makes it so—the target, expectation, previous period, or other comparison. Check the intended message against the actual data. If they disagree, explain that and help me find the story the evidence supports. Briefly state the evidence-supported takeaway and tone for my confirmation before creating anything. Wait for my confirmation or correction unless I have already explicitly confirmed that same takeaway and tone; supplied context alone is not that confirmation.
Use an optional reference
If I have not supplied a reference or said I have none, ask: “Is there a data presentation you like? You can share a screenshot, upload an example, or paste a link here in the chat.” If I share one, check that you can read it. Use any preferences I have already described and ask what appeals to me only if that is unclear. If the format is inaccessible, invite an accessible version or a description. If I have no example, keep going.
3. Create three versions
Choose three contrasting approaches from the list below that fit our story, audience, data, and viewing format. Let any example I shared inform your choices.
- Edward Tufte — quiet precision: Small, aligned comparisons that make a departure from the pattern jump out. https://www.edwardtufte.com/notebook/sparkline-theory-and-practice-edward-tufte/
- Hans Rosling / Gapminder — make change move: Animated journeys or successive frames that let you watch the story unfold. https://www.gapminder.org/tools/assets/guide.pdf
- Giorgia Lupi — make the data human: Distinctive marks and individual detail that give a large dataset texture and personality. https://giorgialupi.com/data-humanism-my-manifesto-for-a-new-data-wold
- David McCandless — make scale click: Bold visual comparisons that make an otherwise abstract number feel comprehensible. https://informationisbeautiful.net/2010/the-billion-dollar-o-gram-2009/
- W. E. B. Du Bois and the Atlanta team — a bold visual argument: Striking color, geometry, and direct comparisons that give the evidence presence. https://www.searchablemuseum.com/w-e-b-du-bois-at-the-1900-paris-exposition/
- Nadieh Bremer — reveal the connections: An explorable web that makes relationships, concentrations, and dependencies visible. https://www.visualcinnamon.com/portfolio/alis-os/
- Shirley Wu — make the pattern unexpected: Flowers, trees, or another visual metaphor whose shapes and colors encode real values. https://www.shirleywu.studio/notebook/2025-04-flowers
Create each finished visual
Where sub-agents are supported and permitted, assign one of your three selected approaches to each sub-agent. Give each the same verified data, takeaway, tone, audience, and viewing format, plus the example I shared and what I liked about it, if any. Tell each agent to follow its assigned approach and create a finished visual. Choose animation or interaction when it serves the story and works in the intended viewing format. If sub-agents are unavailable, create the three versions yourself, following one selected approach for each.
Verify and present the visuals
Keep every encoded quantity faithful to the data. Verify calculations, scales, proportions, labels, and any animation against the source before presenting. Keep references to sources and material limitations available with the output. If the tools cannot render a visual, explain the limitation and help me use an approved option; a design description is not a finished visual.
Show the three finished visuals. Make the central point visible through composition, emphasis, color, readable titles, labels, and necessary legends.
4. Let me choose
Ask which version I like best and what I want to change. Iterate on that version while preserving the data. Provide a usable file or artifact in the agreed format where supported.
Done when
You have seen three distinct visual approaches and refined your favorite into something that conveys the intended, evidence-supported story at a glance.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.
These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.