Munition

AI data visualizer

Early access

CSV, XLSX or XLS

or

Early access. Free to start.

AI data visualizer: find the right chart for your spreadsheet

Upload a CSV or Excel file and say what you want people to see. The AI data visualizer picks a chart type that fits, writes a title that states the takeaway, and shows the chart with the summary table behind it.

From table to chart in 3 steps

  1. Upload your table

    A CSV or Excel file with one header row. Munition shows the columns it found and what they seem to hold.

  2. Say what the chart should show

    “Monthly sales for each shop” or “which channel brings the most sign-ups”. Munition proposes a chart type and explains in one line why it fits.

  3. Adjust, check, then copy

    Change the type or labels and check a few values against your table. Then copy the summary table and chart settings, or download them as Markdown, to rebuild the chart in your spreadsheet.

Early access

Turn your next table into a chart people understand

Join early access to upload your first table and get the chart that makes your point. It's free to start.

Charts that are easy to read, not just quick to make

  • A chart type chosen for the question

    Lines for change over time, bars for comparing categories, scatter plots for two measures, stacked bars for parts of a whole, each with a short reason for the choice. Ask for a second type to compare.

  • Titles that state the takeaway

    “Online sales overtook in-store sales in March” rather than “Sales by channel”. You edit the wording; the chart backs it up.

  • Readable defaults

    Bar charts start at zero, axes carry units, categories are sorted, and color highlights what matters instead of decorating everything.

  • A warning before a misleading chart

    Too many slices, a truncated axis, two scales on one chart: Munition says why it may mislead and offers a clearer alternative.

  • The table and settings behind the chart

    The chart is shown on this page. Copy the summary table it plots, with its type, title and axis labels, and rebuild it in Excel or Google Sheets; for simple bar, line and pie charts, the Mermaid code is there too. All of it downloads as Markdown.

  • Here, in ChatGPT or Claude, or by API

    The plan: use it on this page, inside ChatGPT or Claude through the Munition connector, or from your own code through the Munition API. All three come with early access.

Who it's for

  • A physics student writing a lab report

    Checks that a scatter chart suits the measurements, gets the axis labels and units right, and rebuilds the chart in the spreadsheet used for the report.

  • A PhD student making a conference poster

    Asks how to show a dense results table to people standing a few steps away, and gets a simpler chart with fewer series and larger labels to redraw for the poster.

  • A geography teacher using open data

    Uploads a public dataset on population by region and asks for a chart young teenagers can read, with the source noted under it.

  • A marketing manager with a monthly report

    Turns a year of traffic by channel into a line chart whose title says what changed, instead of a grid of numbers nobody reads.

  • A nonprofit preparing its annual report

    Shows where the money went as sorted bars rather than a crowded pie, so donors can compare the categories.

Make a good chart for free

Free tools make good charts once you know which chart to ask for. Munition saves time when you're unsure which chart fits, or want the takeaway title and labels worked out before you open your spreadsheet.

  • Excel: Recommended Charts

    Excel previews the chart types that suit your selection.

    1. Select your table, headers included.
    2. On the Insert tab, choose Recommended Charts and preview the suggestions.
    3. Add axis titles from the chart elements options, then replace the default title with your takeaway.
  • Google Sheets: Insert, then Chart

    Select the data and choose Insert, then Chart. Sheets suggests a type, which you can change in the editor's Setup tab; titles, axis labels and colors are under Customize.

  • Datawrapper for charts you publish

    Datawrapper is built for newsrooms and has a free plan. Paste or upload your data, pick a chart, add notes and the source, then export a PNG. Charts on the free plan carry a small credit line.

  • Choose the chart on paper first

    Write the one sentence the chart should prove, then pick the form that shows it.

    1. Change over time: a line.
    2. Comparing categories: bars, sorted by value.
    3. Relationship between two measures: a scatter plot.
    4. Parts of a whole: a stacked bar, or a pie with very few slices.

Tips for charts people understand

  1. Write the title last, as a sentence that says what the reader should notice.
  2. Sort bars by value unless the categories have a natural order, like months or age groups.
  3. Use one strong color for the series that matters and grey for the rest.
  4. Put the data source and date under the chart.
  5. If you need a paragraph to explain the chart, try a simpler chart.

Frequently asked questions

It's a tool that turns a table of data into a chart. You upload a CSV or Excel file and say what you want to show; Munition picks a chart type, labels it, writes a takeaway title and shows the chart on the page with the summary table behind it.

Start from the question. Lines show change over time, bars compare categories, scatter plots show how two measures relate, stacked bars show parts of a whole. Munition suggests a type and the reason, and you can always switch.

The usual culprits are a bar chart axis that doesn't start at zero, 3D effects, pies with many slices, two different scales on one chart, and a time period chosen to tell a convenient story. Munition flags these and suggests a clearer alternative.

Munition does not hand you an image file. Paste the summary table into Excel or Google Sheets, insert the chart type it names and apply the title and labels: the chart then lives in your own file and updates with it. Before using it in a thesis or report, compare a few values with your original table and note the data source under the chart.

Convert the table to a spreadsheet first, then upload it. Check the converted numbers before charting, because extraction errors go straight into the chart.

Munition is free to start: joining early access costs nothing, and there will be a free plan when the tool opens, with paid plans for heavier use. Without AI, Excel's Recommended Charts, Google Sheets and the free plan of Datawrapper make good charts once you know which one to ask for.

Not yet. The data visualizer is in early access: the plan is for it to work on this page, inside ChatGPT and Claude through the Munition connector, and through the Munition API for developers. In a chat, you would attach the spreadsheet, say what the chart should show and get the chart type, the title and the table back. Join early access and tell us where you would use it: it helps us decide what to build first.

Nothing is uploaded during early access: the page does not open or send your file. If you join early access, we keep only its type and size, to see what people want to use the tool for. When the tool opens, the Munition privacy policy will describe how files are processed, before you use it.

Benjamin Martin · Written with the help of AI tools. · Updated