Explainer
Stop Pasting Charts as Pictures
The Ritual
A Rational Act of Self-Defense
The workflow is always the same. The analysis happens in Excel, where the data is alive. Then, at publishing time, each chart is copied into the report as a picture — because a hundred live chart objects make the document unstable, because images can at least be sized, because everyone has watched a big file die and learned the lesson. Inside a tool that breaks at scale, flattening your evidence into pixels is the smart play.
But look at what the organization just did: it took its most expensive content — the modeled, checked, argued-over numbers — and published a photograph of them. The data that drove the chart stays behind in a spreadsheet nobody will find in three years. The publication of record carries the souvenir.
The Cost
Pixels In, Nothing Out
Downstream, the picture fails every reader except the one it was styled for. It can't reflow for a phone. A screen reader gets a caption where sighted readers get evidence — a compliance exposure for any public body, and a quiet unfairness everywhere else. Search and analysis tools mine nothing from it; the organization's own analysts end up re-typing figures from screenshots of reports their colleagues wrote. And when the underlying number changes, someone regenerates and re-pastes a dozen images by hand — or, worse, doesn't.
In structured publishing the trade never has to be made: data goes into the system as data, and every chart is a rendering — sized for the PDF page, responsive in the browser, table-backed for assistive technology and APIs, regenerated when the numbers move. Stability isn't the price, because a document assembled from managed components has no fragile monolith to protect. It's one of the clearest examples of the whole humans-and-machines argument in miniature: keep the meaning, and every presentation comes for free.
FAQ
Questions We Hear
Why do teams paste charts as images in the first place?
Self-defense. A hundred live Excel objects make a big Word document heavier, slower, and likelier to crash, and embedded charts resist consistent sizing in the layout. Flattening each one to a picture is the rational move inside that tool — which is exactly the point: the ritual isn't a mistake, it's evidence the authoring environment can't carry live data at the document's real scale.
What actually gets lost when a chart becomes a picture?
The numbers. A pasted chart keeps the pixels and discards the data behind them — nothing can be recalculated, restyled, resized for a phone, read by a screen reader, or extracted by any system mining the publication later. Analysts end up re-keying figures out of screenshots of their own organization's reports. The person who made the chart had the data; the published document has a souvenir of it.
How do charts work in structured publishing?
The data lives in the system as content, and the chart is a rendering of it — regenerated per output, sized for the page in the PDF, responsive on the web, with the underlying table available to assistive technology and APIs. Because the document is assembled from managed components rather than one fragile file, a hundred live charts add no instability; there is no monolith for them to crash.
Does this matter for accessibility?
Enormously. A chart-as-image with a one-line alt text gives a screen-reader user a caption where everyone else gets the evidence — which for public bodies is a compliance exposure, not just a courtesy gap. Charts rendered from live data carry their table with them, so every reader, human or machine, gets the actual numbers.
Get In Touch
Publishing Souvenirs?
If your reports are full of screenshots of your own data, the fix is an authoring environment that can carry the data live. That's buildable.