Quick disclosure before I get into this: I work on PlotSet, so this isn't a neutral third-party review. I'll try to write it the way I'd actually want to read it though, trade-offs included, not just the good parts. The part of chart-making nobody talks about Making a chart from a spreadsheet is a solved problem at this point. Every tool handles the basics fine: bars, lines, pies, pick a template and go. What's still annoying is what happens after the chart exists. Almost every dataset ends up needing to become a few different things: A PNG or SVG for a slide deck or a PDF report A live embed for a blog post or an internal dashboard Increasingly, a short video clip for LinkedIn, Instagram, or the intro of a YouTube video The usual way people handle that last one is exporting a static image, opening a separate video tool, animating it by hand, and rendering it out. It works, but it's slow enough that most people just skip it and post a screenshot instead. And the second the underlying numbers change, you're redoing the whole thing from scratch. What we actually built We call it Video Chart. The idea is that instead of bolting animation onto a chart after it's already built, the chart is a timeline from the start. You upload a CSV, an Excel file, or a Google Sheet (or just paste the data in), pick a template like a bar chart race, an animated line chart, or a map that moves between regions, and the platform handles the motion between data points instead of only plotting a final snapshot. Once it's built, you can export the same chart three ways: a static PNG, SVG, or PDF like any normal chart tool, a live embeddable widget for a webpage, or a rendered MP4 or GIF, up to 4K at 60fps, with your own music or voiceover if you want it. The rendering itself happens on our servers, not your laptop, so it's not chewing through anyone's battery or GPU. You submit the render, go do something else, and get a notification when the file's ready. Most charts finish in under a minute. 90s test — PlotSet | PlotSet A data video created with PlotSet. plotset.com The parts that turned out to be genuinely hard A few things worth mentioning if you're curious how this actually works under the hood. Turning data into motion isn't just interpolation. A static chart only needs final positions. An animated one has to decide how fast values move between points, when labels should swap order (this matters a lot for bar chart races, where the whole appeal is watching rankings flip), and how to avoid the jittery, stuttery look that naive frame-by-frame animation tends to produce. Figuring out the time axis on someone's behalf is trickier than it sounds. Most spreadsheets don't come with a clearly labeled time column. We built schema detection that guesses which column is temporal so people don't have to configure that by hand every time. Pacing turned out to be a design problem more than a technical one. A chart that reaches its conclusion in the first two seconds feels rushed. One that crawls loses people before anything interesting happens. We ended up building pacing presets (slow start, faster middle, clear finish) instead of exposing raw keyframe timing, mostly because almost nobody actually wants to hand-tune easing curves for a chart. I know I wouldn't. Where this makes sense, and where it probably doesn't If all you need is a polished bar chart race video and nothing else, a smaller, more specialized video tool might genuinely be the better call: cheaper, and with more granular video controls. Where PlotSet earns its keep is when the same dataset needs to live in more than one place: an embed on a live dashboard, a static export for a report, and a video for social, all coming from one source instead of three tools that don't talk to each other. And not every chart benefits from motion, honestly. Data that changes over time (rankings, growth curves, geographic spread) genuinely gets easier to understand when it's animated. A single comparison between two numbers usually doesn't need to move at all, and we don't force it to. If you want to poke at it The gallery on plotset.com has real examples if you'd rather look at finished videos before touching your own data. There's a free plan too, if you just want to try it without committing to anything. Curious how other people here are handling the "one dataset, three output formats" problem. Would love to hear what's working for you, or what isn't, in the comments.