It's absurd that adjusting a chart axis still feels like fighting the tool. When a user sets a graph scale to 0-2, the software should respect that instruction. Instead, it reverts to 0-12, forcing a manual override that often fails anyway.
This isn't a niche frustration. Anyone who has built a chart in a traditional spreadsheet has experienced this. You want to show a subtle trend, monthly sales growth of one point, or a temperature shift of two degrees, and the software decides your data is more dramatic than it actually is. It stretches the axis to fit an imagined range, flattening your story into a horizontal line. The result is a graph that misleads at a glance. The user in this post did exactly what any reasonable person would do: they typed the correct limits. The software ignored them. That is not a user error. That is a design failure.
The deeper issue is that legacy tools treat charting as a fixed, rigid process. They assume you will accept their defaults because defaults are easier to code. But data does not always cooperate. A clean, honest visualization depends on precise control over axes, labels, and spacing. When that control is taken away, the user loses the ability to communicate clearly. The graph becomes a source of confusion rather than insight.
An AI-native spreadsheet should not guess what your data looks like. It should listen. When you say the range is 0 to 2, the tool should lock that in, adjust the tick marks accordingly, and leave the rest of your chart intact. It should learn that you prefer tighter scales for certain metrics and apply that preference consistently. The fix is not a better manual override. The fix is a tool that treats your intent as the default. Stop fighting your chart software. Demand one that follows your lead.