How a Wi-Fi heatmap is actually made

A Wi-Fi heatmap looks like a photograph of the signal. It isn't. It's a handful of real measurements, plus a calculation that fills in everything in between. Understanding that split makes you much better at reading one, and much better at making one.

What's real and what's inferred

Two things go into any heatmap:

1. Measurements. At a point on the floor plan, the app records how the connection responds right there: how quickly, how steadily, and whether anything gets lost.
2. Interpolation. For every pixel between those points, a value is estimated from the nearby measurements, weighted so that closer readings count more than distant ones.

That second step is the map's opinion, not its evidence. Where you measured densely, the opinion is well informed. Where you left a big gap, it's a guess drawn from far away.

Why the algorithm doesn't know about your walls

Interpolation works on distance, not on architecture. It has no idea that between two of your points there is a concrete wall, a lift shaft or a fuse box. So it will happily paint a smooth gradient across a barrier that in reality causes a cliff.

This is the single most useful thing to know, because it tells you exactly where to spend your measurements: near boundaries. Two readings, one on each side of a thick wall, turn an invented gradient into a measured drop. See what walls do to Wi-Fi.

How to measure well

• Spread out. Even coverage of the floor plan beats a cluster in one room, however careful.
• Take a reading in every corner you care about, not just in the middle of rooms.
• Add points at doorways and on both sides of thick walls.
• Keep the device at a consistent height, roughly where you'd actually use a phone or laptop.
• Pause for a moment at each point. Signal readings fluctuate; a rushed walk produces noisy data.
• Use the same device for the whole survey. Phones differ in antennas and in how they report; see reading the Wi-Fi signal on an iPhone.

What a heatmap can't show you

A map shows where the connection is good and where it isn't, not why. A spot close to the router that still measures badly is common and means the problem isn't distance: congestion from neighbours, interference from something on 2.4 GHz such as a baby monitor or the microwave, or a device stuck on the wrong band. That's why full bars but slow internet is its own story.

Measure each band separately, too. 2.4 GHz and 5 GHz produce different maps of the same home, and mixing them into one picture tells you very little.

Use it as a before and after

The real value isn't the first map, it's the second one. Measure, change one thing — move the router, add a node, raise it off the floor — and measure again along the same route. WiFi Heatmap keeps both, so you can see whether the change actually did anything instead of trusting the feeling that it did.