How to Map Wi-Fi Coverage in Your Home (Without Buying a Survey Tool)

You already know which room is bad. What you do not know is where the signal actually falls off — which wall it dies at, which outlet still has a usable link back to the router, whether you have one dead corner or one dead half of the house. That is what a coverage map answers, and you can build a useful one with a phone, a sheet of paper and about twenty minutes.

Here are the three methods worth knowing, what each one is honestly good for, and how to read the result.

What a coverage map is for#

A coverage map exists to make one decision: where hardware goes. Router, extender, mesh node, access point, cable. Everything else it shows you is context.

That framing matters because it tells you when to stop. You do not need a smooth, photo-realistic heatmap. You need enough points to know where the signal crosses from "fine" to "not fine", and what it crossed to get there.

The threshold that does most of the work is around -67 dBm. Above that, HD video calls and 4K streams generally hold together. Between -68 and -74 dBm you are in the marginal zone, where things work until they suddenly do not. From -75 to -84 dBm the link connects and barely works, and by -85 dBm the spot is effectively dead. For the full scale and what each band means per activity, see what dBm numbers actually mean.

Method 1: the walk test#

The walk test is the only method that measures what is actually happening in your house today. It is also the slowest, and you have to redo it every time you move anything.

Reading the signal: on Android, RSSI appears in the network details screen or in any analyzer app. On iOS, the operating system does not expose RSSI to third-party apps at all, so use your router's admin page or app — nearly all of them list every connected client with a signal figure — and read your phone's entry while you stand in each spot.

Then follow this literally:

1. Sketch the floor plan on paper. Rough is fine. Mark the router with an X.
2. Mark your test points on a grid roughly 8 to 10 feet apart, plus one extra in every corner and one wherever you actually sit, work or stream in each room. Number them.
3. Pick one band and stay on it. If your 2.4 and 5 GHz networks share an SSID, your phone may switch mid-walk and ruin the comparison. Split the SSIDs temporarily, or note the band at every point.
4. Stand still at point 1 with the phone at chest height, screen facing you, held away from your torso. Your body absorbs several dB.
5. Wait ten seconds, then take three readings a few seconds apart and write down the middle one. RSSI jitters by 3 to 5 dB even when nothing moves.
6. Repeat for every point, walking in a consistent order so you can redo the same walk later.
7. At doorways, record twice — door open and door closed. A solid door is worth several dB, and it changes which room the map says is bad.
8. Write the time of day at the top. Evening congestion from neighbors will not change your RSSI much, but it changes how the same RSSI performs.

Good for: ground truth, confirming a suspicion, before-and-after proof. Bad for: comparing places you have not put hardware yet, which is usually the actual question.

Method 2: the spreadsheet grid#

The walk gives you numbers. The grid is what turns them into a decision instead of a pile of readings. Use one row per test point and these columns:

That table is a worked illustration, not a measurement of anyone's house — your numbers will differ. What matters is the shape of it.

The "walls from router" column is the one people skip, and it is the column that tells you what to do. Two points can read -74 dBm for completely different reasons: one is far away through open air, the other is ten feet away through a chimney breast. The first can be fixed by nudging the router. The second cannot. If you are unsure what your walls cost you, the loss per building material is worth ten minutes.

Keep the sheet. When you move the router or plug something in, add a second RSSI column rather than overwriting the first. A before-and-after column pair is the most persuasive thing you can own when deciding whether to keep a device or return it.

Method 3: floor-plan simulation#

A simulation does not measure anything. It models your home — walls, materials, router position — and predicts signal at every point on the floor plan. That sounds like a downgrade until you notice what it gives you in exchange: you can test positions you have not tried.

That is the difference that matters. A walk test can tell you the back bedroom is at -74 dBm. It cannot tell you what the back bedroom would read if the router moved to the hallway, or if an extender went in the landing outlet, without you physically doing it and walking the whole house again. A simulation compares five candidate positions in the time it takes to tap five spots.

The trade-off is real: a prediction is only as good as the model. Get a wall material wrong — call a brick chimney "drywall" — and the map will be optimistic in exactly the direction you do not want. It also cannot see your neighbor's channel congestion or a microwave, because those are not geometry.

Use it for planning. Use the walk test to verify afterward. The two methods answer different questions and neither replaces the other.

Which method answers which question#

Rules that keep the map honest#

Most bad coverage maps are not wrong about Wi-Fi. They are inconsistent about method.

• One device for the whole map. Two phones side by side can report readings that differ by 5 to 10 dB because of antenna design and calibration. A map built from a mix of devices is unusable.
• One height. Chest height, every point. Floor-level readings are systematically worse.
• One band. Never put a 2.4 GHz and a 5 GHz reading in the same column. At the same spot, 2.4 GHz commonly reads 4 to 8 dB stronger and still delivers less throughput.
• Median of three. A single reading is noise.
• Same walk order, same doors. If you re-run the test after a change, reproduce the conditions or the comparison means nothing.

Turning the map into a placement decision#

Draw the -67 dBm contour on your sketch: a loop around the router enclosing every point stronger than that. Now look at the shape.

• One small pocket outside the loop. A single extender or node is likely enough. Put it where the router still reads -60 to -67 dBm, on the router side of the problem — the rule and its exceptions are covered in the halfway rule and where it breaks.
• The loop is off-center, hugging one end of the house. Your router is in the wrong place. Fix that before you buy anything, because moving the router is free and often worth 10 dB or more across the far half of the plan.
• The loop stops dead at one wall. That is a material problem, not a distance problem. Route around it through a doorway rather than trying to punch through it.
• Two or more separate pockets far apart. One extender will not do it. Plan node positions properly instead of chaining devices.
• No pocket at all — everything inside the loop, still slow. An extender is the wrong tool. Stop here and diagnose throughput, not coverage.

That last one is the most valuable outcome a map can give you, and the one people most want to ignore. A map that says "your coverage is fine" has just saved you the price of a device that would have changed nothing.

Building the map without walking it#

Step 1 of the walk test already has you drawing the floor plan on paper. Draw it in Range Up instead and the same ten minutes produces method 3 as well as method 1. You trace the rooms, set the material preset to whatever is behind the paint and nudge the wall loss if your partitions are heavier than that, and mark the router's real position; the predicted coverage comes back room by room, in the same dBm bands your grid uses, so the two maps sit next to each other in one vocabulary.

Then it does the thing a walk cannot. Name the room that failed, and the app weighs positions for a second radio against that room and hands you the map with and without one — the outlet comparison from the table above, without carrying a phone to each outlet twice. Bear in mind which column of that table it belongs in: it predicts, it does not measure, and it never touches your live network or reads the air. Walk the house afterward to check the prediction. The one-page version of what to do with a position is the placement method, start to finish, and the app is on the download page.