You walk into your apartment and the bars drop from four to one. A video that uploaded fine on the sidewalk now sits at 12% for a minute. Most of us blame the building: thick walls, energy-efficient windows, the elevator shaft next door. A new study from Ookla, the company behind Speedtest, says that’s mostly the wrong suspect.
Ookla looked at a full year of in-building signal readings, from September 2025 through August 2026, across more than 100 large cities in 60 countries. The headline finding is uncomfortable for American carriers. North America sits at the bottom of the global table for indoor 4G signal, next to Oceania, and behind not only East Asia and the Gulf but also Africa and Latin America. And the reason isn’t really your walls. It’s the network outside them.
What Ookla actually measured
The study, led by Ookla lead industry analyst Luke Kehoe, uses signal readings that phones report while people run Speedtest and other background measurements inside buildings. Ookla matched those readings to individual buildings, then compared the median indoor 4G signal city by city. It also compared each building against the outdoor signal around it, which is what lets it split the blame between the network and the bricks.
Signal here is measured in dBm, or decibel-milliwatts. The numbers are negative, and closer to zero is better. The scale is logarithmic, which is the part people miss: every 10 dB is a tenfold change in power. So when Ookla says the best cities sit around -88 dBm and the worst around -105 dBm, that 17 dB gap means the phone in Naples is working with roughly one fiftieth of the signal power a phone in Dubai gets.
The US numbers are worse than the averages suggest
A regional median of -101 dBm doesn’t sound dramatic until you look at how many buildings fall off the cliff. Ookla uses -105 dBm as the line where indoor 4G service becomes unreliable. That’s the zone where calls start to break up, pages stall, and your phone burns battery hunting for a better cell.
Here’s the share of measured buildings below that line in the cities Ookla called out:
Read that again. Roughly two in five buildings in Miami, Seattle and Los Angeles have indoor signal weak enough to be unreliable. New York, the best North American city on this measure, still has a problem in one building out of four. Seoul has it in one out of fifty.
This matters for speed, too, not just bars. Ookla’s global Speedtest data shows that between -90 and -110 dBm, median 4G download speed falls by 65% and upload speed by 87%. That upload number explains a lot of everyday frustration. Sending a video, backing up photos, or joining a video call from your kitchen hits the weakest part of the link first.
So why aren’t the walls to blame?
Building materials do matter. Ookla just found they matter a lot less than most of us assume. In the median city, moving indoors costs about 4 dB of signal. Concrete and stone buildings cost around 5 to 7 dB. In Tokyo, where there’s a nice mix of both, timber-framed buildings knocked off about 2 dB and concrete ones about 7 dB.
Here is the twist for American readers. Phones in North American and Oceanian cities lose the least signal on the way inside, about 2 dB. Lots of US housing is wood frame and drywall, which radio waves pass through fairly easily. Ookla gives Phoenix as an example: it loses noticeably less signal going indoors than Seoul does. Seoul still wins easily, because its outdoor network is so much stronger to begin with.
| Factor | What Ookla found | What it means for you |
|---|---|---|
| Outdoor signal | Explains about 85% of the gap between cities | Weak bars indoors usually mean a weak network nearby |
| Building entry loss | Under a tenth of the gap; median 4 dB | New windows or concrete make it worse, but rarely cause it alone |
| US entry loss | About 2 dB, among the lowest anywhere | Wood-frame homes aren’t the villain |
| Stations and stadiums | About 4 dB stronger than nearby buildings | Dedicated indoor systems work |
| Hospitals and hotels | About 1 dB weaker than nearby buildings | Big, dense buildings without their own gear struggle |
| Low-band spectrum | Phones lean more on sub-1GHz indoors in 97% of cities | Indoors, you’re often on slower, longer-range bands |
That last row is worth a quick explanation. Low frequencies, like T-Mobile’s 600MHz or the 700MHz and 850MHz bands AT&T and Verizon use, travel farther and push through walls better than the mid-band spectrum that carries most fast 5G. When you step inside, your phone often falls back to those low bands. You keep a connection, but there’s less capacity to go around, which is part of why speeds collapse even when you still technically have service.
The buildings that do it right
The most useful finding in the report is about in-building systems. Large buildings served by their own cells on at least two carriers’ networks had stronger indoor signal than the rest of the city’s large buildings in 75 of the 76 cities Ookla tested. That pattern points to shared distributed antenna systems (DAS) or dedicated small cells.
The median improvement was 3.7 dB. In North America, though, it was about 6.5 dB, more than three times the roughly 2 dB gain in East Asia. That makes sense. When the outdoor network is already strong, there’s less to fix. When it’s weak, like in most US cities, indoor gear does a lot of heavy lifting.
Rules, not just radio
Why do Seoul, Hong Kong and Singapore do so well? Kehoe points to regulation. Those cities, along with Riyadh and Jeddah, sit under rules that write indoor connectivity into how buildings are planned. Hong Kong reserves space for mobile equipment in new large buildings. South Korea and Singapore require room for network gear and actively measure indoor coverage. Saudi Arabia mandates shared indoor systems.
Switzerland shows the opposite effect. It has some of the world’s strictest caps on radio exposure, 4 to 6 V/m at places of “sensitive use” like homes and schools. Ookla says Zurich and Geneva both land among the seven weakest cities for indoor signal. Italian cities, which also have tight exposure limits, average -102 dBm, against -99 dBm for the rest of Europe.
The US has a close cousin of the Asian rules, but it isn’t aimed at you. Model fire codes require many new buildings to provide radio coverage for first responders, so a firefighter’s radio works in a stairwell. There’s no matching requirement for ordinary cellphone coverage. Carriers decide where to put indoor systems based on whether a building is worth the money, which is why your stadium has great signal and your dentist’s office doesn’t.
How to check your real indoor signal
Bars are a rough guide that every phone maker draws differently. If you want the actual number, both iPhone and Android will show it.
iPhone: open the Phone app and dial *3001#12345#*, then tap Call. In Field Test mode, look for RSRP under the serving cell info.
Samsung Galaxy: Settings > About phone > Status information > SIM card status > Signal strength.
Pixel and most other Android phones: Settings > About phone > SIM status > Signal strength.
| 4G signal (RSRP) | Rough rating | What to expect |
|---|---|---|
| -80 dBm or better | Excellent | Full speed, no worries |
| -80 to -90 dBm | Good | Where Seoul and Dubai sit indoors |
| -90 to -100 dBm | Fair | Works, but speeds start to slide |
| -100 to -105 dBm | Weak | The typical North American building |
| Below -105 dBm | Unreliable | Dropped calls, stalled uploads, battery drain |
What you can actually do about it
You can’t rezone your city, but you have more options than standing by the window.
- Turn on Wi-Fi Calling. AT&T, Verizon and T-Mobile all support it on current phones. Calls and texts route over your home broadband, and your phone switches back to cellular when you leave. On iPhone it lives in Settings > Cellular > Wi-Fi Calling. On Android, check the Phone app’s settings or your network settings.
- Ask your carrier about a network extender. Some carriers still offer small cells that plug into your internet connection and create a mini cell inside your home. Availability varies, so ask before you buy anything else.
- Consider a signal booster, carefully. Consumer boosters grab the outdoor signal with a roof or window antenna and rebroadcast it inside. They need some usable signal outside to work with, and FCC rules require you to register one with your carrier. Given Ookla’s finding that the outdoor signal is the real problem in US cities, test outside first.
- Check which carrier is strongest where you live. Coverage maps show outdoor signal. Ask neighbors, or try a prepaid eSIM from a rival network for a month before switching for good.
- A newer phone can help a bit. Modems keep getting better at holding weak signals. Early tests suggest Apple’s C2 modem in the iPhone 18 Pro does its best work in exactly these fringe conditions. Still, a better modem gains you a few dB at most. It can’t turn -110 into -88.
One thing that won’t fix your living room: satellite-to-phone service. Direct-to-cell systems like T-Mobile’s T-Satellite, and the network SpaceX is building for Starlink Mobile, need a clear view of the sky. They’re a lifeline on a hiking trail, not an answer to a weak signal inside an apartment block.
The bigger picture
US carriers love to talk about 5G coverage maps and median download speeds, and those numbers have genuinely improved. But almost all of those maps describe outdoor coverage, and we spend most of our time indoors. Ookla’s study is a useful reminder that a network can look great on paper and still leave a third of a city’s buildings with shaky service.
There are two clear fixes, and neither is cheap. Carriers can build denser outdoor networks, with more sites and better use of spectrum. Or they, and building owners, can invest in indoor systems, which deliver a bigger gain in North America than almost anywhere else. Cities that wrote indoor coverage into their building rules did the best. Until something like that happens in the US, Wi-Fi Calling is going to keep doing a lot of quiet work in American homes.
About this article: GeekBlog covers U.S. technology news, AI, phones, smartwatches and gaming. Every story is written and checked under our Editorial Policy. Spotted a mistake or have a story tip? Contact our editors.

