Years later, has 5G really delivered on its biggest promises?

Depending on where you live, you’ve likely had 5G in your pocket for at least a couple of years — or possibly close to half a decade. In any case, the wireless tech has certainly been around long enough to have had time to accomplish the numerous lofty promises that CEOs piped up to upsell us, which included everything from rejuvenating retail to traffic lights pushing updates to your car.

While some of those promises might have come to pass, quite a lot of them have obviously not. I don’t know about you, but my city’s traffic lights aren’t any more in sync than they were a decade ago. But let’s answer the big question: Now that we’re several years into 5G’s global rollout, has it come close to living up to the hype?

For you and me, 5G promised three major improvements: gigabit speeds boasting 10x or greater enhancements over 4G, latency as low as 1ms, and more reliable coverage.

Let’s start with that first point, which was always overly reliant on the promise of expensive, short-range mmWave spectrum. This technology has mostly seen deployment in dense urban hotspots — stadiums, airports, and downtowns — in a few countries like China and the US, but the rest of the world has largely ignored it due to its cost and poor signal range. Even in the US, only Verizon remains highly committed to the technology, with AT&T and T-Mobile preferring localized deployments. Instead, much of the world’s 5G network coverage is based on repurposed 4G spectrum and select 6-GHz bands.

So how does the data compare to the hype?

Ookla’s US speed test report between July and December 2024 shows a huge discrepancy in data rates across the country. In many states, the median download speed seen by consumers remains around the 50Mbps data range, and even in the best case, consumers typically experience sub-200Mbps speeds. That’s serviceable for light work on the go, but falls well shy of the gigabit potential we were promised (although consumer speeds were never going to sustain that level for any length of time). Equally, latency remains comparatively high for our gadgets, averaging 63ms nationwide, fine for browsing but too high for real-time applications like cloud gaming or AR/VR.

The situation is equally varied when we look at the global picture. Western consumers are sitting somewhere between 100Mbps and 300Mbps for typical 5G download speeds, but that’s a very broad range. Once again, typical speeds vary quite widely, with the worst markets scoring hardly much faster than a reasonable 4G at under 100Mbps and the fastest well over 300Mbps, which challenges a decent fibre package. South Korea’s operators score over 400Mbps for 5G downloads, for example.

According to the GSA’s June 2025 report, upload speeds are substantially slower, between 20Mbps and 50Mbps. However, that’s still a step up from the 5Mbps to 15Mbps range common on 4G LTE networks.

Taking a step back, the report reveals that 5G speeds are around 3x faster than 4G LTE on the same networks. That’s obviously a solid benefit, but again, it’s nowhere near the 10x speed boosts marketing at the start of 5G’s deployment, partly because mmWave reaches so few customers. In addition, looking at 5G in isolation is rather generous — you certainly won’t be connected to 5G all of the time, so typical speeds will be slower. Given that some 4G LTE-A networks were already capable of hitting 100Mbps or more, a lot of 5G’s benefits could also be quite heavily exaggerated by this data.

Considering the positives, carriers have some seemingly decent coverage numbers to boast about these days, which you’d hope, given that we are years into the rollout. In the US, at least 75% of customers are now covered with 5G, with similar figures of around 80% in many European countries as well. Other regions range from 60-90%, depending on how early they started their deployment. As you’ve probably experienced, urban areas have the best 5G coverage, while rural areas often remain more limited.

However, coverage doesn’t always mean connected. You might dip in and out of cell areas on your commute, or might not actually receive a 5G signal when sitting in your home.

Rather than measuring geographical reach, Opensignal’s 5G Availability metric records time spent with an active 5G connection, and that number is obviously much lower than area, and varies greatly depending on your country and network. Puerto Rico leads the charts with 57.4% typical 5G connectivity time, followed by India (54.3%), Kuwait (45.6%), Singapore (40%), and the US (37.5%) in fifth, based on data collected between January and March 2025.  This is likely due to the fact that many suburban and rural areas still lack strong mid-band 5G to boost signal coverage.

Even if we give 5G the benefit of the doubt in terms of speed, latency, and geographical area, US consumers typically spend just over 1/3 of their time connected to a 5G network — hardly a convincing case that the technology is a major game changer for our daily lives. EU countries fare even worse.

So while everyday mobile performance is a mixed bag, what about the futuristic use cases 5G was supposed to unlock?”

If consumer coverage and speeds are a little disappointing, the overhyped futuristic applications promised by 5G remain aspirational, to put it politely. Outside of flashy demos, things like remote robotic surgery and fully autonomous smart cities were always more pipe dream than near-term reality.

This is partly because many networks are still Non-Standalone 5G (NSA). NSA relies on 4G infrastructure for core functions, while Standalone (SA) is built from the ground up to enable features like network slicing and ultra-low latency. According to Téral Research, just 74 out of 354 global public 5G networks have migrated to SA — a rather disappointing 21%. Additionally, Omdia and Ookla note that European carriers are notably behind the US, scoring just 2% against 24% in 5G SA availability. As of mid-2025, 163 further operators are investing in SA, but until full Standalone 5G is widely deployed, many advanced features — like ultra-low latency and true network slicing — will remain on the sidelines. Many of the initial 5G promises are also being batted around again with 5G Advanced — only time will tell.

Instead, a few private enterprises have deployed their own private 5G networks for specific use cases, such as security systems and warehouse robotics. Other sectors like ports, airports, and manufacturing plants have adopted private 5G networks for logistics, asset tracking, and low-latency automation. While still limited in scope, these deployments hint at future potential — particularly if costs fall and software ecosystems mature. Notably, nearly 90% of tightly controlled private 5G deployments are based on Standalone.

Skepticism aside, some new 5G-powered capabilities have made their way into the consumer space. Fixed Wireless Access for home use is well established for some consumers in the US and abroad, giving access to reasonably quick data speeds in lieu of often more costly cable setups. However, data speeds and reliability are just as variable as mobile packages. Satellite connectivity in the Pixel, iPhone, and such is only possible because of 5G-NTN, though that will cost consumers extra to use in the long term.

However, these 5G benefits have come at a considerable cost to carriers. In addition to the raw equipment costs, carriers have spent a fortune on new spectrum. AT&T spent some $27 billion in the US C-band action, and Verizon a staggering $55 billion. Yet in October 2024, OpenSignal noted that “5G hasn’t yet had a substantial impact on operators’ profitability.” Partly because plan prices haven’t risen in line with the costs, which is good for consumers, but equally because business use cases haven’t really manifested. Carriers are now prioritizing network efficiency to keep investment costs under control.

Overall, the rollout of 5G has had some benefits, but those perks are certainly limited when compared to the initial promises and costs.

Mobile broadband is undeniably faster than the 4G era, and even though around 3x the speed is far from what was often marketed, this has helped push data speeds from dubious to far more usable. Latency is better too, ensuring activities like streaming and browsing feel noticeably smoother, but it’s nowhere near low enough compared to fibre for consumers to contemplate real-time applications.

All of this has to be carefully weighed against the costs, in terms of carrier investments, data plan costs, and hardware prices. Data plans certainly haven’t become much cheaper, and, if anything, premium 5G tariffs remain expensive. Likewise, 5G modem, radio (particularly for mmWave), and chipset prices continue to rise, in part due to the costs of implementing the complex array for 5G networking brands and technologies. It’s one of the reasons why modern flagship phones are just so pricey.

Of course, the promised future of mass IoT and automation certainly hasn’t appeared. We’ll have to see if those promises ever materialize before 6G networks roll around. But when it comes to 5G, I imagine I’m not alone in feeling that the next-gen network is fine but has hardly been worth the hype, and the data proves it.