Turkey has transitioned to 5G, but not with speed—with errors

I watched a short video on the internet. The man in the video was saying: “A 10 GB data package in the UK does not last as long as a 10 GB package in Turkey.” It is a simple sentence, but it points to a very important difference. The same 10 GB lasts for weeks in one country, while it runs out in a few days in the other. So, why?

The issue is not speed, but latency

5G does not mean speed; it means “latency.” The real promise of this technology is that devices can respond to each other almost instantaneously with millisecond delays. For this to happen, the infrastructure must be extremely clean, interference-free, and low-loss. In Turkey, however, the picture is the exact opposite.

QoS (Quality of Service) is low; a significant portion of the data is lost to interference. Packet losses, routing errors, base station congestion… All of these reduce connection efficiency. In other words, the 10 GB in Turkey is never actually used as 10 GB. The data dies on the way. That is why the same package lasts a month in the UK, while it runs out in a few days here.

There is no latency without fiber

The low latency promised by 5G is only possible with fiber infrastructure. This is because the data leaving the base station must be transported to the backbone via fiber. In Turkey, the fiber access rate is still low, and the pace of investment is slow. If there is no fiber, the biggest advantage of 5G, which is low latency, is already impossible.

In short, 5G means latency, not speed—and that cannot be achieved with these interference-prone, weak lines.

IP sharing at base stations

For years, operators have been using CGNAT systems that share a single IP among multiple users at base stations to cut costs. This not only slows down data traffic but also blurs connection identities. It creates session errors in applications, delays in streaming services, and even confusion in security systems.

In reality, this method blocks the low-latency communication that is at the core of 5G. Because network traffic cannot be separated at the IP level, QoS management is compromised. In other words, the signal you see on a phone with a 5G logo is operating at 4G quality.

And yes, there are some things I know about this IP sharing, but I will not write them here. If I get into it, a lot of people will get hurt.

BTK: Jumping into technology without solving problems

The structural problems that the BTK (Information and Communication Technologies Authority) has failed to solve to this day are still there. Domain name management is still buried in bureaucracy. The domain name authority taken from METU (ODTÜ) has been operated for years with an inefficient, paperwork-heavy system that is behind the times. The “first come, first served” logic is still not applied. It is ironic that an institution that talks about digitalization is still carrying folders in its digital identity system.

Without a change in this mentality, advanced technologies like 5G will only be window dressing. Because before adopting a technology, you must establish the institutional culture to support it.

Tender or sharing?

In the approximately 3 billion dollar 5G tender, the highest frequency went to Turkcell, followed by Türk Telekom and Vodafone. Equal shares were given in the 700 MHz band, while Turkcell and Türk Telekom gained an advantage on the 3.5 GHz side. In short, it was a classic “brotherly share” arrangement where everyone got their cut and no one was upset.

However, selling frequencies is not the same as building infrastructure. Fiber is still missing, IP sharing is still widespread, and QoS is still low. Under these conditions, 5G will just be an expensive label.

Let's move on to 6G already

5G makes no sense with this infrastructure anyway. This matter is finished. The era of artificial intelligence has now begun in the world. The era of applications is over; the era of algorithms has begun. We are in an era where smart devices, sensors, production lines, and artificial intelligence systems will communicate.

In this new era, response time is more important than speed, and intelligence density is more important than capacity. This requires a much lower-latency and much more powerful infrastructure that 5G cannot provide.

Therefore, we need to move on to 6G now.

Because this time, the issue is not just about installing base stations. It is about redesigning the backbone and connecting it to space.

In the world of artificial intelligence, no infrastructure that does not move fast will be able to survive. While we are still trying to lay fiber, the world is talking about network architectures integrated with satellites.

Conclusion

The 5G tender is over, but Turkey's digital transformation has not begun.

• Fiber is still not widespread.

• QoS and interference problems persist.

• IP sharing is killing performance.

• BTK is still managed with the reflexes of the 2000s.

As a result, Turkey has transitioned to 5G, but it is moving forward on 4G infrastructure with a 3G management mindset.

My personal notes

First, Turkey should have solved the fiber infrastructure before transitioning to 5G. If there is no fiber, there is no latency.

Second, 5G means interference-free communication, not speed. And that is impossible with this network structure.

Third, IP sharing at base stations is the most invisible but most critical bottleneck of the system.

And finally: The 10 GB in the UK works, while our 10 GB runs out. Because we still cannot transport data. Let's move on to 6G now, because it is time to save the signals lost on the ground in space.