Kenya’s fixed internet market has recently entered one of those moments when headlines about speed and price start to obscure the deeper question: what actually makes internet good? The current wave of aggressive fibre pricing and higher-bandwidth plans from Safaricom, Airtel, Savanna Fibre, Zuku, Faiba, and other players has created the impression that faster and cheaper automatically means better. But that is only part of the story, and possibly not even the most important part.

In a conversation I had with a senior leader at one of Kenya’s leading internet service providers, the point that came through clearly was that bandwidth matters, but so do latency, contention, caching, device readiness, jitter, and the way people actually use the internet. In simple terms, that means speed is only one piece of the puzzle. How quickly data responds, how many other people are sharing the same connection, where content is stored, whether your devices can actually take advantage of the service, how steady the connection feels, and how your household uses the internet all matter too.

What the market is doing

The current market makes the pricing arms race impossible to ignore.

Safaricom’s latest plans now stretch from:

• Bronze: 40 megabits per second at Ksh 2,250.
• Silver: 60 megabits per second at Ksh 3,075.
• Gold: 150 megabits per second at Ksh 4,725.
• Diamond: 500 megabits per second at Ksh 9,375.
• Platinum: 1 gigabit per second at Ksh 20,000.

Savanna Fibre has pushed even harder on price and speed:

• 100 megabits per second at Ksh 2,000.
• 250 megabits per second at Ksh 4,500.
• 500 megabits per second at Ksh 6,000.
• 1 gigabit per second at Ksh 10,000.

Airtel’s XSteam Fibre, which recently launched, is also now part of the picture:

• 15 megabits per second at Ksh 1,999.
• 30 megabits per second at Ksh 2,999.
• 60 megabits per second at Ksh 3,999.
• 100 megabits per second at Ksh 4,999.

Zuku is also reinforcing its position with a broader mid-market offerings:

• 30 megabits per second at Ksh 2,799.
• 80 megabits per second at Ksh 3,799.
• 100 megabits per second at Ksh 4,399.
• 200 megabits per second at Ksh 9,999.

JTL (Faiba) is also showing that it wants to compete across both the mainstream and premium segments:

• 35 megabits per second at Ksh 3,000.
• 60 megabits per second at Ksh 4,000.
• 90 megabits per second at Ksh 5,000.
• 150 megabits per second at Ksh 10,000.
• 200 megabits per second at Ksh 15,000.
• 300 megabits per second at Ksh 20,000.
• 1 gigabit per second at Ksh 30,000.

That matters because it shows that Kenya’s fibre market is now developing multiple and highly competitive pricing options at once. Some players are fighting for the lower and middle tiers. Others are pushing premium and ultra-premium positioning. Some are trying to do both.

Why speed is not enough

The problem with the current debate is that speed has become the easiest metric to market, but not necessarily the best one to judge quality.

From an operator’s point of view, latency matters, especially for real-time use cases such as gaming, telemedicine, video calls, and interactive services. Latency simply means the delay between doing something online and getting a response back. So even if a connection is fast on paper, it can still feel slow if that delay is too high.

There is also the question of network density and power consumption. Faster next-generation networks require more base stations, which are the signal points that help carry internet and mobile traffic. More of those are needed when the network is expected to serve many users at high speed. That means the economics are not linear. As speeds rise, the cost of delivering and sustaining those speeds often rises too.

This is where the “faster is always better” story starts to fracture. A consumer may happily pay less for more bandwidth, but the operator still has to deliver that experience sustainably. If the market turns into a pure price war, the danger is not just shrinking margins. It is that operators will be forced to cut corners in areas customers only notice when things go wrong: congestion, support quality, uptime, latency, and service consistency.

The hidden infrastructure layer

There is another layer to this story that often gets glossed over: content delivery networks, local caching, and internet exchange points. These are not abstract technical details. They are part of what makes an internet connection feel fast, stable, and responsive in the real world.

A content delivery network or CDN is a system that stores content closer to users, so websites and videos load faster. Caching is similar: it means frequently used content is kept nearby instead of being fetched from far away every single time. An internet exchange point is a local meeting point where internet networks exchange traffic directly, which keeps data from taking a longer and slower route abroad.

A lot of what users experience as “fast internet” is actually the result of content being delivered locally, traffic being exchanged domestically, and video platforms adapting intelligently to available bandwidth. If traffic has to travel far unnecessarily, or if the local routing and peering environment is weak, then even a high-speed plan can feel disappointing. In that sense, the best network is not just the one with the biggest number on the flyer. It is the one that moves content efficiently and keeps delays low.

This matters because internet quality is a stack, not a single metric. Speed matters, but so do the delivery systems that sit behind it.

Shared versus dedicated

One of the least discussed factors in internet quality is whether the connection is shared or dedicated. That distinction matters because two plans with the same advertised speed can feel completely different depending on how the network is designed.

A shared connection means many users are using the same available capacity. That is why speeds can drop during busy times, especially in the evening when many people are streaming, downloading, or gaming. This effect is called contention, which simply means too many users are trying to use the same pipe at once. When contention is high, the connection can feel slower even if the plan on paper looks fast.

A dedicated internet access connection, on the other hand, is reserved for one customer, usually a business. It is more consistent because fewer people are competing for the same capacity. That is why companies that need reliable service often pay for dedicated internet rather than cheaper shared service.

This is an important distinction for businesses in particular. A company may not just need speed. It may need consistency, uptime, and lower jitter, which means less wobble or variation in delay. In simple terms, jitter is what makes video calls feel choppy or gaming feel unstable, even when the speed test looks fine. A home user may tolerate more variation than an enterprise customer, but even household users are increasingly sensitive to congestion when they are streaming, gaming, uploading files, or working from home. The real issue is not just what the plan says on paper. It is what kind of product it actually is.

What people actually do online

Another blind spot in the speed-first narrative is how people actually use the internet.

The operator perspective makes this clear. Fixed networks tend to support longer-form content, while mobile usage leans more toward short-form, high-frequency consumption. That means home fibre is often being used very differently from mobile data, even if the consumer talks about both in the same way.

Video now dominates usage. YouTube, TikTok, Facebook, WhatsApp, Instagram, Netflix, and gaming updates all eat bandwidth in different ways. In practice, many households are not just using the internet to browse. They are streaming, updating, transferring, calling, watching, sharing, and doing all of that across multiple devices at once.

That matters because many consumers do not actually know how much bandwidth they need. They know they want a better experience, but not necessarily what the best specification is. So when providers simply race to sell more megabits per second, they may be solving the wrong problem. The better question is whether the household gets a smoother, more reliable, more responsive experience for the applications it actually uses.

The hidden consumer issue

The market also needs to be honest about fair usage policy. A fair usage policy is the point at which an “unlimited” plan may start slowing down after heavy use.

If a plan is fast but slows sharply after a usage threshold, its value proposition may be weaker than a slower plan with better sustained performance. That is why customers need to think beyond the headline figure. What matters is not just the maximum advertised speed, but whether that speed holds up in a real household with streaming, gaming, uploads, cloud syncing, and multiple users.

The same is true for device readiness. If a home or small business does not have the right devices, a faster plan may not meaningfully improve the experience. The operator insight here is straightforward: connectivity only creates value when the devices exist and the behavior changes. That remains one of the most important truths in Kenya’s internet market today.

Where the market is heading

Safaricom is clearly defending its leadership. Savanna Fibre is trying to reset price expectations. Airtel is moving toward a more complete fixed broadband proposition. Zuku and Faiba are also visibly reinforcing their own positions, with Zuku leaning into a broader mass-market and upper-mid-tier ladder, and Faiba signaling both mainstream relevance and a serious premium ambition through its gigabit positioning.

The Communications Authority’s latest statistics show why this matters at scale. Kenya had 2,461,981 fixed internet subscriptions by December 2025, of which 1,378,198 were fibre subscriptions. Safaricom led with 858,394 subscriptions, followed by JTL at 494,150 and Wananchi at 272,802, while smaller players such as Ahadi and Vilcom still represented meaningful slices of the market. This is not a small niche. It is a large and growing market where pricing moves ripple across households, small businesses, and institutions.

But scale alone does not mean the market can sustain a permanent price war. That is the key business question. Aggressive pricing can drive customer acquisition and force rivals to respond, but it can also damage profitability if it becomes the default strategy. The broader point is that no one truly wins a destructive price war if the result is a weakened industry, thinner service quality, and underinvestment in the very networks customers depend on.

The bigger question

So do faster internet speeds and lower prices matter in Kenya? Yes, absolutely. But they are not the whole answer.

The real issue is whether operators can deliver a better overall experience: low latency, stable performance, sensible fair usage policies, good support, appropriate devices, clear shared-versus-dedicated options, and network architectures that match how people actually use the internet. The winners in this next phase will not just be the cheapest or fastest. They will be the providers that understand the whole user experience and the economics behind it.

That is the deeper lesson from Kenya’s new fixed internet moment. Speed gets attention, but experience keeps customers. In a fixed internet market as competitive as Kenya’s, that distinction is becoming more important by the week.