Skip to content Portal

The Difference Between Tier 1 and Tier 2 ISPs

The three tiers of Internet Service Providers serve as a foundational concept within Internet connectivity. In this article, we’ll help you understand the three types of Internet connections, the three tiers of ISPs (plus one specialized Internet provider), and explain the difference between Tier 1 ISPs and Tier 2 ISPs, particularly for IP Transit buyers. 

Three Types of Internet Connections 

There are two types of Internet Protocol (IP) connections that you can procure from an ISP:  

1. Internet access is a connection that enables end users to consume applications, content, and data from the Internet. Internet access is purchased by consumers (home broadband and mobile Internet subscribers). Businesses also purchase Internet access on behalf of their employees. Internet access services are often delivered to residential or business locations such as offices, factories, and warehouses.

2. IP Transit is a connection that enables an organization to provide applications, content, and data as Internet traffic to end users across the Internet. When an organization purchases IP Transit as its network access to Internet bandwidth, it must register for an Autonomous System Number (ASN). When it does so, the organization is known as an Autonomous System (AS). Just so you know, Autonomous Systems are not limited to ISPs. However, most ASNs provide services via the Internet, such as cloud providers, hosting providers, content providers, e-commerce companies, SaaS providers, gaming companies, etc. 

All Autonomous Systems must use the Border Gateway Protocol (BGP) for routing between their computer networks and the Internet and announce their IP network address blocks to the Internet associated with its ASN. BGP routing communicates the routes between all ASNs and the IP address blocks their networks manage. BGP routing not only allows ASNs. to announce their addresses but also to receive the complete set of routes to the entire Internet. 

In addition to the two Internet connection types above that are offered to consumers and all businesses, there is a third type that is only used by Autonomous Systems: 

3. Internet Peering is a physical interconnection and a BGP routing relationship between two Autonomous Systems that want to exchange traffic. ASNs can set up peering agreements that are commercial or non-commercial. Non-commercial peering, or settlement-free peering, implies that no more traffic is exchanged than can be done as an equal trade. Internet peering relationships are most commonly physically arranged in Internet Exchange Points, shared networking facilities in data center locations where ISPs and other Autonomous Systems run portions of their networks, including their Internet routers. In some cases, Internet peering relationships are set up on a bilateral basis between ASNs as a Private Network Interconnect (PNI).  

 Note that a common terminology confusion occurs when talking about “peering” because this term is also used for a type of Internet connection.

Three Tiers of ISPs Plus One Specialized Internet Provider 

Tier 3 ISPs 

To ease understanding, we’ll start with the simplest type of ISP—Tier 3 ISPs. Strictly speaking, the definition of a Tier 3 ISP is that they must purchase IP Transit services to send traffic to the rest of the Internet. Tier 3 ISPs generally only sell IP access to end-user businesses and consumers. Rural telecoms, local Internet companies, and small managed IT services companies operate as Tier 3 ISPs. 

Tier 1 vs Tier 2 ISPs 

When defining the difference between Tier 1 and Tier 2 ISPs, it is essential to begin by understanding that by definition, both Tier 1s and Tier 2s own and operate networks that connect other Autonomous Systems to the Internet. There is a term for this—an ISP’s “AS Cone”, meaning how many AS are connected to the Internet via the ISP. You can see the numerical metric and ranking for AS Cone on websites like This means that Tier 1s and Tier 2s operate as part of the “backbone” of the Internet. 

No ISP has a network that physically reaches everywhere in the world. So Tier 1s and Tier 2s must use other providers’ networks to get access to the rest of the Internet beyond their network boundaries. The technical difference is in how it is done. 

Tier 2 ISPs are technically defined as ISPs that gain access to the rest of the Internet through a combination of peering relationships and paying for IP Transit services from Tier 1 networks and other Tier 2 Providers. Tier 2 providers include many prominent national telecommunications and multi-national carriers, including Comcast, Cox, Charter Communications, China Telecom, British Telecom, Vodafone, Virgin Media, and Softbank Japan. 

Tier 1 ISPs are technically defined as ISPs that gain access to the rest of the Internet solely through settlement-free peering relationships with other Tier 1 ISPs. Examples of Tier 1 ISPs include AT&T, Verizon, NTT, GTT, Cogent, Tata Communications, Deutsche Telekom, Arelion (formerly Telia), and Lumen (previously Level 3 and Centurylink). 

Note that beyond using peering or transit, every new downstream IP transit customer adds to the reach of a Tier 1 or Tier 2 ISP’s network. 

Internet Exchange Providers are Specialized ISPs 

Internet Exchange Providers are another type of Internet provider that doesn’t belong to a tier. Remember how we explained that peering relationships are generally set up in shared facilities called Internet Exchange Points? Well, the industry decided in all its wisdom to use nearly the same term and precisely the same acronym (IXP) to stand for two different things. In short, the business of an Internet Exchange Provider is to build and facilitate Internet Exchange Points on a commercial basis. As a result, IX Providers only offer services to Autonomous Systems. 

How Tier 1 vs Tier 2 ISPs Engineer Their Networks 

One of the ways that Tier 1s differ from Tier 2s is in how they make network engineering decisions. The result is that Tier 2 networks often perform better for customers. 

Every network provider must engineer its network to deliver traffic with sufficiently good performance to satisfy customer requirements. Of course, sufficient doesn’t necessarily mean excellent. In any network, traffic can be deliverable while suffering latency, packet loss, and jitter issues. Fundamentally, the shorter the paths and the more bandwidth capacity available in a network, the better traffic performance will be. 

Tier 1 network engineering is heavily influenced by brand marketing rather than just performance goals. The reason is that the very definition of Tier 1 ISP means they can’t buy transit from any other provider.  

As noted above, only ISPs that own high-value audiences or routes can gain this position, but once they do, they must adhere to this “marketing” rule to stay in the Tier 1 branding club. This branding club has a lot of value in the market, so it’s important to them. 

However, the imperative of only doing settlement-free peering comes into conflict with trying to engineer optimal paths or provide sufficient bandwidth for those paths. In many cases, when Tier 1 ISPs establish peering relationships with one another, there is an imbalance of traffic flowing between their networks at a given interconnection. If there were no Tier 1 branding considerations, one of the ISPs might make up for the imbalance by purchasing transit from the other to ensure that its customers can get all their traffic delivered with strong performance. However, to preserve its Tier 1 branding, the ISP can’t do that, so it ends up with less bandwidth than it needs in that peering relationship. As a result, customers experience packet loss in their traffic delivery. This is more common than you might expect.  

For Tier 1 ISPs, because they can use their branding to foster trust in their services despite not always giving the best performance, this scenario works for them.  

Tier 2 ISPs don’t possess Tier 1 brand equity. But they aren’t constrained by Tier 1 branding considerations on how they engineer their network. As a result, Tier 2s tend to engineer for consistently high performance because this is their value-added differentiation. 

Tier 2s are Leading Innovation in IP Transit with Cloud Models 

Another way Tier 2s are increasingly different from Tier 1s is in the pace of innovation. The new category of Network-as-a-Service (NaaS) is where network service providers adopt cloud models that incorporate: 

  • Self-service portals 
  • Open REST APIs 
  • Transparent pricing 
  • On-demand provisioning functionality 

This new wave of innovation is difficult for Tier 1s to emulate due to their long legacy practices and traditional business models.  

While most NaaS providers have focused on private layer 2 services, getting IP Transit on demand from a NaaS provider is now possible.  


Whether you’re a content provider, hosting provider, or Tier 3 ISP, smart buying of IP Transit ensures the best service quality to your users. 

Learn more about our highly ranked network ( AS 5405), IP Transit, and IP Access services, as well as integrated DDoS protection and multi-service support for Flex Ethernet connectivity. You can check out our transparent pricing and order/provision via our self-service portal.