T-Mobile/Sprint is Decommissioning MPLS - Why & What Next?

Cover photo image source: T-Mobile/ Sprint

We recently heard from our customers that T-Mobile plans to sunset its Multiprotocol Label Switching (MPLS) services. 

Chances are that you know T-Mobile as the bright pink, consumer wireless provider, so hearing “T-Mobile” and a wide area networking solution in the same sentence might have you scratching your head. 

T-Mobile entered the wireline services industry when it completed its merger with Sprint back in 2020. While Sprint’s wireless subscriber base is what made them an attractive M&A target, at one point Sprint was considered an elite MPLS and wireline connectivity provider as well. Post-merger, it was possible to procure MPLS services through T-Mobile that were provisioned on the legacy Sprint backbone (you could do this even as recently as a few months ago). 

But now, T-Mobile/Sprint’s move to decommission its MPLS services is forcing customers to make big decisions regarding their wide area network design: “rip and replace” your T-Mobile/Sprint MPLS solution with a new provider or transition your MPLS wide area network to a new technology such as software defined WAN (SD-WAN). 

This post discusses T-Mobile/Sprint’s decision to decommission its MPLS service and walks through the options available to customers who are impacted by this change. 

T-Mobile / Sprint’s MPLS Decommissioning

Why?

To reiterate, T-Mobile/Sprint have made no formal announcements regarding the MPLS decommissioning. However, we’ve heard from multiple customers that have received formal notice of the MPLS decommissioning (we also started this Reddit thread to investigate further). So, here we discuss our opinion of why they would do this. 

As mentioned in the introduction, T-Mobile was attracted to Sprint primarily for their wireless customer base as well as the radio spectrum that they own rights to. 

Sprint had been winding down its wireline segment in the years leading up to the merger; wireline services made up just ~5% of its total service revenue in FY2019 and this business segment operated at a loss in the two years prior to the T-Mobile/Sprint merger. 

T-Mobile was not a wireline provider before the merger, and in 2021 (post-merger) wireline services made up just ~1% of T-Mobile’s service revenue. 

All of this to say, the decision to decommission the T-Mobile/Sprint MPLS services was made so the company could remove the associated overhead costs that go into being an MPLS operator while also allowing the business to focus on its core competency: wireless. They have yet to release any official statements regarding this decommissioning, but this is our take on what we’ve heard so far. 

Is this foreshadowing of what’s to come?

As cloud centric networks become the norm, MPLS usage has lost market share to new technologies. So, the next question we pose is if T-Mobile’s decision is potentially indicative of things to come from other major MPLS/wireline providers?

In our opinion - no. 

MPLS still has a place in enterprise networks today. It is a fully managed, dedicated service that provides firm service level agreements (SLAs) for availability, jitter, packet loss and latency. 

Since MPLS is carrier managed, it also means a single throat to choke (or back to pat) in terms of network management and billing for users; you need to make a change to your MPLS configuration, call your provider; you need to add a location, call your provider. This convenience often comes at a significant cost, but it is unarguably operationally efficient for medium to large enterprise networks. 

So while most MPLS carriers are potentially feeling a lag in demand for their service, the carriers also know that MPLS remains a critical network solution for many enterprises. 

If you work with a legacy, big-time wireline provider today for your MPLS - such as Verizon, AT&T, Lumen, Telefonica, or Orange - we don’t see your MPLS service being decommissioned anytime soon. 

What’s Next?

Over the last 30 days, we have spoken to customers who have as few as 30 locations and as many as 3,000 that are being impacted by T-Mobile’s MPLS decommissioning. 

The big question is what is next for these companies? That answer is simply, “it depends”.

Ultimately there are not a lot of options, so let's take a minute and look at two of the most likely solutions. 

Option 1: Rip and replace T-Mobile’s MPLS with an alternative MPLS provider

There are a multitude of carriers who can provide MPLS, and if your edge sites are already fiber fed by a LEC or CableCo (aka “on-net”) you may be surprised at how easily a replacement MPLS solution could be implemented. However, the chances that a second provider is 100% on-net to replace your T-Mobile/Sprint solution are very slim.

The next best solution would be to find a provider who is partially on-net who can deploy Type 2 connections at the locations where they are off-net (also known as circuit aggregation). However, we hear too often that this process does not go very smoothly with U.S. based LECs. So, even in a partially on-net situation, the “rip and replace” option can still be a cumbersome and expensive process to complete. 

If your network is majority off-net with any MPLS partners (aside from T-Mobile/Sprint), then “rip and replace” becomes a “rip and rebuild”. Depending on your geography and your new provider’s willingness to build - this path could become cost prohibitive very quickly. 

Whether you find yourself in a “simple” rip and replace or a “rip and rebuild” position, now is a good time to evaluate if MPLS is the optimal solution for your network. Take a read through the pros and cons of MPLS here if you need help evaluating that decision.   

Option 2: Transition from MPLS to a SD-WAN. 

If you currently utilize the Sprint/T-Mobile MPLS service and have been considering making the switch to SD-WAN, now is the best time to kick off your network transition. 

When well planned and architected, SD-WAN can accomplish almost anything that MPLS can, including predictable routing of performance sensitive traffic whether it be VoIP, video, ERP (or most anything else). Additionally, SD-WAN solutions add lots of insight and reporting that MPLS deployments simply do not offer. Finally, SD-WAN allows for centralized orchestration of WAN and Internet traffic, with tuning available on an individual application level. SD-WAN does this across a provider agnostic underlay network that, in many cases, not only increases network availability and bandwidth, but also reduces cost. 

Life After Sprint’s MPLS 

For most enterprises, SD-WAN is a flexible, economical and forward looking network solution -  but it requires effort and understanding to do it right. We can’t cover all of the nuances here, but you can read further in this guide to transitioning from MPLS to SD-WAN. 

If you’d rather speak with a Lightyear expert on the topic, reach out to us here. Lightyear’s free software has helped hundreds of customers navigate network design and procurement, saving them time and money in the process. 

Considering other WAN topologies or technologies? Check out this guide to designing the optimal WAN.