U Turn Road Sign

T-Mobile Continues To Walk Back On Forced Plan Migrations

In T-Mobile’s earnings call earlier today, CEO Mike Sievert reiterated that the company is walking away from forced plan migrations and associated price increases.

By the way, that was sort of not very accurately reported. So let me just kind of clear it up…we tend to do tests and pilots of things quite a bit to try to figure out what’s the right answer. In this case, we had a test cell to try to understand customer interest in and acceptance of migrating off old legacy rate plans to something that’s higher value for them and for us. And we had planned to test and did some training around that. And then it leaked. And it leaked as if it was a broad national thing, and it kind of wasn’t.

Now I don’t know that we still have to do that test cell because, to your point, we did get plenty of feedback thanks to the erroneous context of the leak. And I think we’ve learned that particular test cell isn’t something that our customers are going to love.

Now exactly none have rolled out. So even to your question that we recently rolled out, we didn’t. We had planned it. We had planned it as a test cell and then we aren’t doing it because I think we’ve got plenty of feedback.

Every company has bad ideas. T-Mobile put a stop to this bad idea before it adversely affected customers. Maybe that says something positive about T-Mobile. On the other hand, the suggestion that this was just a small test is frustratingly evasive.

A test was only going to take place because T-Mobile considered making forced plan migrations “a broad national thing.

Other Changes To Legacy Plans

Sievert later hinted that T-Mobile is still considering eventual adjustments to legacy plans:

We remain very interested in rationalizing our legacy rate plans for IT purposes, simplification purposes, revenue realization purposes, customer satisfaction and retention purposes. So we’re going to stay at it. But that particular idea is — we’ll probably do something different. Good. Okay.
That use of “rationalizing” confused me. Apparently, there’s a business meaning of the term that’s something like “adjusting products and reorganizing things to increase efficiency”.

If T-Mobile moves forward with changes to legacy plans, I hope they’ll go about it a bit more honestly.

Image of a finger pointing at a frowning face

T-Mobile’s Price Increase Saga

Earlier this month, a now-deleted post on Reddit and an article from The Mobile Report broke a story about an impending price increase for T-Mobile customers. Leaked documents suggested that T-Mobile would switch some customers on older plans to different plans with monthly prices $5 to $10 higher. Customers would be alerted of the upcoming changes and given an option to opt out of the automatic migration.

The documents included suggested lines for T-Mobile representatives fielding calls from customers. The lines include a gem that’s emblematic of the sort of bullshit consumers have to deal with in the cell phone industry:

We are not raising the price of any of our plans; we are moving you to a newer plan with more benefits at a different cost.

Raising prices of certain plans would violate a promise T-Mobile made not to raise prices for existing customers. T-Mobile appears to be weaseling out of its commitment by switching customers’ plans.

Most Customers Are Unaffected

Initially, it sounded like the automatic migrations would affect customers on the following plans:

  • Magenta
  • One
  • Magenta 55+
  • Simple Choice / Select Choice
  • Simple Choice Business

I’d guess tens of millions of customers are on those plans. However, documents released later suggested only 1% of T-Mobile’s customer base, about a million people, would be affected.

The leaks generated pushback and confusion. T-Mobile’s CEO, Mike Sievert, sent a company-wide email clarifying the situation. Sievert explained that reporting around the leaks missed context. Allegedly, price increases and plan switches were part of a small test. The email made me more sympathetic to T-Mobile, but it still reeked of bullshit. It didn’t acknowledge that the changes are, in practice, a price increase.

We hope our customers will be thrilled with the new benefits and service they will eventually receive…We continue to remain committed to being the Un-carrier.

Did T-Mobile Walk Back?

I’m unsure what to make of the whole saga. Maybe a price increase for a tiny segment of T-Mobile’s customers got blown out of proportion. Or perhaps T-Mobile planned to rollout the price increase more broadly but backed off after bad press. It’s hard to say. T-Mobile may scale up the plan migrations to a much larger portion of its customer base.

Image of a traffic jam on one side of a highway

Don’t Hide Throttling!

Cell phone carriers regularly impose limits to reign in heavy data users. A carrier may throttle a user’s data, deprioritize a user’s data, or both.

Throttling limits a user to a specific maximum speed (e.g., 256kbps). Deprioritizing lowers the priority of a user’s data transfer relative to other users on a network. If a network is not congested, lower priority has no effect. When a network is overburdened, deprioritized users experience slower speeds than other users on the network.

Carriers sometimes disclose their policies with statements along the lines of:

After [X]GB of data use in a single month, users may experience lower speeds.”
The disclosure is vague, but it’s an accurate description of deprioritization. It’s not a good description of throttling. If users are throttled, they will almost certainly experience lower speeds.

Boost Infinite’s Throttle

Boost Infinite throttles heavy users on its unlimited plans to 512kbps. Here’s the disclosure on the homepage:
Screenshot reading: "Members using >30GB/mo. may experience lower speeds."Boost Infinite is not the only carrier that has used language that describes deprioritization when it ought to be disclosing throttling. I’m partly picking on Boost Infinite because it’s the most recent example I’ve encountered. But Boost Infinite further frustrated me while communicating about its policy on Twitter X.

The most bothersome message from the Boost Infinite account has since been deleted. Here’s roughly how I remember it:

Boost Infinite subscribers can use unlimited data.

I like to think of it like traffic. You can travel however far you want on the highway, but if there are lots of other people, it will take a bit longer to get to your destination.

Maybe there’s an analogy to be made between prioritization on cell networks and traffic during a long journey. If you have bad traffic on the road, parts of your journey will be at roughly half the usual speed, some moments you’ll be able to drive at normal speeds, and at times you’ll be caught bumper-to-bumper in traffic jams.

Throttling is different. If your car is forced to stay under 1 mile per hour, there’s an explicit upper bound on how far you can travel in a month. You’re probably not going to bother with a trip from New York City to San Francisco.

An Exception?

While writing this post, Joe Paonessa of BestMVNO.com alerted me to reports of Boost Infinite subscribers not being throttled when accessing Dish’s network.

Boost Infinite’s service almost always piggybacks on AT&T or T-Mobile’s network. However, Boost is owned by Dish. A fraction of customers using certain devices and living in particular markets can sometimes access Dish’s nascent network.

Across Boost Infinite’s customer base, I expect subscribers are connected to Dish’s network less than 1% of the time. However, there’s technically a possibility of avoiding throttling. Perhaps that makes Boost Infinite’s “may reduce speeds” language more defensible.

But we shouldn’t have ended up in this scenario in the first place. Carriers ought to clearly disclose their throttling and congestion-management policies. Consumers should be able to make sense of their cell phone plans’ policies without connecting the dots between random Reddit posts, reports on Twitter, and posts like this one.