Blog

Consumers Need More Information On Congestion & Prioritization Policies

Christian

Wireless networks have finite capacities. If enough users try to send data over a network at the same time, the network will become congested and deliver slower speeds.

Not all users will see the same decreases in speeds during congestion. Prioritization policies govern how different people on a network are affected by congestion. In many cases, subscribers with premium service plans will be prioritized ahead of subscribers on low-cost plans. When low-priority subscribers are experiencing sluggish speeds during congestion, they’re often described as being “deprioritized.”

There’s a shocking lack of public information about prioritization policies. Over the last year, I’ve dug into legalese to figure out how major carriers prioritize different plans, looked into the technical mechanisms behind prioritization, and spoken with industry experts. I’ve created what I believe is some of the most detailed content about prioritization policies among carriers in the U.S. Despite all that, I regularly find myself confused about prioritization.

Vague deprioritization disclosures

Carriers typically disclose the possibility of deprioritization in fine-print statements along these lines:

During periods of congestion, subscribers on this plan may experience data speeds slower than those received by other subscribers on the network.
Disclosures tend to be vague. Carriers almost never discuss the frequency of deprioritization, the severity of speed decreases, or the locations where subscribers will be especially prone to deprioritization.

Is deprioritization a big issue for low-cost plans?

Mobile virtual network operators (MVNOs) are often prohibited from discussing specific terms of their arrangements with host networks. In many cases, MVNOs can’t even make clear statements about which networks they operate over. In preparation for this post, I talked with a few people who are knowledgeable about the MVNO industry. None of them wanted to be quoted.

The general view among participants on online forums about the wireless industry is that almost all MVNO subscribers are deprioritized. However, there are competing claims. The MVNO Ting published a blog post titled Do MVNOs get second class cell service? The post explicitly states that Ting subscribers on Sprint’s network have priority on-par with typical Sprint subscribers. Ting’s post also seems to carry an implicit suggestion that subscribers using other MVNOs are usually not deprioritized:

The truth of the matter is, Sprint’s MVNO contract states that Sprint must provide its Customer MVNOs with service parity to traditional Sprint wireless voice and data service. It’s all laid out in very clear terms…The problem is, the discussion of whether or not carriers throttle and traffic shape MVNOs on their network takes on a conspiratorial tone online. I know, it’s shocking! Suppositions get accepted as fact. Assumptions leap off from suppositions and next thing you know, it’s all true because someone read it on the Internet. Hopefully this helps to dispel the myth.

Wirecutter has written about prioritization, and it looks like the company reached out to a handful of carriers about their policies (emphasis mine):1

[Some carriers] prioritize their own customers over third-party prepaid traffic, as happens with the Metro by T-Mobile subsidiary. A T-Mobile spokesperson confirmed that policy, saying that although postpaid and prepaid T-Mobile service have the same priority, Metro by T-Mobile and other resellers ‘may notice slower speeds in times of network congestion’…However, AT&T, Sprint, and Verizon told us that they don’t impose any such prioritization.

I don’t buy it. Xfinity Mobile, a popular Verizon reseller, explicitly acknowledges deprioritization:

In times of congestion, your data may be temporarily slower than other traffic.

Verizon makes it clear that its own prepaid subscribers will be deprioritized during congestion. Verizon’s flanker brand, Visible, also deprioritizes subscribers’ data.2

While these facts don’t rule out the possibility that Verizon gives high-priority access to some resellers, I’d be awfully surprised if subscribers with Verizon resellers typically have higher priority than a large portion of Verizon’s own subscribers.

Does deprioritization matter?

There are a lot of reasons people experience slow speeds, and people may be too quick to assume that deprioritization is the source of lousy speeds. Unfortunately, I don’t know of any publicly available data that sheds light on how often deprioritization causes trouble for consumers. As far as tell, drive tests assessing network performance typically use high-priority services. I’d be interested to see how assessments would come out tests were run with low-priority services.

Sharing better information

I’m aiming to offer the best public-facing content about prioritization policies. If you work in the wireless industry and would like to talk publicly or privately about prioritization policies, please reach out.

Mint Mobile’s Super Bowl Promo: A Retrospective

Christian

During the Super Bowl, Mint Mobile was offering new customers three months of free service. In my earlier post about the promotion, I said that I thought the deal wouldn’t have any serious catches or gotchas that made it less appealing. Sure enough, that’s how things panned out. I really appreciate that Mint Mobile doesn’t engage in gimmicks to nickel-and-dime its customers.

As expected, new customers were able to get free service, a free SIM card, and free shipping:

Mint Checkout

Before the promotion launched, I wondered whether Mint Mobile would try to push people into plans that renewed automatically. I’m happy with the approach Mint took. On the checkout page for the promotion, Mint conspicuously featured details about its auto recharge program. While the auto recharge box was checked by default, the program was clearly explained, and the box was easy to uncheck:

Mint Mobile Auto Recharge

Promotion duration

In my last post, I wrote:

It sounds like the promotion will last for something like 3-5 hours. Mint is suggesting that the promotion will end when the final whistle blows at the end of the Superbowl (though I won’t be shocked if the promotion stays available for a bit longer).

Sure enough, Mint extended the promotion. Here’s a screenshot from Mint’s website last night:

When I checked Mint’s website shortly after midnight on the West Coast, the promotion looked like it was still running. When I checked Mint’s website this morning, the promotion was over.

Proof & Network Evaluations

Christian

RootMetrics recently tweeted about how its latest analyses prove which networks are the fastest and most reliable:

RootMetrics' Tweet

I’m a big fan of RootMetrics, but the tweet annoyed me. There’s a ton of flexibility in how network evaluators can approach their work. Will performance data be crowdsourced from consumers or collected via in-house testing? How will data be cleaned and aggregated? What regions will be included? Etc.

Phrases like “fastest network” and “most reliable network” are ambiguous. Do you determine the fastest network based on average download speeds, median download speeds, or something else?

RootMetrics’ tweet is especially odd in light of their latest report. Depending on which speed metric you choose to look at, you could argue that either Verizon or AT&T has the fastest network. AT&T narrowly beats Verizon in terms of median download speed. Verizon narrowly beats AT&T in RootMetrics’ overall speed scores.1

RootMetrics’ Results In Dense City Centers

Christian

RootMetrics’ most recent report includes information about network speeds in city centers. Here’s how RootMetrics’ describes the tests:1

Capacity is critical for a good mobile experience in highly congested areas. To show you how the carriers performed in high-traffic areas of select cities, we measured each carrier’s median download speed outdoors in the dense urban cores of 13 major cities across the country.

The results are presented with the following graphic that lists the median download speed in megabits per second from the fastest and slowest network in each city:2

RootMetrics' median download speeds in city centers

Despite Sprint’s lousy performance in RootMetrics’ overall, national-level results, Sprint still managed to offer the fastest speeds in 3 of the 13 city centers RootMetrics considered. The results are consistent with a point I’ve tried to emphasize in the past: if you tend to stay in one area, you shouldn’t worry too much about national-level network performance.

Prioritization

I haven’t seen information about which plans RootMetrics’ test devices use on each network. In fact, I’m not entirely sure RootMetrics uses service plans that are available to regular consumers. My guess is that RootMetrics’ test devices have prioritization and quality of service levels at the high-end of what’s available to regular consumers. If my guess is correct, RootMetrics’ test devices have higher priority than many budget-friendly services that run over the Big Four networks. A few examples of those services:

  • AT&T: Cricket Wireless unlimited plans
  • T-Mobile: Metro by T-Mobile plans
  • Sprint: Mobile hotspot plans and Boost Mobile plans
  • Verizon: Verizon Prepaid plans, the Start Unlimited plan, and Visible plans

Subscribers using low-priority services are especially likely to experience slow speeds in congested areas. I’d be interested in seeing RootMetrics rerun its city-center tests with low-priority services.

RootMetrics’ Report For Late 2019

Christian

Yesterday, RootMetrics released its latest report on the performance of U.S. wireless networks. I’d been looking forward to this report. RootMetrics’ drive testing methodology has some advantages over the approaches used by other companies that evaluate network performance.

Results

RootMetrics’ results were generally unsurprising. As with the last report, Verizon was the big winner, followed by AT&T in second place, T-Mobile in third, and Sprint in fourth.

Here are the overall, national scores out of 100 for each of the major networks:

  • Verizon – 94.6 points
  • AT&T – 93.2 points
  • T-Mobile – 86.5 points
  • Sprint – 83.2 points

RootMetrics also reports which carriers scored the best on each of its metrics within individual metro areas. Here’s how many metro area awards each carrier won (along with the change in the number of rewards received since the last report):

  • Verizon – 660 awards (-12)
  • AT&T – 401 awards (+21)
  • T-Mobile – 217 awards (-20)
  • Sprint – 80 awards (-9)

AT&T’s improvements

RootMetrics’ results align with the results of other recent evaluations suggesting aspects of AT&T’s network are becoming more competitive. AT&T fared particularly well in RootMetrics’ latest speed metrics. While Verizon narrowly beat AT&T in the final speed score out of 100 (90.7/100 for Verizon vs. 90.2/100 for AT&T), AT&T narrowly beat Verizon in aggregate median download speed (33.1 Mbps for AT&T vs. 32.7 Mbps for Verizon).

It appears that RootMetrics’ final speed scores are based on something more than median download speed. That may be a good thing: having consistent speeds is arguably much more important than having high average or median speeds. Still, I’m frustrated that I can’t figure out exactly how the final speed scores are derived. RootMetrics continues to be non-transparent about the math underlying its analyses.

A section of the latest report suggests that Verizon may do a particularly good job of avoiding sluggish speeds:

Verizon’s ‘slowest’ median download speed of 17.9 Mbps, recorded in Fresno, CA, was still quite strong and would allow end users to complete the majority of data tasks with ease. In fact, Fresno was the only market in which Verizon registered a median download speed below 20 Mbps. No other carrier came close to matching Verizon’s consistency of delivering fast speeds in metros across the US.

5G performance

The new report includes details about RootMetrics’ recent tests on 5G networks. I found the 5G results unsurprising, and I’m not going to comment on them further at this time. I think 5G deployments are still in too early a stage for the results to be of much interest.

Link Roundup – 1/25/20

Christian

This post is the first of hopefully many link roundup posts.

  • Masnick of TechDirt explains how Apple dropped plans to encrypt its customers’ backups after the FBI pushed back on the company’s plans. It seems like there’s a lot of misinformation going on in this situation.
  • Physicist Casey Handmer is extremely optimistic about SpaceX’s Starlink. If you’re not already familiar, Starlink is a satellite constellation SpaceX is working on to provide global internet connectivity. At the moment, the company plans to have over 10,000 satellites deployed in the constellation.
  • Tim Farrar is very critical of Handmer’s post. In my opinion, Farrar is way too negative on Starlink, but many of his criticisms seem correct.
  • The U.K. government may place some limitations on Huawei, but it looks like the company may have an easier time in the U.K. than in the U.S.
  • There seems to be ongoing confusion about how TracFone’s SmartSIM product will work.

MVNOs Hiding Their Host Operators

Christian

Mobile virtual network operators (MVNOs) often appear to be prohibited (or at least discouraged) from explicitly acknowledging which networks they run over. Still, it seems that most MVNOs don’t have to keep their host networks entirely secret. The effects of these policies can be kind of funny.

Red Pocket

The MVNO Red Pocket operates over all of the major U.S. networks, but it only mentions Sprint by name. The following screenshot comes from a part of Red Pocket’s website that lists the networks the company offers service over:

The unnamed networks are color-coded to match the colors used in the host networks’ branding: AT&T in blue, T-Mobile in pink, and Verizon in red.

Ting

The MVNO Ting currently operates over Sprint and T-Mobile’s networks. Verizon will be added to the list soon. Today, Ting published a blog post about the upcoming addition. Ting still isn’t naming the networks it works with, but the company is making easy-to-interpret statements like: “In 2020, Ting will be on every major coast-to-coast network except AT&T.” The blog post included a video where the names of each network the company works with were censored out:

Xfinity Mobile Growing Quickly

Christian

An SEC filing Comcast published today showed that Xfinity Mobile has been adding subscribers quickly. In the last quarter, Xfinity Mobile added 261,000 new lines.1 Overall, Xfinity Mobile had a net increase of 816,000 lines in 2019, bringing its total subscriber count at the end of the year to slightly over two million lines.2 It’s now fair to say that Xfinity Mobile is one of the largest MVNOs in the U.S.

By my math, Xfinity Mobile’s subscriber base grew by about 66% in 2019.3 It will be interesting to see whether Comcast can keep that kind of growth rate going forward.

By one method of accounting, Xfinity Mobile looks unprofitable. However, the brand seems to be moving in the right direction towards profitability:4

Cable Communications results include a loss of $116 million from our wireless business, compared to a loss of $191 million in the prior period.
One could argue that way of looking at the financials leaves out something important. Subscribing to Xfinity Mobile makes it harder for a Comcast customer to cancel their internet service. Looking exclusively at the balance sheet of Comcast’s wireless segment leaves out the benefits Comcast receives as Xfinity Mobile reduces churn in the company’s other product segments.

At the moment, Xfinity Mobile presents a pretty good value proposition for current Xfinity Internet customers. Going forward, I’m not sure whether Comcast will keep Xfinity Mobile’s prices competitive to encourage growth or increase prices to make each subscriber more profitable.

Opensignal’s 2020 U.S. Mobile Performance Report

Christian

Today, Opensignal released a new report on the performance of U.S. wireless networks. The report details results on seven different metrics.

Here are the networks that took the top spot for each metric at the national level:

  • Video Experience – Verizon
  • Voice App Experience – T-Mobile/AT&T (draw)
  • Download Speed Experience – AT&T
  • Upload Speed Experience – T-Mobile
  • Latency Experience – AT&T
  • 4G Availability – Verizon
  • 4G Coverage Experience – Verizon

It’s important to interpret these results cautiously due to limitations in Opensignal’s crowdsourcing approach. Since performance data is collected from actual subscribers’ devices, factors not directly related to underlying network quality may impact the organization’s results. For example, if subscribers on a network are unusually likely to use low-end devices or live in rural areas, that will affect the outcome of Opensignal’s analyses. Still, Opensignal’s results are interesting; they’re drawn from a huge data set involving primarily automated performance tests.

Download speed findings

The most notable result in the latest report might be AT&T’s first-place finish on the download speed metric. In the previous Opensignal report, T-Mobile won first place for download speeds, and AT&T took third place. I’ve recently been critical of the methodologies used in some other evaluations that suggested AT&T had the nation’s fastest network. While many of those methodological criticisms still stand, the fact that Opensignal’s arguably more reliable methodology also found AT&T to have the fastest network leads me to believe I was too harsh. I’ll be interested to see whether AT&T also takes the top spot for speeds in RootMetrics’ upcoming report.

New metrics

Two new metrics were introduced in this report: Voice App Experience and 4G Coverage Experience. The Voice App Experience metric assesses the quality of voice calls via apps like Skype and Facebook Messenger. I’m not exactly sure how the metric works, but it looks like all four networks received similar scores. Opensignal deemed all these scores as indicative of “acceptable” quality.

The 4G Coverage Experience metric adds a bit of complexity to the previously existing 4G Availability metric. The coverage metric assesses 4G availability across areas all Opensignal’s users find themselves in, regardless of their network.