Rogers Explains Why Its Netflix Streaming Speeds Fared Poorly in Tests


Earlier this week Netflix released for the first time Canadian ISP data when it came to streaming speeds. Out of our incumbents, Bell ranked first, followed by Telus in ninth and Rogers at the bottom of the pack ranking 14th.

A follow up post on the Rogers Redboard explains why the company didn’t do very well in the Netflix tests:

Netflix’s test was done just before we virtually doubled Netflix capacity and we’ll continue to add more capacity as it’s needed. These results only apply to customers’ specific Netflix connection and not overall internet speeds.

Independent third party testing continues to show that Rogers offers top internet speeds.

Rogers, along with Telus and Bell have been part of the Netflix Open Connect network since early 2013. This network allows ISPs to connect to Netflix’s content delivery network for free to ensure the fastest and most efficient streams.

If Rogers has been part of the Open Connect network, why would its speeds fare so poorly? This was a question posed by a Rogers customer on the Redboard post’s comments section. Rogers explained:

Think of it like this. Let’s say we have a 2-lane highway to the Netflix building. Over time, the traffic gets more congested and more people are on the road, therefore the speed of the traffic slows down. “Throttling” would be removing a lane. “Adding capacity” is building another lane to allow traffic to move at a better, faster rate. Makes sense?

University of Ottawa Law professor Michael Geist gives his take on the responses by Rogers which state the company does not throttle Netflix traffic and that it has doubled capacity:

While these responses are meant to be reassuring, they raise troubling questions about how Rogers manages its network and whether the slow Netflix speeds could have been used to create a competitive advantage for its own online video services. While the company says that it does not throttle Netflix traffic (ie. deliberately slow it down), its response also suggests that it knew that the service was being slowed by insufficient capacity.

Joris Evers from Netflix told us he is aware of Rogers doubling its capacity and says he expects the next tests to show improved speeds by the company.


  • Hot Toronto Deals

    I tweeted @RogersBuzz about this, the day the results came out. Asking them how they balance their marketing, pushing the fastest broadband versus scoring the lowest on Netflix video streaming.

    The reply I received from @RogersElise was:

    “Netflix test was done just before we virtually doubled Netflix capacity, we will continue to add more capacity as required.”

    This definitely suggests that while they may not be “Throttling” (adding/removing lanes on a highway), they definitely are actively managing the bandwidth with traffic shaping tools (speed bumps on the highway). By speaking specifically about “Netflix” capacity, it alludes to specific active management/prioritization around Netflix (and maybe other high bandwidth apps: Skype, Youtube, ustream, etc.).

    Let’s see how ROGERS fares in the next months results… We all know how much ROGERS hates to be ranked, yet loves to promote the results of studies it commissions via “independent” third-parties.

  • DoctorT

    It doesn’t really work that way. Think of it like a router: Say they had 10 ethernet links connected (from their routing center to netflix’s connection center) before, that could be fine and all, but they probably got saturated over time, giving them bad speed results. Basically, they probably had too many people trying to move data through those 10 links. Now they say they’ve doubled the links. So according to that, they would have 20 links (so now everyone has enough bandwidth and more people can get on too). Each link is really expensive to upkeep, that’s why it’s a big deal to double the amount of them (it literally doubles your costs for Netflix routing). This is what they mean by capacity! (a doubling could also come from physically increasing the link speed – i.e. moving from a 1gbit link to a 10gbit on – this saves money too because you need less links now).

    In reality, there’s probably no throttling, it’s just the slow speeds were from too many users. It’s like when you have a few people in your house trying to download a big file all at the same time – making everyones speed slower as a result.
    Regardless of the method, they both ruin any possible speeds.

    Just as a side note, I don’t work for Rogers. I’m actually really against them (more because of their stupid expensive cell plans). I use Teksavvy DSL so I don’t know what the problem could really be. I just really don’t like when people assume something wrong.