cokemaster: My 2c:
I suspect the larger telcos are dragged down by:
* a larger copper footprint than the smaller players
* Wireless broadband being slower than Fibre delivered plans
* likelihood that more customers are on entry level fibre plans. Eg. 30/50mbps, 100mbps.

Unless you can segregate access and speed profiles out - Ookla speed data is likely to be meaningless.

Customers equipment/devices as well and that either being poorly setup or faulty will absolutely skew results too.

I say this without my ISP hat on, as well these results do not tell the full story.

That doesn't stop myrepublic from trumpeting on about being the fastest ISP in New Zealand 🙄

This doesn't also tell the story of how big the sample size is.

The bigger players will have many more customers in their sample size but many are likely on entry level plans or copper (as @cokemaster points out). If you average it all out because of this the slower customers (or customers with crappy routers) will bring their average down.

But, the ISP that is #1 won't mention this. Instead they'll run with it once again. This is where I really miss Truenet as they did consider this also and actually made a change in the ISP market.

Seriously, why give this rubbish any attention? The methodology is fatally flawed and... I'll say it, actively misleading.

For people that know better to post this is disappointing - at least all the commenters so far see this.

Cheers - N

@Talkiet:

 

Seriously, why give this rubbish any attention? The methodology is fatally flawed and... I'll say it, actively misleading.

 

For people that know better to post this is disappointing - at least all the commenters so far see this.

 

Cheers - N

 

I think it's good to have this posted and discussed, because when a company put out claims we know where they are getting it from and what are the actual factors behind the numbers - and how those can influence the results.

michaelmurfy:

 

This doesn't also tell the story of how big the sample size is.

 

The bigger players will have many more customers in their sample size but many are likely on entry level plans or copper (as @cokemaster points out). If you average it all out because of this the slower customers (or customers with crappy routers) will bring their average down.

 

But, the ISP that is #1 won't mention this. Instead they'll run with it once again. This is where I really miss Truenet as they did consider this also and actually made a change in the ISP market.

 

From reading up about this - there isnt a strict sample size. Its based on consumer tests done against speedtest.net and then collated and summarised.

I have tests that run of my multiple fibre connections hourly and I have also developed my own scripts to run these tests automatically and the issue with speedtest.net is the servers you test against. There are HUGE variances in speeds againsts the servers and the times you test against them.
I run tests against servers in NZ and can find some of the bigger providers speedtest servers very variable. ie. ~ 940 on one test and ~ 600 on the next etc.

There are a few that I use which are accurate every time so I tend to hook onto those. 

For the purposes of this speed index - I can see this skewing the results heavily.

cokemaster: I concur with @freitasm. I bet that this is going to be used in marketing (and it has in the past)… so at least leaving some discussions online qualifying this is a “good thing” TM.

And bingo. They're running with it:

*slow clap gif*

I like the "Ookla trademarks used under license and reprinted with permission"... When you "win" this "award", Ookla allow you to BUY the results so you  can use it in marketing :-)

COMPLETELY UNBIASED AND NEUTRAL I TELLS YA!

Cheers - N

Talkiet:

I like the "Ookla trademarks used under license and reprinted with permission"... When you "win" this "award", Ookla allow you to BUY the results so you  can use it in marketing :-)

COMPLETELY UNBIASED AND NEUTRAL I TELLS YA!

Cheers - N

 

insane:

Talkiet:

 

I like the "Ookla trademarks used under license and reprinted with permission"... When you "win" this "award", Ookla allow you to BUY the results so you  can use it in marketing :-)

 

COMPLETELY UNBIASED AND NEUTRAL I TELLS YA!

 

Cheers - N

 

And yet it is your (Spark's) recommend means for testing speed ;)

https://www.spark.co.nz/help/internet/troubleshoot/check-broadband-speed/

The funny thing is each major RSP thinks they are the fastest and has commissioned some "independent" organisation to certify them as such. They can't all be right.

Ookla are the default standard and once you remove the local issues (wifi, rubbish browser, outdated hardware, uploading to iCloud at the same time etc), it's an easy test for end users to run and gives them a number that they understand.

It's entirely possible to endorse their individual test (with a few caveats) procedures yet maintain their "Speedtest awards" are fundamentally flawed and in several provable cases are actively misleading. That's where I sit.

To me at least, it seems like the ISPs with the genuinely best performance don't actually need to go around using dodgy metrics to try and prove it.

N