Troll marketing: the broadband wars

769

If you’ve been paying attention to the scenery in Colombo, you’ll have seen Etisalat adverts ALL over the place, promising this:

This, by the way, is taken from Etisalat’s Facebook page.
This, by the way, is taken from Etisalat’s Facebook page.

                               
A quick glance at http://etisalat.lk/Dual_Carrier_Service.cfm confirms: they say they’re six times faster than the competition. 
 

Six times faster than everybody else? Now that kind of speed is too good to miss, so we tested the claim ourselves in an effort to separate fact from marketing and see how fast “6 times faster” really is.

 
Firstly, some background on the matter. Etisalat, the communications giant, has been providing high-speed HSPA+ services for some time now. They recently upgraded their tech to DC HSPA+ – Dual Carrier HSPA+. This is a technology that works by creating and allowing access to two simultaneous HSPA+ data streams: it’s fully backward compatible with the normal service.

 

Like Etisalat.lk says:

5).What Can I do with DC-HSPA+?

You can double whatever the speed you receive from HSPA+ with DC-HSPA+. You will experience really fast download speeds.

 

Two data streams: twice the bandwidth: twice the download speed – in theory. In practice?

As any internet user knows, the only way to actually measure speed is to download something. We conducted a survey of mobile broadband speeds on TRCSL premises, using identical hardware and software – nothing flashy, but mid-range consumer laptops like the average user would have.  The towers – not just Etisalat, but others – are just 500 meters away from the test site,  right opposite to the Elvitigala flats: perfect, consistent reception, and as good as you’ll ever get in Sri Lanka.

 

Here are the top three contenders (as of the time of writing):

Speed test (note: results are in Megabits per second)

Network

Server

1st Test

2nd Test

3rd Test

Average

Etisalat DC HSPA+

USA

6.38

5.16

6.44

5.99

GER

10.37

10.15

9.21

9.91

Airtel HSPA+

USA

8.08

8.24

7.67

7.99

GER

7.31

5.67

5.66

6.21

Etisalat 7.2Mbps

USA

5.97

6.03

6.13

6.04

GER

5.18

6.07

6.08

5.77

 

Here the facts come into play. As you can see, Etisalat fluctuates all over the board. Airtel’s HSPA+ service solidly beats the Etisalat 7.2 Mbps, performing close to 1 Mbps better in all tests. Now look at the Etisalat DC HSPA+ service.  There’s two HSPA+ lanes operating there, so one would expect it to be exactly twice as fast as the normal service: in fact, it’s not even close. It even consistently loses to the single-lane Airtel HSPA+ when downloading from the furthest server (USA).
 

So no, Etisalat DC-HSPA+ is NOT “six times faster” as their advertisements claim. They aren’t really even twice as fast: looking at the data, the single-lane Airtel HSPA+ performs, on average, just as well as the new dual-lane system.
 

So is Etisalat’s marketing department lying to the consumers with their promise of dual lanes? No, as it turns out, they’re not. We also conducted a bandwidth test to confirm:
 

Bandwidth test (note: results are in Megabits per second)

Network

Server

1st Test

2nd Test

3rd Test

Average

Etisalat 7.2Mbps

USA

7.03

5.29

6.98

6.43

GER

7.12

5.94

6.76

6.60

Etisalat DC HSPA+

USA

9.94

9.91

10.44

10.09

GER

13.54

13.52

13.58

13.54

The bandwidth is there: two lanes, twice as fast as the existing Etisalat service. The problem is the marketing: it’s still not six times as fast as the competition. Users will know that there’s a vast difference between bandwidth and download speed – it’s like water in a pipe: if the water is your data, then bandwidth is the width of the pipe. Speed is how fast the water (data) goes through that pipe. Simply upping the bandwidth does not give more speed. In Etisalat’s case, when you actually download something, you’re not seeing speeds that are six times faster than the competition (not even close): neither is it actually twice as fast as their own existing, stable HSPA+ service.
 

So what’s Etisalat’s justification for the advertisements?
Theory.

HSPA+’s theoretically computed maximum is 21 Mbps (Megabits per second; remember the bits and divide by 8). Dual-carrier HSPA+ works by basically creating two lanes, doubling this theoretical value. Theoretically, you can download at 42 Mbps. Theoretically. The Etisalat website clearly says the word “theoretically”, but the public advertisements don’t. They probably have word limits on those billboards – after all, why put another word when you can get away with less, right?
 

Factually, as the tests indicate, you can’t even get close.  The maximum download speed we actually got was 9.91 Mbps off the German test server. The 42 Mbps that Etisalat claims to deliver is purely theoretical. In practice, you’re missing 32.09 Mbps. 
 

And of course, this doesn’t matter when it comes to pricing.  As Daily FT says:

[quote] “The Dual-Carrier HSPA+ will give users a download speed of 42 Mbps and an upload speed of 5.76 Mbps. The dongle and Wi-Fi routers are now being brought to market by Etisalat where a DC-HSPA+ dongle will cost approximately Rs. 16,000 and routers from Rs. 25,000.”[/quote]
 

Whoops. You have to pay Rs. 16,000 for “six times faster”. Contrast that with the closest competition, Airtel HSPA+, where a dongle is somewhere in the vicinity of Rs. 3000. That’s 5 times more money for speeds that beat the single-lane service by only a few KBps. This isn’t a comparison of Etisalat vs Airtel: it’s a look at Etisalat’s own claim to be six times faster than its competition.
 

Regretfully, we have to say: they’re good, but not as good as their advertising!

65 COMMENTS

  1. Agree! Thanks for highlighting these issues, most of the media talk the good and ignore these kind of adverts, b’coz E’s theire clients. Appreciate Read Me for the independent reviews. These are just mere marketing lies…..!

  2. This seems to be an excellent, independent test. One question to clarify something which is not clear about in your article.

    Did you use a DC HSPA+ supported dongle, and a DC HSPA+ enabled Etisalat broadband package when you do the testing? Or, did you use a normal dongle with a normal BB package and connected through DC HSPA+ APN?

    Just asking, because it wasn’t clear in your post. If it wasnt the case, then you have not really tested the DC HSPA+ network, but the same HSPA+ thing you tested with the 7.2Mbps dongle.

    Disclaimer: I work for [email protected], the Social Media agency for Etisalat so my comments may be biased.

  3. Hi,

    Could you also tell us the times at which the three tests were conducted? Assume they were fixed time slots, spread across a week?

    Agree all ISPs misguide customers by advertising maximum theoretical throughputs and don’t educate consumers on real world factors that impact throughput, I’ve just tested the connection on their new dongle and got the following throughput.

    Germany – http://cl.ly/image/3U1m27381W28
    USA – http://cl.ly/image/472r2033422O

  4. Agree fully, it’s all misleading advertisements. 6 times faster is 42 mbps which should be really really super utra fast. Faster than my SLT ADSL? Constant speed?

    Was surprised to such on independent review on such a web site. nowadays noboday talks about these stuff except the individual blog posts. As a responsible IT magazine hat’s off for you guys. When’s the print out? Sure it would be superb!

  5. First congratulations and good job Etislart for introducing DC service to Sri Lanka. But where it went wrong is your marketing and the way you guys challenged the rest. See, as a non technical person I don’t care about your theory or any, all I see it 6 times faster advertisements and I assume I get 6 times faster internet than my current connection, as you say 6 time faster than other providers and your own other connections.

    Now researching more on-line after this post, I’ve found out no network provider can give 42 mps. Never.

    So as a leading and innovative network provider please be more responsible on what you advertise, don’t lie. If your product is so good people will just buy it, you don’t have to brag about it.

  6. Hi Miuru,

    Kudos for your ‘experiment’. I have a question and I’m asking this out of sheer curiosity… It so happened that I was invited to one of Etisalat’s press events, targeting bloggers and social media personal to introduce this product/service. Now I’m not a major technical person but during a demo, they displayed that their DC HSPA connection with the dongle can actually download almost twice as faster. The bandwidth was clearly twice as much, I think it went from around 15Mbps upto 30Mbps. Can you explain this? Thanks!

    disclaimer: I work for a social media company called Loops Solutions and we manage social media for a different mobile network service provider.

    • Hi Indulekha,

      Thank you for your appreciation. There are many factors affecting mobile broadband speeds unlike fixed broadband.

      1. Mobile broadband, 3G HSPA speed varies widely time to time. This is the nature of this technology. (Highly unreliable)
      2. This was a random test done on mobile broadband. (Snapshot)
      3. Once in a while (1-4% of time) you’ll hit speeds closer to the advertised speed. But this is nowhere near the speed of theoretical upper limit. It is not fair to conduct advertising to make public believe that these speeds are real and achievable.
      4. Dedicated 30Mbps Internet circuit (constant speed) would cost around 2.5 to 3 million
      SLR/per month.

      When measuring speeds, we considered the average and not speed spikes (Abrupt upward variation of speed). 15Mbps or 30Mbps can be speed spikes. Usually operators do their maximum to show speeds especially during their demos and there is no evidence of testing methodology or tools used. And like you witnessed 30Mbps can be possible, but the possibility is very low.

  7. If Etislart really does 6 times faster than the rest as seen everywhere. No theory or other crap here. Just as they say “we are 6 times faster than the rest” they should get readme, trc sri lanka and do the same test at trc sri lanka and prove they are 6 times faster. sounds too serious ah? but why does a leading provider want to do fake advertising? (confused face)

  8. Hi Miru

    Facts read well!

    Readme.lk good work guys – unbiased in the way forward. Unfortunately though freedom of writing and expressing through facts and figures is not welcomed in our society I am glad that there are groups such as you guys to do the job – with courage!

    Thumbs up! and good luck with fighting ‘so called ‘counteractive’ comments as well 🙂

  9. Isn’t there an advertising regulations commissions? or a consumers authority to crack these type of misleading advertisements? Dam I relly though I would get 6 times faster net, very dissapointed after reading this on the provider too!

  10. Actually there are ISP’s who provide such service, for example i use Lanka Bell 8mbps unlimited connection and they dont pool me saving heavy use or anything , i use 150+ GBs monthly without any problem.

  11. I don’t see why Sri Lankan ISPs are only concern about internet speed ? it doesn’t matter how fast they, unless they provide reasonable data limit.. if I watch few 1080p youtube videos for few hours my data limit is over in few hours.. ooops..

    I know some isps around the world are now putting data limits since they are too lazy and greedy to upgrade their infrastructure.. as an example now Comcast is putting data caps but their minimum data cap is 300GB.. and additional 50GB will only cost you only $10.. this is reasonable comparing to Sri Lankan scam internet..
    http://arstechnica.com/business/2012/09/comcast-d

    And in here in SL we have a crazy bandwidth limit after the lame low data limits.. they can dor p your speed up to 50kbps.. even in India they provide 300kbps after the data caps… why are you not talking about these issues ?? why are you guys only talking about these useless internet speeds of our lame lame isps ?

    you know what, I m still using my old unlimited 512k SLT connection and they already blacklisted me illegally as a heavy user for twice now.. I had to contact TRSL twice to fix my connection.. I barely use 40gb per month. not to mention I have 5 members in my household. I m so disappointed to say I cant get an high speed SLT connection because their ridiculous data caps and high prices .. what is the use of a 8mbps connection if they put a lame data caps like 20gb ??? this is a complete joke comparing with other countries…

    Miuru, I really appreciate if you can talk about these issues more in your upcoming articles .. because we desperately need a to have more serious discussion about these issues more than talking about internet speeds … because we cant let these greedy isps to scam our people ..cant we ?

  12. Does anybody know, if TRCSL monitors and approve these marketing taglines? I don’t really care if six times or seven times, cuz I don’t have time and money to test out all marketing claims advertised by companies. I’d expect the regulatory bodies to monitor it and regulate, without letting companies to say what ever they want to say on ads.

  13. Looking at their advertisments I really though it’s 6 times faster. All fake 🙁 Now any place I complain this fake misleading advertising ways? TRC? I really wish Etislart would learn a very good lesson and the rest too would see it and learn a lesson.

  14. Hi Miuri,

    Great post, just curious why the other mobile broadband providers are not tested here,

    I would love to see a complete review with Dialog, Mobitel and Hutch also tested!

    Thanks.

  15. Just as I was about to make a comment I saw this. Seriously… out of curiosity, why isn’t Dialog and Mobitel brought into the picture?

    By far they have a larger base of users. I’m using a Dialog connection which sucks! After seeing your post, I’m actually thinking of shifting to Etisalat or Airtel.

    They promised us 4G about a year ago but so far they fail at least to give us 2 Mbps consistently.

  16. I noticed that you guys are doing a great job these days! please keep it up, However I am also using an Etisalat Dual Carrier service, purchased just two weeks back with a wifi router, because internet speed really matters to my business, and like you pointed out it is not exactly six times faster but i should admit that their speeds are much better than any other networks I have used (dialog and airtel), on average I get about 14 mbps (while sharing the speed with a group of 4 people in my office) which is actually way better than airtel or dialog i have used.

  17. Hi Miuru, this is Damitha from Etisalat Marketing Team.

    I think your test is a good initiative to open a discussion about mobile broadband speeds in Sri Lanka.

    Well, what our ads claimed was etisalat is “up to 6 times faster”. But yes it may not be the case in some occasions because of the limiting factors that you have also already pointed out in one of your comments above.

    However In majority of the cases if you compare our speeds against other mobile broadband providers, I am sure that we are delivering closer to 6 times faster speed compared to others.

    We cannot simply judge if etisalat is 6 times faster than others or not by conducting just 1 test against only a selected service provider at a single pre-defined location.

    If we are to conduct a test to verify if etisalat is up to six times faster or not compared to other mobile broadband providers, we will have to do at least 10-15 similar tests in different locations (ideally 30 locations) against all the mobile service providers.

    Then we can see on what percentage of time etisalat is faster than other providers (and by how many times), and on what percentage of time etisalat will be slower than others.

    I think this will be a better simulation of an actual user’s behavior than just coming to conclusions based on the results of just a single test like this.

  18. I’ve gone through the entire thread just to see a mysterious biasness in the author towards the TRCSL’s speed test. Looks to me like the author is avoiding any concerns towards trustworthiness of this test. He seems to be taking it for granted. None of the concerns regarding TRCSL testing tool’s trustworthiness was answered properly by the author.

    Speedtest.net may not be a trustworthy tool but it doesn’t give an automatic license for TRCSL tool to be more trustworthy.

    On the other hand it is crystal clear Etisalat is promising something that no one can ever guarantee to deliver. Who can vouch for X-Times faster claim with mobile broadband? Every novice knows mobile broadband speeds are volatile.

    If the author is genuine, and not being influenced by the TRCSL authorities (who are having a big PR crisis among average internet users in Sri Lanka after failing to resolve SLT unlimited broadband package problem) he should eliminate this biasness by testing an actual user scenario to do this experiment.

    To be honest this entire article sounds to me like an arranged publicity stunt by TRCSL to become a hero among Sri Lankan internet users. Readme.lk would have fallen to their trap unknowingly or knowingly.

    Prove me if I am wrong.

  19. Yo Bro, The whole point here is “Are they 6 times faster are not as their advertising” It’s a NO! that’s the whole point. Maybe you would assume all novice knows it, but NO being hte in the IT field my entire office didn’t know it. Also I would be very happy if the FUP is implemented in all users, then the heavy download users won’t dominated the internet speed. It would be equally shared among users.What is Unlimited after all? Also Etislart “Advertises” on speednet, what grantee speednetis is no baised for Etislart? So better test on a nutrual ground. Als after seeing this post, and in one comment someone said as a normal pulic, i can get appointment and walk into TRC and see the tests myself and ask any questions. AND YES i did it, He was very kinda to spends 2 hours with me, it’s their I learnt the SLT issue. I would advise you to make an appointment with TRC broadband head, and meet him.

  20. @Readme.lk – You have been trolled here 🙂

    Amount of comments white washing the TRC is extremely strange.

    @Namal Jayathunga – I am not saying Eti’s 6 times faster thing is an absolute truth. All I say is, the method Miuru used for testing it is having some bias.

  21. well, @Sujeewa, white washing TRC? Lets keep this topic on track. ETISLART is the culprit here dude! advertising fakely as 6 times faster!

  22. @Miuru Nice article

    So these test are done in a kind of a unbiased location and slight difference between speeds let alone 6 times faster, I thing the best place to start testing is with the most biased place etisalat it self . They should get 6 times faster or more if that is not the case then serious problem.

LEAVE A REPLY

Please enter your comment!
Please enter your name here