A Blog From Behind the Trenches

I guess the growth will slow down a little because of Opera Mobile. Maybe June 2011

Oh, I forgot all the iPhone users. So, in that case it compensate the growing popularity of Opera Mobile, and thus we can expect May 2011 is more likely after all

Anonim writes: It looks like non-accelerating growth from august till now. Have you guys performed analysis where the saturation level is? And how does the smartphone (with their full-blown browsers) dominance affects demand for opera mini (a light- or limited-browser)?

Originally posted by anonymous:

Have you guys performed analysis where the saturation level is?

Since there is no sign of slowing down, you can safely assume that saturation at least two times the current value, say some 100-150 mln.
Theoretical saturation is the number of mobile phones in the world, and that is approx. 4 bln.

Note, however, that the ceiling will hardly ever be reached, it's the growth rate that may decrease.

Anonymous writes: "looks like non-accelerating growth from august till now" Looks like you are talking out of your ass.

Anonymous writes: "And remember, these numbers are only for end-users. People who are using preinstalled and bundled versions of Opera Mini are not part of these statistics." How? The article says these are unique monthly users, so I guess they're counted by the proxy servers. Do those who get bundled O-minis go through different servers?

State of Mobile Web reports also show an increasing growth of transcoded data, so no matter what version (bundled or not) of Opera Mini is used - it is getting more and more popular every month.

Congrats on Opera Mini growth. I was using it on w995, now use Opera Mobile on my X1 and wouldn't trade Opera for any other (Skyfire, Fenec, IE,) browser in the world.

Originally posted by anonymous:

How? The article says these are unique monthly users, so I guess they're counted by the proxy servers. Do those who get bundled O-minis go through different servers?

Do you really think Opera is unable to figure out which edition of Opera Mini someone is using?

Originally posted by Kriolyth:

State of Mobile Web reports also show an increasing growth of transcoded data, so no matter what version (bundled or not) of Opera Mini is used - it is getting more and more popular every month.

AFAIK, that report only deals with the regular version, not bundled versions.

I have both Operamini and Opera Mobile on my phone but I tend to use Operamini as my default browser. The compression increases speed and reduces data usage making Operamini my prefered option over all other browsers (including other thin client browsers like Bolt which I also have)

Originally posted by haavard:

Does anyone want to take a guess as to which month Opera Mini will pass the 100 million users mark?

February 2011.

Originally posted by haavard:

Does anyone want to take a guess as to which month Opera Mini will pass the 100 million users mark?

April 2011.

Originally posted by Purdi:

AFAIK, that report only deals with the regular version, not bundled versions.

I can't find any proof, but I guess it's based on statistics from Opera Mini servers, and all versions of Mini on all platforms connect through them. There's also no point in discarding bundled versions data, though it might be interesting to know how many bundled versions are out there.

another anonim writes: " Anonim # 12. February 2010, 13:31 Anonymous writes: "looks like non-accelerating growth from august till now" Looks like you are talking out of your ass." well, it rather looks as you dont know what 'acceleration' means. from aug 2009 till now, there is no acceleration shown in this graph. it is something you cannot skew or interpret at will. acceleration is, or isnt. this graph shows acceleration not. simple. anyway, I really would like to know the answer to question, how smartphones with their no-compromise browsers affect demand for lite-browser like opera mini.

Please don't fight folks

Any more acceleration and the curve falls backwards, starts to change the history, and there's suddenly more Opera Mini users than there are mobile phone users

Originally posted by Kriolyth:

I can't find any proof, but I guess it's based on statistics from Opera Mini servers, and all versions of Mini on all platforms connect through them. There's also no point in discarding bundled versions data, though it might be interesting to know how many bundled versions are out there.

They are discarding preinstalled versions:

"The data presented in "Part 1: Growth" represents people who have downloaded Opera Mini. Those numbers do not reflect users of operator pre-installed versions of Opera Mini."

Originally posted by anonymous:

well, it rather looks as you dont know what 'acceleration' means. from aug 2009 till now, there is no acceleration shown in this graph. it is something you cannot skew or interpret at will. acceleration is, or isnt. this graph shows acceleration not.

Are you drunk or something? Look at May 2008 to August 2009, then look at the part after that. The graph is CLEARLY showing accelerated growth over time.

Only a dishonest drunkard would look at a tiny piece of the graph and make conclusions based on that. Look at the whole thing.

Anonim writes: "Are you drunk or something? Look at May 2008 to August 2009, then look at the part after that. The graph is CLEARLY showing accelerated growth over time. " acceleration is a 'growth of a growth rate'. in the most recent period - and a half year one - starting from aug 2009, there is no acceleration. growth rate is steady. it WAS accelerating before that, it isnt now. for a half year, that is an epoch in internet times and please stop calling everybody else names. I'm really worried that this kind of behaviour is tolerated here.

Originally posted by anonymous:

in the most recent period - and a half year one - starting from aug 2009, there is no acceleration. growth rate is steady. it WAS accelerating before that, it isnt now. for a half year, that is an epoch in internet times

So you ARE drunk and dishonest. Look at the graph since May 2008. Notice the accelerating growth? Indeed.

Anonim writes: and look at the most recent period. not accelerating growth. indeed.

How many times do I have to tell you that you can't just pick a tiny part of the graph and conclude from that? Look at the whole graph. The growth is clearly accelerating.

Anonim writes: "Look at the whole graph. The growth is clearly accelerating." skewing statistics by choosing the apropriate timespan is quite common among fanboys. when recent data (aug2009-now) is 'not convinient' they often go back trought the timeline to ancient times, some most daring opera and apple fanboys even talk about Crusade times when needed. nobody denies, that opera mini usage grows, but one want true information - is the growth stagnant? where is the saturation level? what is the impact of full-browsers now in each and every smartphone on mini-browsers like opera mini? these are the questions that raise from this graph. I'm sure that people inside opera ask them also and answers will determine what kind of budget opera mini will get. applying fanboy logic to business decisions? fail.

Originally posted by anonymous:

skewing statistics by choosing the apropriate timespan is quite common among fanboys

Indeed, which is why you are narrowly looking at a tiny part instead of looking at the trend, which clearly shows accelerating growth.

when recent data (aug2009-now) is 'not convinient' they often go back trought the timeline to ancient times

Yeah, "ancient times" like 2009? LOL, you are indeed drunk.

nobody denies, that opera mini usage grows, but one want true information - is the growth stagnant?

Clearly not, considering that it's growing faster than ever.

applying fanboy logic to business decisions? fail.

Being a drunken ignoramus? Fail.

Look, now you kids are disputing the exact nature of Opera's growth pattern? Is this going to devolve into the Hitler argument or the Star Trek argument? The important thing is that the little numbers are going *up* not *down*, and growth will pick-up further when Mini 5 lands in Final. So sit down.

The graph includes only voluntary, after-market users of Opera Mini. This is done for two main reasons.

1) Opera has shipped, preinstalled, on roughly 200M phones, thus being the original browser on those devices. Opera wishes to feature the trend of SWITCHING from another phone browser INTO using Opera.

2) Opera has revenue-sharing contracts with cellular operators worldwide. These preinstalled browsers are often customized with non-standard changes to the interface or features (See ATT.NET browser for one example). Rather than tally up these varied (and often compulsory) usage shares, the company has chosen to highlight only the aftermarket users.

Originally posted by Purdi:

Those numbers do not reflect users of operator pre-installed versions of Opera Mini.

Thanks, now it's clear.

Cutting Spoon Nice point

Whether 'anonymous' is drunk or not, you have to admit the graph is not very detailed, and if taken too seriously it's easy to say that, based on what is seen in the graph only, growth acceleration is somewhat 'jerky'. In very basic terms, the graph could look a bit nicer. This wouldn't really convey much extra meaning, but could for instance smooth out the curve and therefore eradicate the possibility of the assertion that "there was no growth upon growth between August '09 and February '10".

@anonymous, we simply cannot see what went on in those months because the graph doesn't show us the data that finely.

Before the end of the 2010

November, I guess

A month-by-month breakdown rather than by quarter would be a little more illuminating of the recent growth trends.

Originally posted by theoddbod:

A month-by-month breakdown rather than by quarter would be a little more illuminating of the recent growth trends.

http://www.opera.com/smw/

I bet Opera will be more and more popular. A Vietnamese cellphone company set Opera Mini as their phones' default web browser and the sales of that company is increasing.

The SMW data can be nicely fitted to a monoexponential growth curve y = A*exp(B*x) where y = no of users, x = months since launch (I assumed Nov 2005 as the launch date), with A = 0.835 and B = 0.082.

This leads to 100 M users after 58.35 months, i.e. on the 10th of September 2010.

Already!

http://files.myopera.com/HansKiefer/files/Opera%20Mini%20Users.jpg

edit: included the data for 10 and 20 M users. As a result, postponed 100 M user date by 3 weeks. The growth rate seems fairly robust, and I believe we are so far away from the 4 b theoretical users that there is no saturation yet.

Thanks Hans, that's very precise! We will wait and see about the level of accuracy.

I'm going for September 2010

What an interesting discussion...

A friend somehow stumbled across this, and emailed me about it. I agree with his opinion: User share has more than doubled every year. That being the case, (especially with Mini 5 and potential iPhone release) usage may blow right past the 100M mark in 2010.

We'll need to wait until the end of March before we can get February's numbers, but I wonder if publicity from MWC has influenced the growth rate. I'd say we could see 100M by the end of August if all goes well, or by November if the iPhone thing falls through.

If Opera gets on the iPhone you have maybe 5-15 million or more just from that within a few weeks.

Hey! I think we just agreed on something! Thank you! I can't see 100M taking more than 10 months, and 6 months is possible with an extra 10M+ iPhone/iTouch/iPad users.

The reason why such things grow exponentially with a constant rate over several years is most likely that everybody who is happy with the product (Opera Mini) will convince a certain number of friends within a certain time with a certain probability. People are just the same in letting themselves convince to try something new, no matter what phone they use. Therefore, I don't expect a noticeable effect if OM gets into the iPhone app store. But we'll see (or not). Actually, I have to correct my bet slightly, because I used the wrong function. Correctly, it should read y = A*exp(B*x)+ C. I refitted the data and ended up on the 20th instead of the 10th of September. Still in time to celebrate 5 years of Opera Mini!

Originally posted by HansKiefer:

Therefore, I don't expect a noticeable effect if OM gets into the iPhone app store.

Define "noticeable".

Originally posted by Purdi:

Define "noticeable".

According to your assumption, we should see a positive deviation from the predicition function by 5-15 million users within a few weeks from launch on the iphone. This is what I would call "noticeable".

I'd second for October:
http://files.myopera.com/Kriolyth/files/operamini_users.png
Logarithmic scale, 100 mln users at the top. Green line points to September, red line to October. With a month and a half for report to be created, by the New Year (or Christmas) we'll know that Opera reached 100 millions

Non-traditional growth and extraordinary circumstances will often have a significant one-time impact on the curve. A launch of Mini on the iPhone may see wide media coverage, and spawn a lot of "curiosity" downloads.

Opera gets 3-5 mln users every month; even if we account for speed increase, "instant" 15 mln users will "save" two or three months. Cooool

Prediction updated

http://files.myopera.com/HansKiefer/files/Opera%20Mini%20Users.jpg

100 M users in Oct/Nov with latest data available.