When there is a mobile phone for half the planet: Understanding the biggest technology

We periodically report on the state of various technologies at this site, and my posting from last Spring reflecting on the final 2006 statistics "Putting 2.7 Billion in Context: Mobile Phones" has remained one of our best-read blog entries throughout last year. Now we have most of the relevant numbers for mobile phones for the end of the year 2007 and can make an update. And yes, as of November 2007 there were 3.3 billion mobile phone subscriptions around the world. At 6.6 billion people globally, yes, there is a mobile phone subscription for half of the planet. But that is not really the whole story. So lets delve deeper into these numbers.


UDATE Dec 10 2008 - I've just published my annual statistical review about the overall size of the mobile industry for end-of-year 2008. All new numbers so if you are interested in the "big picture", you may want to read this blog entitled Trillion with a T: Newest Giant Industry is Mobile

NOTE - that new blog story covers all of the topics in this, older blog posting, so unless you want to read about the history of mobile, I suggest you hop over to the newer blog where the current numbers are discussed.


OTHER TECHNOLOGIES

For a quick review of the context. There are about 1.3 billion fixed landline phones. So there are 2.5 times as many phone subscriptions as there are fixed landline phones. In Europe about a quarter of the homes have abandoned the fixed landline altogether and all family members use only their mobile phones (In Finland its more than half, even in the USA its already one in ten households where they have cut the cord). Contrasting with internet use, about 1.3 billion people use the internet worldwide. Or the other major technologies we could consider and compare, there are about 850 million automobiles in use worldwide. There are 1.5 billion TV sets in use, and 1.4 billion people own at least one credit card globally. But there are 3.3 billion mobile phone subscriptions !!

SUBSCRIBERS AND PHONES

So lets get the first point cleared. I make a point of talking about subscriptions, not phones. 3.3 billion mobile phone subscriptions as of November 2007 (Informa, Nov 2007) does not mean 3.3 billion different people have a phone. As we told you last year, already 28% of all mobile phone owners around the world have two or more subscriptions (Informa, Apr 2007). Or to put it in another way, 2.6 billion people have at least one phone and the remaining about 700 million subscriptions are with those people who have two or three or even more subscriptions.

SECOND SUBSCRIPTIONS

Some of the people with two or more are business execs who may have a business phone like a Blackberry for business use, and a personal phone for personal use. Others are people who have modest or low incomes and have multiple subscriptions to get the optimal telecoms pricing, with one network offering low cost evening calls, another low cost weekend minutes, etc. With the GSM based systems at over 80% of all subscribers (GSM Association Nov 2007), GSM has the SIM card module, which allows easy switching from one network to another without needing several phones. In some countries GSM phones are locked to one network such as mostly in the USA for example, so this may not be as easy, but in most GSM countries the phones are unlocked and users can easily change networks many times any day. As I said, more than one in four mobile phone owners has two or more phone subscriptions. Note that the other stats I mentioned also include multiple ownership or use. Many business execs may have one desktop PC for work, and a second laptop for occasional travel. The PC population of 900 million PCs actually hides a total unique PC owner number that is smaller than 900 million. The same is true of cars (some rich people own several cars) and TV sets (many young single people in the industrialized world have two TV sets, one in the living room, another in the bedroom) etc.

But as I've said in my telecoms marketing, churn, loyalty and segmentation workshops since 2003, the biggest group of new customers to get a second subscription is not business executive or the poor low income customer, but rather the first time employed. Half of British young employed adults report that they think its a sign of being important, or cool, to have two or more phones. Not subscriptions. Phones (Office Angels, Oct 2006). And now mobile operators admit publically what they've been saying in private to me in my workshops that it is true, the biggest customer group to get second subscriptions is the young adults as the Swiss mobile operator/carrier Sunrise reported at the Loyalty and Churn conference that I chaired in London in Nov 2006.

REPLACEMENT CYCLES

Why do we care? Well, the young employed adult is the ultimate target audience. They are young enough to still want to try all cool new things, but have the disposable income to spend on anything they consider new and cool. So young first-time employed are the biggest group to have two or more subscriptions. They are also fully addicted to mobile phones. So for them definitely two subscriptions means two phones. Which brings us to replacement cycles. Where the average replacement cycle for PCs is about three and a half years, the replacement cycle for mobile phones is 18 months and still shrinking (Semiconductor Industry Association May 2006). If you have two phones on two networks, you then have an effective replacement cycle of 9 months. But in the most advanced markets like South Korea, Japan and Hong Kong, the market has already adjusted so far that young adults replace phones every 6 months as Jim O'Reilly and I report in my latest book Digital Korea. Note that with a 6 month replacement cycle for a new phone for young employed adults, that means that mobile phones are now being replaced in harmony with the fashion industry cycle: a Spring Fashion cycle and an Autumn Fashion cycle. As I was in Tokyo in January and September, I witnessed that very pairing of fashion cycles and the new flashy phones to reflect that pattern.

HANDSETS

Now as the phone handset makers like Nokia, Samsung, Motorola, SonyEricsson and LG ship over a billion phones annually (IDC, Jan 2007), we have a colossus of an industry of high tech pushing ever more powerful gadgets into our pockets. And yes, Nokia alone ships one million phones every day of the year, Saturdays, Sundays and Holidays included. For contrast note that the PC industry shipped 250 million new PCs in 2007, of which about 100 million are laptops (Computer Industry Almanac Jul 2007). TomiAhonen Consulting has updated the handset features numbers against all shipments, replacement cycles and installed base, and reports that at the end of 2007 of all world phones in use, 72% were data-capable so-called 2.5G phones on data technologies like GPRS, CDMA 2000 1xRTT, and EDGE - that is 2.4 billion phones or almost three times as many data capable mobile phones in use, as personal computers, laptops and desktops worldwide. Over half of mobile phones have bluetooth capacity (1.7 billion) and already 12% or 400 million mobile phones in use are fully 3G capable (or faster still). Again, that 72% of phones can access the internet or 12% are 3G capable does of course not mean that the phones are used that way, just like most music-capable phones are not used to consume music. But the numbers nonetheless are very impressive.

MOBILE INTERNET

But as I mention data-capable phones and internet use, lets talk internet. There are about 900 million personal computers in use and during this year 2008 the world will pass the 1 billion PCs in use level (Forrester June 2007). The growth of the PC population in use numbers have slowed for several years now and are nearly static (In Japan the annual PC/laptop sales numbers are already declining), while obviously mobile phone subscription and phone shipment numbers have kept on growing rapidly (we added 600 million new phone subscriptions last year which reflects a growth rate of 22%. But yes, what of the famous 1.3 billion internet users? How come there are more internet users than total PCs in use? There are three primary reasons for that. First is shared computers. A family may have a family computer for the children and parents to share. A small office may have a computer that is shared by two employees etc. The second factor is access by those who do not have their own computers so for example internet cafes and university computer labs. Finally there are of course those who access the internet via a mobile phone, which is already the majority of internet access in countries such as South Korea, Japan and India.

What? Mobile phone access to the internet? Yes. Real Networks reported in 2007 that 25% of all mobile phone users around the world access the internet on their phones. That is a staggering 825 million people already. And yes, please note that most of these users also have a PC to access the web, and use both a PC and a mobile phone. And early on, as users start the migration from PC use to partial (and for some even total) internet use on mobile, it will be a transition, and the usage on PCs will be more than on mobile phones. And as internet use on phones is often costly and internet speeds on (broadband) PC connections are faster, then the total internet traffic on PC based internet access will remain much larger than that on the mobile.

TomiAhonen Consulting tracking numbers are also updated for mobile, PC based, and combined access to the internet. So out of all 1.3 billion internet users, only 37% access exclusively by personal computers (desktops and laptops). This includes all access from internet cafes and computer labs and shared PCs. Another 33% of internet users access by both PC and mobile. And already 30% of all internet access in 2007 was exclusively from mobile phones. If you want to see the glass as half empty, you can fairly use the numbers and say that 70% of all internet users have access by PC/laptop and only 30% of the internet access comes from those users whose only access is a mobile phone. Or for the glass-is-half-full crowd, we can equally fairly use the same data and say that already 63% of all people who access the internet do so from their phones at least part of the time, and only 37% of internet users are accessing only from a personal computer/laptop.

Nonetheless, the transition is very powerful. Remember that user numbers are the first shift. The more relevant shift comes with usage. In Japan the regulator reported that 2006 was the first year when more internet access was also by mobile phones than by PCs. Flirtomatic, the UK based online dating service on PCs and mobile phones reports that they get 4 times more traffic from mobile internet (WAP) users than PC based web users of the same service. In America a November 2006 report by Telephia found that the first category of internet content to have migrated from PC based internet to mobile internet is weather information, with web sites such as Accuweather, The Weather Channel and Yahoo Weather all reporting much larger mobile usage than PC based internet use, up to four times heavier use from mobile phones than PCs.

Do not misunderstand me. There still are a little bit of more users accessing the internet from a PC than mobile phone at the end of 2007, but the trend is unstoppable now, and during 2008 we will see the cross-over point for users, where more internet access will be from phones than PCs. That is users, not usage. Where people have access to both methods, probably for many years more usage sessions and the duration of their internet access will be from PCs than from mobile phones. But as Japan often leads in these numbers, we've already seen the next stage, that more total usage will start to come from mobile phones, rather than PCs. And then after that, probably, the total traffic will also eventually migrate to the mobile internet. But do bear in mind, that this change will be more by phones similar to the Apple iPhone and the like, than the early WAP phones of the beginning of this decade. Google incidentially reports that access by iPhone users is much larger than by users of phones from other makers. This is part of that Era of Before iPhone and After iPhone that we blogged about a lot last year. There is a global transition going on to the newest mass media, mobile (or like I say, mobile is the 7th mass media). This is the biggest economic opportunity in our lifetimes. You don't want to miss it.

MOBILE CONTENT

Which brings me to the revenues of the mobile internet (or VAS, Value-Add Service) data services versus content revenues on the older PC based internet. We saw the cross-over point in 2004 where more content revenues were earned by mobile than fixed PC based internet. In 2006 the numbers were 25 billion dollars of internet content revenues vs 31 billion on mobile content (Guardian, May 2007). Now the latest numbers tell of the enormous growth in the mobile content revenues for 2007 with the total value of VAS content on mobile being worth 45 billion dollars (Informa Jan 2008). What does it consist of? The biggest internet content revenues are still the classic first categories for any new media, pornography and gambling. But even thought the mobile internet is much younger than the PC based internet, it has also matured much faster. Adult entertainment and gambling are on mobile phones as well, of course, but now the two largest categories of mobile content are music at 8.8 billion dollars (about 5 billion of that is ringing tones) and mobile social networking at 5.05 billion dollars worldwide.

SMS TEXT MESSAGING

But there is the truly enormous data application which I have not even mentioned yet. SMS text messaging. The most used data application on the planet. As addictive as cigarette smoking as the Queensland University study found and we've reported here and in my various books. CMG Logica reported that in January 2007 the world passed the 2 billion active uses of SMS text messaging threshold. That was 74% of all mobile phone users at the time (growing from 72% the year before). TomiAhonen Consulting has projected that to be 2.4 billion active users of SMS texting at the end of 2007. Even Americans get it now, as more than half of American cellphone owners have become active users of texting according to the CTIA, and a major driver of that has been voting for reality TV shows like American Idol, as Alan Moore reports in the SMLXL White Paper on Pop Idol.

So one in three people on the planet already uses SMS text messaging. Mobile phone users in the UK prefer SMS text messages to making voice calls (JD Power May 2007). And for most of the decade all use of SMS is preferred over email and instant messaging, even in America as their youth discovered this two years ago and now that trend is spreading (Comscore Jun 2006). But what of the contrast? The numbers are overwhelming. Internet based email has been around for almost two decades and has nearly 800 million individual email users, who maintain 1.45 billion email accounts (Vision Gain May 2007). eMail user numbers are very static, modest growth at best. So 800 million active users of email and 2.4 billion active users of SMS text messaging. That is yes three times as big. And what of IM Instant Messaging. I don't have the very latest numbers, but the magnitude is about half of email users, so if we're generous, lets call it 500 million IM users. SMS is nearly five times as big.

And what of the money involved? SMS user numbers grew by 16% from 2006. SMS traffic numbers grew by 50% (!!!!!) and inspite of big price wars in telecoms and the offers of big buckets of free messages, SMS revenues grew by 23% (Informa Q1, 2007). Duh... did someone say SMS is addictive? Will this growth stop? The total annual revenues earned by SMS text messaging passed 100 billion dollars in 2007 according to TomiAhonen Consulting. Yes, that is as big as total Hollywood box office, total Hollywood DVD sales and rentals, total music industry revenues and total videogaming software revenues in 2007 - combined. By contrast the total messaging revenues of email and IM are less than 5 billion dollars globally.

And why is SMS text messaging so addictive? it is the most discrete (secret) form of communication and it is also the fastest way to communicate. It is preferred by kids in school attempting to cheat in class to busy business executives who need something more powerful than wireless email on a Blackberry. A May 2007 survey by 160 Characters found that 84% of active users of SMS text messaging expect a reply within 5 minutes !!!  On email we're happy to get a reply within 24 hours. On voicemail who knows if we ever get a reply. Like we've reported, even the Finnish Prime Minister says on his voicemail recording, don't leave me voicemail, send me SMS. Or how about the Finnish libraries who send alerts via SMS and the Finnish dentists who replace cancelled appointments via SMS. One in five London car drivers pays the congestion charge by mobile phone using SMS text messaging. One in two Helsinki public transportation user pay for the single tram tickets using SMS text messaging. Or the heavy users, 10% of British students thumb out 100 SMS text messages per day - in South Korea 30% of students average 100 SMS text messages sent per day. What is the global average? 2.6 SMS messages sent per day. The leading countries, Singapore users average 12 per day and phone owners in the Philippines are the world leaders averaging 15 SMS sent per day. Even laggard USA is following in lock-step with this growth curve and now USA cellphone owners average over 1 SMS sent per day.

INDUSTRY SIZE

So - it is the monster technology on the planet. A subscription for half of the planet. The total service revenues for mobile already passed 720 billion dollars last year (Informa Jan 2006). Toss in another 100 billion for handset sales and some more for network infrastructure and we're at a total industry of over 875 billion in size by 2007 - well on target to hit a trillion dollars as an industry by the end of the decade - then in size with the global automobile industry or global armaments industry for example.

NEAR FUTURE

Every year we've heard many pundits and so-called telecoms experts (often based in the USA) forecast that the mobile industry is about to get into a downturn, that the growth rate has to stop, by whatever distorted logic they may have used. I said in my second book, m-Profits, in 2002 that this industry was poised for enormous growth until the industrialized world would hit averages of 125% mobile phone subscription penetrations when measured per capita. Well, today 59 countries have passed the 100% per capita subscrition penetration (Informa Nov 2007). Hong Kong, Taiwan, Israel, Italy etc have penetration rates of 130% and above and European aveage penetration is about 110% (USA is at about 85%). I also reported in my book at the time that SMS was addictive, so there was no slowing this trend.

How long can it continue? I am not sure. But don't be fooled into thinking that somehow the "laptop" internet on wireless technologies etc will have any chance of catching up - much less than half of the world's PCs are laptops and the PC industry sold about 100 million laptops in 2007. The laptop industry is the very small little brother to the giant mobile telecoms industry, in users, in devices deployed, in revenues, in growth rates, in profits. Whatever measure you want. The PC and internet industry is focusing on broadband migration as its saviour. Even broadband is migrating to mobile now with so-called 3.5G technologies (HSPDA) led by countries like South Korea, Singapore and Hong Kong and about 20% of all global broadband internet access is already mobile as well (Informa 2008).

ALSO MORE INFO IN THOUGHT PIECE

So, that is my update to the status of the mobile industry for the new year 2008. If you would like a collection of the main thoughts and statistics and facts, in one of my Thought Pieces (a concise two page intense white paper) - please send me an email to   tomi at tomiahonen dot com  and I'll send you the free Thought Piece "Mobile Telecoms Industry Size 2008" in return (please don't leave a comment asking for the Thought Piece, its faster if you send me an email so I can respond directly to your email. But obviously we'd welcome any discussion about these numbers and this blog - as any blog we post - so please do post discussion comments here, of course). Oh, and if you have received the older Thought Piece last year, this is thorougly updated, so you may want the newer version.

Comments

[Krzysztof]

Hi Tomi,
Brilliant and colorful summary!
Pls send me Thought Piece and SMS about 3G in BCN
Cheers
K.

[Giff Gfroerer, i2SMS]

Tomi,

SMS in the United States is certainly picking up. Our company generated more interest in the final two months of 2007 then the previous ten months combined.

The iPhone has generated huge amounts of interest in the mobile arena, most likely due to the unlimited data charge thus allowing consumers to surf without the question being asked if they want the service or not. Their UI is nothing short then brilliant. Some pundits will lay criticism for the iPhone's use GPRS and the simplistic features. However, when battery life and an uninformed public are taken into account, the iPhone was exactly what the US needed to jump-start the mobile industry.

As always, thanks so much for your thoughts, comments and statistics.

[Dean Bubley]

Hi Tomi

Some great stuff here - SMS stats in particular are always fascinating.

But.... as always, we don't quite see eye-to-eye on everything.

I'm really unconvinced that "already 30% of all internet access in 2007 was exclusively from mobile phones" - I reckon the number's more like 5%.

30% of 1.3bn = about 400m - so where are they?

There's maybe 20-25m mobile-only users in Japan, 12-15m in India, 3-5m in China and hardly any in Korea. I can't see the rest of the world contributing another 350m given high PC penetration & usage in N America and W Europe. Most prepay users in emerging markets don't get Internet access, either. [I've put my sources/analysis on my own http://disruptivewireless.blogspot.com]

Curious about your thoughts on who these mobile Internet pioneers are...

Dean

[john]

i Think SMS remains a major growth area for mobile operators. However, revenue growth is only a fraction of the growth in messages

[Tomi T Ahonen]

Hi Krzysztof, Giff, Dean and john

Thank you for writing

Krzysztof - thanks. I've written to you already via email

Giff - Yes, I hear that a lot (SMS picking up dramatically in USA) from my various contacts. Its a very healthy sign for the American cellular telecoms industry to move past the essentially 1G era of "just voice". And yes, the iPhone has now strongly energized many parts of the American industry from telecoms to the IT industry to the mass media and advertising. I was there in October doing the keynote at Mobile Monday San Francisco and the energy and enthusiasm for mobile was clearly more than I've witnessed before in North America. That speaks of good things for the whole industry. We needed the "sleeping giant" to wake up to this opportunity.

Dean - I'll answer you separately next. It'll be a bit longer reply (as you no doubt knew and expected)

john - Yes, I agree. SMS traffic is growing at alarming rates (50% year on year) but revenues also grow, as you say, a fraction of that of traffic. SMS revenues grew by something like 23% last year, under half the growth rate of traffic.

Thank you for writing

Tomi Ahonen :-)

[Tomi T Ahonen]

Hi Dean

Ok. We've been doing this before at many venues and debates and discussions, but seriously, Dean, I think its about time you start to accept the facts. Every trend is away from "your position" and towards mine.

What did I say. 30% of all internet users access exclusively from mobile; 33% access from mobile and fixed; 37% access from fixed exclusively. Is this really difficult for you to accept. At least previously Dean you offer some facts and research to support your view, now you give none.

So. Lets start with some undisputed facts. 1.3 billion internet users. 900 million PC users. So right there we know there is at least 400 million internet users who are unaccounted for in the PC stats. And as you know perfectly well, this discrepancy between less PC users and more internet users has been growing this whole decade. And that mobile internet use started in 1999 when NTT DoCoMo launched iMode in Japan.

Some of the missing user numbers are from internet cafes and university campuses etc, but an increasing number of the missing numbers are mobile users. Not just in South Korea and Japan, but all over, from USA and UK to Finland and Sweden to Italy and Spain to South Africa and Brazil to Russia and Poland to China and India.

So lets dig a bit deeper. PCs. Most are desktop PCs, the rest are laptops. There is a trend to more laptops and older desktops are increasingly replaced by laptops.

PC access. Not all PCs are even connected to the internet. Yes, its an ever shrinking part of all PCs but still there are some, mostly in Africa, Latin America and other parts of the developing world, where given PCs are still stand-alone, used by smaller businesses etc. So of the total 900 million PCs, the actual internet connected population is (slightly) smaller than 900 million.

Then desktop PC internet access. Most of that is by fixed connection, dial-up or broadband. But there are service providers for example in India and Pakistan that offer wireless internet connections via mobile networks to home/office PCs. There is some of the internet access even for these modems that are mobile network based. And now with HSPDA ie 3.5G there is a powerful rival for broadband access, such as M1 in Singapore for example, who is targetting clearly the home broadband users of Starhub, with very compelling internet modem pricing. Similar services offered from Spain to South Korea on HSDPA for home/office internet use.

So of the desktop PCs, it has become possible to connect their internet access via the mobile networks a couple of years ago, and now with HSPDA there is a clear trend to connect ever more this way.

Then there are the laptops. On a laptop there is a compelling reason to get a wireless modem or data card to allow the internet connectivity to be as mobile as the mobility of the laptop computer. For example I've had my 3G data card for my laptop since 2003.

Which brings me to broadband connections. On Broadband there are limitations to the fixed landline telecoms network and numerous rivals from cable modems to WiMax to HSDPA. Informa just reports this January 2008 that 20% of all broadband internet connections worldwide are wireless. Obviously not all of those are mobile, but a significant part are. For example in South Korea the two rival wireless broadband offers are both on national coverage of WiBro (a variant of WiMax) and national coverage of HSDPA.

So out of the 900 million PCs, some (very few) are not connected to the internet at all. Some (few) of the desktops use a mobile network modem especially now HSDPA. And of the laptops a significant part has mobile connectivity as you yourself have reported as you've analyzed Vodafone's 3G data for example.

It is absolutely certain the total fixed internet connected PC population is less than 900 million - and the trend is away from fixed to mobile connectivity.

Dean, the trend is clearly AWAY from your position, even with PC based internet access, closer to my position.

Then of the mobile users. You say you can't find the users. You mention "hardly any in South Korea". Why? South Korea is - as you well know - the world's most advanced digital country where all internet connections were upgraded to broadband and now they already offer 100 Mbit/s high speed broadband as a standard. But also as you know, I released my fifth book, Digital Korea, last August (with a launch party in Seoul).

As we were meeting with the government officials I met up with NIDA experts who gave me their latest study which showed how dramatically the Koreans are moving away from fixed internet connections to mobile connections. NIDA's numbers for 2007 showed that 15% of all homes with internet access had only mobile access, while only 1% of homes with internet access had only fixed access. The rest had both fixed and mobile access.

I don't buy your position at all, that there are no mobile-only internet users in Korea when already 15% of homes have abandoned the fixed/broadband internet connection for a purely mobile internet connection.

I also think your Japanese numbers are too low as definitely are your Indian numbers. But that is not enough to make up for the numbers.

So, where do we start. USA and Canada are the most extreme region at one end - where PC penetration is highest, fixed internet connectivity has been around for the longest; while mobile phone penetration is the lowest, 3G mobile is lagging and mobile data plans are very nasty, etc.

Northern Europe is the next most "PC friendly" part of the picture where PC penetration is high as is mobile penetration. Southern Europe (Italy, Spain etc) is already much more mobile-friendly, where PC penetrations are severely below mobile phone penetrations.

Then comes Eastern Europe and Russia where the picture is already very distorted in favour of mobile. The Asian Tiger Economies, Australia and NZ and Japan come next. Then comes China. Then India. Then the rest of the developing world, Latin America, Rest of Asia and last Africa.

So if the USA and UK are not among the most eager to migrate fixed internet access to mobile internet access, even Italy and Spain will be ahead of those, with Russia, Australia, China, India, Africa etc far far ahead of those migration numbers.

So lets explore the big picture. There are 3.3 billion mobile phone subscriptions. The numbers from Real Networks (an internet company, not a mobile telecoms company, by the way) say 25% of all mobile phone subscribers access the internet on their phones. Out of 3.3 billion that is 825 million.

Oh, its important to mention that a mobile phone user will not connect to the internet via a wired connection. So all of the mobile phone users who do access the internet on their phones, access only wirelessly, where a growing minority of PC users do access wirelessly.

If there are "less than" 900 million PC users who access via fixed internet; and there are 825 million mobile users of the internet, then in rough terms the one third/one third/one third split of "only fixed access, both fixed/mobile access, and only mobile access" is reasonable (by mathematics, not by actual usage obviously). And this split should be slightly skewed in favour of fixed connections and not in favour of mobile connections. My numbers of 37% fixed, 33% both and 30% of mobile are very consistent with the Real Networks numbers.

Why is this important. If we can trust the two numbers, 900 million and 825 million, we cannot have an overlap of say 700 million. Because it would leave 200 million PC only, and 125 million mobile only, and the total user numbers would max out only at 1,025 million internet users. To get to 1.3 billion, the overlap between PC/fixed and mobile has to be about 430M/430M/430M, plus-minus of about 50 M each.

Then lets see if there are any stats to support the Real Networks 825 million mobile internet user position?

Lets start with installed base of devices. 3.3 billion subscriptions equals 2.6 billion actual unique mobile phone users (according to Informa 2007). Thats 2.6 billion definite phone owners (some of the multiple subscriptions have only a SIM card). But today 2.4 billion of those phones are GPRS capable or faster. So out of all phones that have reasonable speed and price internet capable phones, 825 million is only 34%.

I do not have a study of how many GPRS capable phone owners do use mobile internet on their phones, but I do have a study on 3G phone owners. TNS reported in October 2007 that about half of all 3G phone users use the internet on their phones. Out of 400 million 3G phones in use today, thats about 200 million for you. If the "non-3G" unique phone owners are 2.2 billion, and out of those, 2 billion have "non 3G" GPRS/2.5G phones, and we're looking for 625 million users of the internet, thats 31%.

If 50% of 3G users access the internet, it is reasonable that a smaller percentage - but much more than zero - will also use the internet. For example iPhone users are all 2.5G users. And as you know, recent data reveal that the biggest traffic to Google before Christmas came from iPhone users.

So I would argue the 825 million user number is reasonable as a percentage of 400 million 3G users and 2 billion non-3G users with GPRS or other 2.5G phones.

But lets find more facts to support the position. Is Real Networks' 25% number reasonable and within reported stats? Telephia reported in May 2007 that 19% of British internet usage and 17% of American internet usage had migrated to mobile access. (Note that these are the countries at the PC-end of the spectrum) This is consistent with earlier research by Comscore Matrix which reported in Octber 2006 that 34% of Italian and 26% of Spanish internet users accessed via mobile. A related study of Asian users by Mindshare in six Asian countries found that 49% of Asian mobile phone owners download content to mobile phones. That is obviously not the same as internet use, but illustrates that the Asian mobile phone owners are eagerly adopting mobile data services.

These three studies all are well in line with the Real Networks number of 25%.

Then again, the trends. The number of PC sales has been stagnant the past two years but has grown dramatically for mobile. PC devices are migrating from desktop to laptop (more attractive to mobile data cards) and all internet migrating to broadband where now HSDPA and other high speed mobile networks are competing with broadband providers both on data capacity and data pricing, very competivitely. The trends on the PC side favour my position, and are totally against your position, Dean.

The on the phones. Mobile phones have become much more friendly for internet use, with large colour screens and ever faster speeds. Mobile networks have reduced prices of data plans and increased mobile internet speeds. Content owners offer ever more compelling content to mobile internet. All trends suggest mobile internet use is growing dramatically.

But finally, I want to point out a perception issue. If you take Flirtomatic, the service offered in the UK both on the web and WAP. If you ask a web user of Flirtomatic, do they use the internet, the user will definitely equate Flirtomatic web use as internet use. But if you ask the Flirtomatic WAP user does he/she use the mobile internet, most would not think of it as such. Perhaps some think it is WAP use (many would not) and more likely they would understand it as use of an operator's portal services, such as Vodafone Live. Even so, many would not select any of the above, and only understand that they do use Flirtomatic on the phone.

So some of your doubts about mobile internet use, Dean, may be that many of the users themselves dont' even know. Take WAP Push for example. In many cases where a user receives a WAP Push type of SMS text message, and they respond to it, they don't even know that they have moved from SMS to WAP. The network knows. But the user easily thinks that the user has just been involved in SMS.

Dean - you attack these numbers often. You have previously given some numbers to defend your position. But now the data is becoming overwhelming. Clearly there are about 800 million who access the internet via mobile and under 900 who access via PC (of whom some access via mobile). That means that we are very near the point when more access is from mobile than fixed.

You don't seem to like that fact. But all trends are in that direction and you know it Dean. Where exactly are the 400 million? I can of course do a total country-by-country analysis but I doubt you'd pay for my time for that, ha-ha..

1.3 billion internet users. Under 900 are by PC. That means over 400 are something else. Some of those are from internet cafes and university computer labs. Those were estimated to be about 50 million two years ago. 825 million are mobile phone users. You do the math. The numbers in the middle who are mobile users are something between 300 million and 500 million.

Thanks for writing

Tomi :-)

[alex]

Tomi's top-down and Dean's bottom-up analysis for the "mobile-only Internet users" differ by an order of magnitude, so something is not right.

So let's do some math.
The 825 M figure proves only how relatively few "PC only" Internet users there are: 1300 - 825 = 475 M
It has no impact on the "Mobile only" figure, which was disputed by Dean. So no point to argue about those 825 M here.
"Mobile only" is given by about 1300 - 900 = 400 M, so let's look at the 1300 and 900 M figures.
1300 M "Internet users" seems quite reliable and undisputed.
Tomi lists factors why PC Internet users are less than 900 M.
What I am missing is family usage of household PCs. Isn't this is a strong factor also in Western countries? E.g. young children often have own mobiles but use Internet on family (and school) PCs.

How many PC and persons do we have per household?
Maybe Tomi or someone has better figures. I'll just make a guess for Western countries (which skew PC statistics) to get a feeling for the numbers:
300 M housholds
2 PCs per houshold
3 PC Internet users per houshold
-> 900 M PC Internet users with 600 M PCs
Let's count the remaining 300 M PCs with 1 Internet user.
-> all together 1200 M PC Internet users
-> 100 M "mobile only" Internet users

As said, I'm not claiming this is a substantiated estimate, but it seems to me that shared family PCs could explain a much lower figure for "mobile only" Internet users as Tomi estimates.

(Tomi, this is not about whether there is a trend from PC to mobile or not - I am just trying to establish the current state of affairs, without judging a trend.)

[Tomi T Ahonen]

Hi alex

Thanks, good points. But by that same analysis, we get also the dual use by one PC internet user. So the family has for example 2 PCs and 3 users in your example. But the dad also uses a PC at work (is one user on two PCs). The child has a PC at school (is also one user on two PCs). Maybe the mother is the only one of the family for whom the family PC is the only PC.

I see where you're going, but then the absolute math won't square up. If there are 1.3 b total users and about 800-900 M each who access from a PC and who access from a mobile, then there will be rougly 300-400 M who must be PC only, and another 300-400 M who must be mobile-only.

I took a quick look for another debate we have with Dean about this same topic at Forum Oxford, and dug through a dozen countries and their mobile user internet access. These are from the 2007 Netsize Guide quoting 3Q 2006 national statistics by Informa. The numbers range from 71% in Austria (Mobilkom Austria) and 67% in Denmark (TDC) to 19% in Poland (Centertel) and 21% in Switzerland (Swisscom) in Europe. So of all European users between about 20% and 70% of all mobile phone subscribers are already active users of the mobile internet. Remembering that European mobile phone subscriptions already exceed 110%, we do get many countries in Europe where the mobile internet user numbers are greater than total PC penetrations in those countries.

Then if we look at only real developing world country mentioned in the Netsize Guide. In India mobile internet use ranges from 3% with Bharti to 59% with Reliance. It makes perfect sense that in the developing world where you have a very hard time getting a fixed landline even connected, of course most internet users will connect wirelessly, mostly through mobile networks. The ratio in India is about 6 to 1 more mobile than fixed landline connections to the web. This pattern repeats all through Pakistan, Bangladesh, Indonesia; all of Africa; all of Latin America.

alex, I am very confident in these numbers. The big picture numbers bear it out - I've been tracking these stats and they are consistent with both a top-down and a bottom-up analysis. And the "balancing math" how much is overlap and how much are users only on one or the other method, they too are reasonably consistent.

Of course I'd love to see some big industrial analyst company come up with total counts worldwide. But even looking at what Informa was able to provide for the Netsize Guide for 2007, in many countries they didn't have the mobile data user numbers or if they did, at times it was numbers that are pretty meaningless for this purpose.

Thanks for writing

Tomi Ahonen :-)

[Tomi]

Tomi

I think there may be some middle ground here.

And I think that it is around the difference between:

- accessing the real Internet
- using a browser for any network-based application

I've tried to do 2 sample matrices for these, breaking down total user numbers by device type(s) and access mechanism(s). Also I've split out WiFi & cellular as I suspect that consumers surveys may confuse people about the difference between "mobile" and "wireless"

http://disruptivewireless.blogspot.com/2008/01/mobile-vs-fixed-internet-users-2-its.html

I can quite believe that 25-30% of total browser users are phone-only, or phone-mostly, because of walled gardens. But in terms of 'real' Internet access, I reckon it's more like 11% who are "phone-primary", and lower still for people who _never_ use a PC.

Dean

[Andreas - news of the future]

Do you have a reliable reference for those numbers (e.g. number of mobile phone subscribers?

[Tomi T Ahonen]

Hi Dean (posting as Tomi? Are we schitzophrenic, no we are not, ha-ha) and Andreas

Dean - yes, I think we really do see this very much in a similar way on the big picture, and the argument is more on the semantics and precision - which I do admire in your careful approach to all things relating to our industry and its metrics.

I fully acknowledge that for "my numbers" to hold, I have to include WAP usage (and iMode etc). Then we are in the realm of walled gardens. Not that all WAP is walled gardens, but most still currently are.

So if we take the view of "classic" existing TCP/IP internet use via a mobile network connection, the 11% number seems very solid. But yes, if we include the WAP users, then the larger user number is valid.

I also appreciate it greatly Dean that you do the sanity checks etc. And I think I could be more precise, to mention WAP in my numbers so it is clear what definition I use.

I know I've read your blog entries about the numbers at Disruptive Wireless and at Forum Oxford. I was somewhat distracted as the Sprint Nextel story breaking this past week, so I only had time to respond at Forox, but I will be coming to your blog as well to continue the dialogue there as well.

Andreas - the total mobile phone subscriber numbers were from Informa, two major reports last year, the multiple SIM card report which came out in the spring and the 3.3 B number and related stats which came out end of November.

Thanks for writing

Tomi Ahonen :-)

[meera]

How would you say the Indian market is faring? Could you tell me if mobiles are eating into laptop sales - by how much? Any statistical evidence to show this trend?

[jaspreet]

Dear Tomi,

GREAT WORK !!!!

Can you please email me your thought piece for the "Mobile Telecoms Industry Size 2008"?

I am really interested in reading it.

Great Thanks Tomi.

Kind Regards,
Jaspreet

[Erika Beede - Goomzee]

These are GREAT statistics. Thanks for sharing your findings!!!

[Darren]

This was a very interesting article Tomi and reinfoirces just why brands need to start incorporating mobile as part of their marketing mix. Its no longer looking at the potential mobile has to offer, it's now looking at the great opportunity!

Darren

[Pravin Tamkhane]

Hi Tomi,
I must say, it is the brilliant post, great analysis and you really made us to look at the BIG opportunity!

Pravin

[Collet Kudze]

Tomi,

A lot of insight in this blog indeed.
Am just wondering if you have any statistics for Asia , Middle east and Africa.
It would be interesting to see how the digital divide reveals.

Collet Kudze

[Tomi T Ahonen]

Hi meera, jaspreet, Erika, Darren, Pravin and Collet

Thank you all for the kind words. I'll respond to those with specific comments or questions.

meera - Good question (mobile eating into PC sales), we don't see global numbers yet of that. The only local market to report that was Japan with 2006 numbers, by which for the first year ever, PC sales declined in Japan while phone sales kept growing. I am expecting some advanced markets (like South Korea, Italy, Israel, the Scandinavian countries) to start to show that kind of trend soon, perhaps even from 2007 data, probably from 2008 numbers. I'll blog about it definitely when such numbers appear (if they do, ha-ha).

The overall sales of PCs have been rather flat the past four years, but in 2007 the PC industry expected growth, so its not "game over" by any means anytime soon. And there is of course within the PC industry the shift away from desktops onto laptops as the form factor, and the expectation was that perhaps in 2007 we would have the first year ever that more laptops were sold than PCs worldwide.

Darren - yes, it seems that the marketing/advertising industry has woken up to mobile over the past year or so. I've had several good workshops with advertising and media companies specifically on that dimension over the past 18 months. There is a buzz about it, ha-ha.

Collet - About the digital divide. Good question. There are numbers, but it gets awefully difficult for me to gain access to all that data for a timely analysis (I don't have a deep reserach budget to buy all the expensive reports from the industry analysts), in particular as the develping world is growing so incredibly fast in this space. I do occasionally post about the digital divide related topics but more on a country-specific or industry sector-specific way.

I might be able to do a posting about the issue a bit later, as I do obviously have a lot of customers in Africa, Latin America and developing countries of Asia. There was a great article about it in the Economist I think a week or two ago, and I'd definitely agree with their summary that its a chance to leap-frog technologies, going straight from no telecommunciations, to cellular (and digital) mobile telecoms, without the fixed networks and analogue mobile phones that the Western World built earlier. Some of by best examples of industry disruption comes from the developing world, such as the mobile bank accounts in Kenya or paying your utility bills by mobile in India etc.

Thank you all for the warm sentiments

Tomi Ahonen :-)

[Tilly]

Hi Tomi,

I think you're research very enlightening - do you have any views on the growth in SMS advertising? I am doing my thesis on this and would like to know your thoughts.

Kind regards.

[Tilly]

Hi Tomi,

I think you're research very enlightening - do you have any views on the growth in SMS advertising? I am doing my thesis on this and would like to know your thoughts.

Kind regards.