Get me - ever the professional, I was on holiday today but nonetheless wandered up to the NOC/160Characters event on I-mode - and I'm glad I went, it was ace.
Mike Grenville chaired it ably - I'd not attended one of these before but particularly liked the audience introducing themselves before everything kicked off. And the presentations - by a guy called Wojtek from DoCoMo whose surname I missed, and Jag Minhas of O2 - were interesting. I suspect I'm more familiar with I-mode than most people (having read everything I can find out about it for the last 6 years) but there was a lot of info here that I'd not come across before. And the whole event had a relaxed, chatty feel to it. Both speakers dealt amiably with an audience that was at times sceptical (I have to confess I don't quite follow some of the models for how I-mode is way better than alternatives, despite secretly thinking that this might in fact be the case).
A pox on Brighton - I have a very early start tomorrow so dashed out at 9 to get the 9:30 back. Notes follow... my comments are in italics.
Disclaimer: he's not in the team specifically supporting UK I-mode launch.
Before coming to UK, he spent 8 years with DoCoMo in Japan; he worked there during the launch, sourcing English-language content for I-mode and 3G before they existed.
Now: 45m subscribers in Japan, 7m outside.
(Is I-mode a way for DoCoMo to claim subscribers and revenue outside of its own customer base? What's the commercial model for licensing I-mode, I wonder - is it in any way tied to the success of I-mode in an individual licensees territory?)
90% of DoCoMo customers are active users of data services; this makes them different from other operators.
80% of I-mode content providers are individuals or small venture companies. There are 85k web sites accessible via I-mode.
Service started with 67 content providers - it was tough convincing the initial ones to invest. Initially large companies in Japan: banks, transportation, entertainment, etc, then they tried to build up more international content. The more content they had, the more appealing they were to users... and the more users, the more attractive they became to content providers.
Official content providers: linked from portal menu: 4642 sites.
Unofficial content providers: accessible via address, search engine, or (increasingly) via barcode readers in handsets linking through to sites. 81,979 sites.
There is free and paid content in both official and unofficial content providers. Most content in both is free - users pay traffic fee but not subscription. So why provide content? It's an "internet business model": it's promotional. Transactions can be done via credit cards or pre-existing accounts. The speakers both went on to emphasise that I-mode isn't a general billing system for goods and services - but DoCoMo have done a lot of work in this area in Japan. I wonder if O2 have plans to adopt any of the payment schemes DoCoMo have rolled out (presuming they're successful)?
DoCoMo provide billing for content to official providers. For 300 yen (3 euroes) a month, users are connected to I-mode in the first place. Plus traffic fees: 0.3 yen (0.3 euro cents) per 128 bytes of traffic. (that's 2.4 cents per kilobyte, 41kb per Euro - about 25 euros per megabyte! A later questioner got this restated as 0.3 euro cents per 128K which seems way too low - 2.4 euro cents for a megabyte?). Plus monthly subscription fees. DoCoMo cap single subscription fee at 3 euros. Why the cap? Content providers wanted to charge too much and DoCoMo felt they would restrict growth of the market unless they were held back.
Bandai started working with DoCoMo very early. At peak they were making 7m euros per month from a single server and 2-man staff.
Some say this success is only due to DoCoMo's dominant position and unique characteristics of the market - but no. This business model can apply to any size of customer base.
3 simple things which were the essential success factors behind I-mode:
1. Reuse of existing internet standards: subset of HTML. (disagree here myself but it does make a good story when you're persuading content providers to get on board) I-mode creators came from outside the telecomms industry and didn't understand the standards.
2. They never bought content, signed exclusive deals, etc. They allowed all third parties to come to the platform, provided an incentive and let them compete with each other. (I'd not appreciated this, but one of the things I took away from the day was the hands-off approach DoCoMo take when it comes to multiple competing content providers)
3. Appeal to "normal" people. They never advertised I-mode as the wireless internet; it was always going to be a different experience to the Internet people previously knew. Sold on the benefits.
Worked in the Internet business for last 10 years - mainly on the fixed side. (ISP?) In late 1998, he worked with Cellnet who were embarking on a Freeserve-style venture which later became Genie.
Documents best practice etc for all data products in the UK - tending to concentrate more on the R&D side of the business.
I-mode is the next step in driving data revenues for O2. Good vehicle for acquiring cusotmers.
WAP has been more successful in the UK than in other markets across Europe. O2 Active generates 50% of all non-SMS data traffic in the UK, with 1b page impressions pcm.
O2 active: youth, mix CP/operator branded, some paid-for content, event-driven billing. Brand experience. No control over specs of devices.
I-mode: mass market, multiple brands/category, no paid-for content, subscription model. Supermarket-like experience. Validated devices. 86/14 split which accounts for a coverage for risk and fraud which doesn't existing in Japan (?)
I-mode will not replace O2 Active. O2 Active is important right now because of the scarcity of devices which support I-mode. I-mode is the "next step". This is a very consistent story coming out of O2 at the moment, but I have to confess I'm sceptical: having 2 mobile Internet services must be confusing to a customer-base who are only beginning to get the concept in the first place...???
O2 have a license agreement from NTT DoCoMo - licensing the brand and know-how.
Q: how are PRS etc paid for ringtones, given that they're sold on a subscription, not event-driven basis?
A: Not sure.
Q: How can O2 manage a huge number of content providers?
A: They have 100 today. From a personal curiosity point of view I'd like to know how they scale up to thousands - this is probably somewhere that UK operators can learn a lot.
They're launching in Ireland too.
I-mode is differentiated by an open-access philosophy and a wide breadth of content. There won't be a 2-3x charge for going off-portal.
Q: Does part of that 14% go back to I-mode?
A: No - all of it stays with O2, along with messaging and traffic revenue.
Q: Are there any alternative billing methods?
A: No. Premium rate IVR is popular in Europe; SMS could be used; credit cards.
Popular services on 50/50 rev shares are ringtones and downloads.
On I-mode lifestyle services become more popular.
Assumption: customers naturally want to spend some of their disposable income in data services. Jag mentioned a figure of £5 per month.
I-mode tends to favour a value chain with minimal intermediaries. I-mode enables producers of content to start selling direct.
UK data pricing is lower than most of the rest of the world.
Q: What are you aiming for in terms of subscriptions? Who's going to be on the portal and what sort of transactions will be done?
A: A broad selection of content providers; magazine-subscription style. It's not about operators collecting money for hard goods.
Q: How long a life does I-mode have with web'n'walk, 3G, etc? How big a percentage of O2 customers will go for I-mode?
A: On the second one, it's too early to tell. Web'n'walk is about internet content over the air; I-mode is about content provisioned for mobile. There's no differentiation between 2G and 3G - it's all I-mode. Underlying network matters only to the extent of how fast you can transfer data.
Q: How does mobile TV fit into I-mode? Will there DVB-H, MediaFLO handsets?
A: Maybe. They way operators make money on mobile TV will not be per-megabyte.
Q: How do CPs get to the top of the menu?
A: It's first come first served for now. At some point they'll move away from this to formula-driven approach, where category orders are refreshed weekly/monthly, and ordering will be based on something tracking usage. There won't be paid placement.
Q: What about roaming?
A: All works fine. Regular content can be accessed via roaming fee. Unofficial content doesn't involve paying a roaming fee. (?)
Q: Why is traffic revenue not shared with the folks who can generate so much?
A: There's no parallel in the fixed world for traffic sharing. (but this isn't a good analogy as broadband traffic isn't metered). The industry has learned from the Crazy Frog debacle and there's a lot of hand-holding, reminders, etc. with easy access to subscription listing and easy cancellation. They mandate every content provider must provide something for free.
Q: Why did WAP banking fail?
A: The issue was security. Banks wouldn't go for the WAP 1.0 model with protocol conversion. With I-mode security is end-to-end.
I-mode does page caching which makes a big difference.
There's full connectivity between email on I-mode and the fixed internet.
I-mode users on some European operators have seen an increase in ARPU on voice and text too.
Q: How does the purchase cost of an I-mode handset for O2 compare to that for a typical European handset?
Q: How will I-mode protect its customers from unofficial content providers who may not be scrupulous?
A: There is an upper limit on PSMS per day at the moment - so O2 already enforce a degree of protection. There are other mechanisms too.
Questions I didn't get around to asking
- What does Wojtek think of O2 wireless advertising ("faster than WAP") given I-modes avoidance of tech language?
- How do you account for the use of "credits" on existing UK I-mode services, where your subscription buys 20 points - effectively a currency overlaid onto the subscription model. Is this a weakness in the subscription model? Is it something we see in Japan?
- How does cost of purchase of I-mode handsets compare to European ones?
- Are there really 4642 sites linked from the menu? How is this architected such that it's navigable or understandable? What techniques are used here?
Tom,
As you know, the best source of information about the total i-mode ecosytem is at http://www.imodestrategy.com which I coauthor with Pascal Lorne - the Founder of the i-mode Content Forum.
I can answer some of your questions, and perhaps muddy others.
1. On Charges. A packet is 128bytes and from this chart of charges [http://www.nttdocomo.co.jp/english/p_s/imode/charges/standard/home.html] in Japan you will see that on 3G 1MB could cost about 17 euro and lower on plans. But of course this is way flat-rate is so popular and has changed the content landscape and user behaviour. I once asked the Director of research for DoCoMo's 3G network what would happen, at the time just when they introduced flat rates, if every went crazy for downloads, and he said "we don't know". I thought that was a good answer.
2. On Site Numbers. From my more than one year old notes from my last meeting in Japan with DoCoMo I noted over 8800 official sites and 80,000+ unofficial sites and the comment that the unofficial sites were the fastest growing in numbers and revenue generation (for the paid site owners). About 50% of official sites were free.
3. On Subscriptions and Charging. DoCoMo set a maximum subscription fee for the reasons you quote but no minimum - there is no set fee. The maxiumum was set at the price of a typical monthly magazine, which was seen as the amount that people would pay without thinking about it. The subscription model was carefully chosen and is delberately promoted as setting the i-mode model apart, one of the rationals being that to get someone to buy on a pay per service basis takes as much marketing effort and money as getting then to make a monthly subscription. And you only have to get them to make a monthly subscription ones. DoCoMo said that 50.6% of their i-mode subscribers are actually inactive - meaning that they have subscriptions for which they are billed each month but they have forgotten or choose not to use the service. This is a HUGE cash cow from a subscription service
4. On HTML and Creating i-mode Sites. The choice of a subset of HTML (Compact-HTML or cHTML) was very deliberate, in the face of the largest telco research and engineering group in the world. There would have been overwhelming pressure to go with a telco solution, designed by engineers. The i-mode team resisted this, not because they did not understand it but that they knew the customers didn't understand it - both the content providers and the subscribers. They succeeded in introducing a system which consistently is praised by site developers and content providers as taking around 60% less time and effort to implement a site than WAP, and is able to reuse internet web development skips. As well and importantly i-mode was implemented on a 9600baud network and cHTML and the network protocols ensures light loads and good response times, with less expensive equipment and far less network control traffic than WAP. DoCoMo already had an extensive and well used packet network, used for the juen-mail - 10 cent email system. And you're right about the security bit, which is why banks like Citibank quickly and with confidence build i-mode sites that allow transfers of money between accounts etc etc but refused have sites on earlier versions of WAP.
5. O2 and Active Portal. I have a recent quote [http://www.imodestrategy.com/2005/10/_051002_weekly_.html#51002-3] in which O2 state that they will not spend any more money on Active and will phase it out. I also cannot understand the dual portal strategies of carriers, such as Telstra, where in a single market they promote two portals to a confused public and also a confused sales staff (I experienced this today).
6. DoCoMo's Licensing Model. DoCoMo's earlier attempts were through the quaint idea of owning a minor piece of equity and expecting to get some influence over major strategy decisions. We know that Western enterprises and governance doesn't work that way and DoCoMo lost about $10 billion. Deals with first movers such as KPN were most likely license fees only. DoCoMo more recently, such as with Telstra, has a "clip the ticket" on i-mode revenue - a small part of high volume targets. And if carriers do not meet their subscriber targets the country exclusivity (but probably not the i-mode license) would be at risk. There is also a 3rd silent partner in Mitsui Corporation the trading house. Since DoCoMo is not an international company and hardly a sales company they had huge difficulties firstly understanding how to and secondly making any sales overseas. On the other hand Mitsui makes literally 100s of $$$ billions of sales. Mitsu said "why don't se sell i-mode and treat each item of i-mode income as a lump of coal and we will take a minute fraction of a trading commission?" - and DoCoMo said "OK, please do". So under the surface Mitsui have made most of the recent sales, with the exception of the UK where DoCoMo had some experience. The O2 sale is also different because DoCoMo were desperate and O2 may have nothing but a license fee and not a clip of the ticket. But in any case the clip of the ticket explains a little of the increase in share that the overseas carriers take of the total fee. Whereas in Japan it is a 91:9 split in the CPs favour, outside Japan it is typically 85:15 - reflecting increased financial carry risk, commissions to DoCoMo and carrier regret at not being able to get 40%.
7. Miscellaneous. Menu structure is an issue and in the classic DoCoMo systems those who succeed succeed by being rewarded with better menu spots, but the whole issue is one of debate and people are trying to figure out how to keep the "i-mode experience" and manage the menu structure. Web'N'Walk delvers that whole question as a open question to the browser community. The comment about Web'N'Walk being "internet content over the air" while i-mode being content provisioned for mobile is interesting because Telsta's whole i-mode advertising and PR catch-phrase has been "just like the internet on your phone"! Data roaming for unofficial sites is "free" because you simply enter the URL into the browser. i.e. there is no "roaming" per se, unless the carrier slips on an artificial charge because they know that you are not in your country of origin.
The challenges to i-mode are out there, and new models will threaten its growth both in Japan and abroad, but is has a good track record and is still a thought-leader in many respects. The greatest challenge outside Japan, in my experience, is in the telcos' tendencies to (1) adopt a half-pregnant model, and (2) in the sales process, from beginning to end - in which we witness the absurdity that even where a customer actually asks for an i-mode handset that they walk out of the shop not knowing how to activate it, how to use it, and with no subscriptions and even no obvious numbers to call for help.
Cheers, Walter
Posted by: Walter Adamson | November 26, 2005 at 12:15 PM
Tom, will offer some of my own thoughts/responses to your questions and also some of Walter's responses. Starting with thoughts on Walter's responses:
1. The link that Walter supplied for DoCoMo's Japan i-mode charging appears to be broken. Here is a good one: http://www.nttdocomo.co.jp/english/p_s/imode/charges/standard/index.html
The undiscounted packet charges are indeed around the 18 Euro per megabyte mark as Walter says, but if you pay monthly "packet-pack" charges - then the price per packet drops (note these "packet packs" are not inclusive packet bundles - they are a way of reducing the price per packet by paying a monthly fee). So - by way of an example: in order to get a per-packet price equivalent to about 3 Euro per megabyte - you have to pay a "packet pack" charge of about 21 Euro per month. (Didn't I say that the packet charges in the UK are amongst the lowest in the world? :-) )
2. Another thing to add is that in the mature i-mode markets (e.g. Japan and France) there is roughly 50-50 balance of user *traffic* between official and unofficial sites.
3. Agree with Walter on most of his point about subscription vs event cost vs smoother revenue profile encouraging greater loyalty etc. - that's exactly what I was trying to get across in the presentation. Regarding recurring revenue from inactivity - I think this is probably too negative a way of looking at it - it's no different to the Sky subscription example I gave during the meeting - I pay Sky some money every month regardless of whether I consume - and, for me anyway, most months I am not a heavy consumer of TV through Sky - but some months I might be - and the subscription gives me this comfort that I can get at it when I want to. Compare this with pay per view - which Sky will tell you that they have had very little success with - and I certainly wouldn't want to consume my TV that way either. Personally I think customers are clever enough to cancel subscriptions if they feel that they are getting little or no value out of it - so I don't think that this is mainly revenue being derived from inactivity and forgotten subscriptions. Well it certainly won't be in the UK anyway! (In O2 UK i-mode we are going out of our way to make it very easy to cancel a subscription- and we are reminding customer every month for every subscription that their subscription is being renewed by sending them a message.)
4. on Markup. Agreed. I personally worked very closely with the banks that actually did try on WAP: Cahoot and Halifax for example in the Genie days (and we did actually have them on our WAP portals) but it was extremely hard work - involved lots of auditing - and eventually the services were withdrawn due to lack of usage - and lack of will to improve them ongoing.
5. We get this assertion about O2 Active and i-mode all the time - and yes you are right Tom that we have given a very consistent message about how the two services are absolutely key. The "O2 will dump O2 active" is a complete myth - and I think I spotted that dodgy story circulating after it was first published in International Herald and Tribune a couple of months back. I have been at almost every journalist briefing and trade/developer press interview about O2 i-mode - and we have never stated this. And we wouldn't either. O2 Active is a very successful and popular product - accounting for about half of all non-SMS mobile data traffic in the UK (not just O2's network - but all UK) and generating in the billions of page impressions per month - putting O2 Active portal in the top ten of UK websites. It is almost certainly the most successfull mobile data service ever outside Asia. We are not dumping this success. O2 Active is the branded portal offered to our customers who buy from our very extensive handset portfolio - most of which you know is currently non i-mode. So long as a significant number of handsets in the shop are non-i-mode - O2 Active will be available. Many of the content partners in our i-mode lineup are also available in some way shape or form in Active - especially the info and entertainment ones - but obviously using a different charging model. And as you know - for any successful portal service - it needs continued investment if it's going to maintain success. i.e. keeping it fresh and sticky etc. So I hope that puts paid to the debate around O2 Active and i-mode. As for confusing users - yes that does represent a challenge - but again I don't think it's hard to confuse a user about this in a market where the choice of *device* still is the key factor in the purchasing/upgrade decision. In other words - "if you buy this device then you get this service - if you buy that one you get i-mode" - our store staff are competent enough to uphold the differentiation - and our customers are hopefully intelligent enough to understand the difference. If, however, we find that people are making their purchasing decision on the content as the main criteria - then that's a different world altogether - a world that we'd love to get to - but the reality is that we're not there yet - and so a combination of strategies in pricing, marketing and portfolio/channel management will have to be applied in order to get "data service" to become an important factor in the customer's buying decision. The i-mode content model clearly represents a better business model for portal/content (as witness in other i-mode markets) - and we would like for our customers to benefit from that too (i.e. higher quality and better value for money content) - but right now we must offer the customer choice around the things they value the most.
6. I don't really have any comment on Walter's thoughts on Licence Agreements etc. The Mitsui thing something I know nothing about either. The 86:14 split that we are applying is to cover costs of billing and bad debt. Just elaborating on the latter for a moment: bad debt (in which term I also subsume fraud for now) is something that the carrier typically bears the risk of in Europe. By this I mean for example that if a customer subscribes to a Content Provider on O2 UK i-mode - but they don't pay the bill at the end of the month - then O2 still owes the Content Provider the money. So O2 bears the risk of the bad debt (or fraud) - not the Content Provider. As I understand it - the legislation in Japan is different regading this type of service - and so I believe that DoCoMo do not bear that risk. And, of course, if you don't have that risk - then you don't have to pass on the risk into the cost line. And typically the risk cost for bad debt in most textbook financial business modelling is around about 5%. Therefore - outside Japan - 9% cost becomes 14% cost. And that's exactly what it is for us in O2 - it's a *cost* - not a commission cover to DoCoMo - and not a "regret" charge for not being able to get 40% (can't comment for other operators) - we want our Content Providers to succeed in i-mode - because if they do - then we do too.
7. When the menu structure gets deeper and more complicated (as it is in Japan) then there are several other things that rise in importance terms: 1. What's New? 2. Search and 3. Personalisation. The first 2 have become critical in Japan - Content Providers see enormous value in What's New placement - and customers go to it a lot. And i-mode has spawned an entire new "search" industry in Japan - mostly based around magazines and directories. We have had a lot of experience with Personalisation in O2 Active - and we would want to see what aspects of these can cross over into i-mode without harming the business model. Yes I think there will be a lot to learn - but thankfully we have sufficient time to plan and do it properly.
Regarding roaming - I think several people were getting confused about this when it was brought up the other day. Basically - if you are an O2 UK customer of i-mode - and you take your handset overseas - then you will be connecting to the i-mode platform in the UK for everything - whether you go on-portal or off-portal - it doesn't matter. All data traffic will be charged at data roaming rates.
And finally some answers to questions in your posting:
The guy from DoCoMo - his name was/is Voytek K. Siewierski.
You said we are launching in Ireland too - just a minor point: we have already launched in Ireland, the Irish launch took place one week after UK in early October.
The £5 a month I mentioned was derived from the Mobinet 2005 survey results - which intimated that customers are (at the moment) placing a ceiling on what non-voice services they consume on their mobiles. So - if this is the case - and if this acceptable ceiling is about £5 a month - then as per my worked example of the i-mode revenue splits - in an event-charged world you are at most only going to get the user to download ONE £3 game per month. In i-mode - you are more likely to get the customer to spend the full £5 - because of the subscription value-for-money element of the model.
The "credits" overlay on the subscription fee acts as a way of limiting what you can do for the subscription fee - i.e. x games for £3 a month - as opposed to "as many games as you like". I don't really see a problem with this approach so long as the value-for-money per game (or "thing") is better than if you were to buy that thing ia non-subscription world. I'm fairly certain this approach is used in Japan - but I'm guessing that it is restricted to download objects - e.g. games, ringtones, wallpapers etc.
I can't see any reason for the cost of i-mode handsets to be any different to non-i-mode ones. There may be some R&D cost to be recovered - but I would expect that to be quite low is it's mostly a fairly straightforward software porting exercise (and only really for browser and email applications) - and some button changes.
Best regards - Jag
Posted by: Jag | November 26, 2005 at 02:56 PM
Guys
First off - thanks for hopping in. It's great to get such informed feedback.
I guess I still have a few questions:
1. There are lots of official sites on the DoCoMo menu. How do DoCoMo manage a coherent menu containing 8,800 official sites? I'm interested in the mechanics of doing this: from this perspective I-mode seems way larger than any European portal I've seen so far. What changes as you grow to this size?
2. Traffic charges do seem to be very high in Japan (based on 128 byte packets). Will this charging be replicated in the UK? The charge-per-mb stills seems high to me, perhaps I'm not understanding the model though.
3. Sorry Walter, but I don't agree with your analysis of cHTML; the writing of markup language is a small percentage of the overall effort of putting together a site, but we don't see that cHTML is necessarily quicker to work with than WML. It's not any slower, either - we just don't see a difference. You still need to test on different devices, and there are other aspects of working with I-mode that do require I-mode specific knowledge and practice - which adds to development effort. We're building out first UK I-mode service right now, so I've got good visibility of the process here. To reiterate: I like the processes and architecture of I-mode, but the choice of CHTML doesn't make a significant practical difference. I'm sure that when it comes to talking to content providers though, being able to talk about "a subset of HTML" is a much better story.
And a lot of what makes WAP different is designed around increasing the efficiency of transferring markup over low-bandwidth connections. Perversely, my theory now is that the overhead of WAP proxies slows things down more than the WAP layer actually gains you in performance: I-mode does indeed seem speedier to me.
4. *Very* interesting to hear that DoCoMo get a commission of content revenues in some cirumstances. So I-mode effectively becomes a means of their make a profit out of customers in territories where DoCoMo have no presence - beautiful.
5. O2 active vs I-mode: Jag, interesting that you say "If, however, we find that people are making their purchasing decision on the content as the main criteria": isn't this exactly the basis of the new I-mode advertising? Focus on the benefits not of I-mode itself, but of the content you can access through it.
Thanks very much!
Posted by: Tom Hume | November 26, 2005 at 04:38 PM
Tom, Jag, thanks this is great feedback and insights.
Tom, if you could be so kind to edit my posts and move the square brackets from touching the URLs this will fix up the "bad links", thanks.
The issue of making subscribers aware of their subscriptions and giving them easy ways to unsubscribe has certainly been taken seriously by many i-mode alliance partners including Telstra. In terms of integrity, goood governance and in the face of the Crazy Frog episode, it's clear that everyone now wants best practice, although I'd say that Japan in fact lags behind in this regard.
Terrific clarity about the O2 Active portal and also even more insightful about the choice of *device* still being the key factor in the purchasing/upgrade decision. That makes sense when I step down from the dream of where we would like to be and look at where we are today. This is exactly what I witnessed in going with a friend on this weekend who wanted specifically wanted to buy a phone with an i-mode service. The store sales staff were totally confused by this request and kept asking two questions only: (1) how much do you want to spend, (2) what handset do you like? To them i-mode was not only an unnecessary complication but it threw their sales pattern into disarray. I was astounded actually, but it brings home the reality of Jag's comments.
In menu structures, Jag's comments all spot on and I also strongly agree that O2 has a lot to offer the i-mode world in their experience in this area. I've actually written about this on my blog and people like Changingworlds http://www.imodestrategy.com/2004/12/changingworlds_.html and also in Australia the Bullant solution on SingTel/Optus, see http://www.goobile.com/2005/11/bullant_powers_.html The What's New has become very important, as Jag says, and this process should be studied and understood by forward-thinking content providers in order to maximise their potential visibility.
Re the ease of cHTML/i-mode development versus WAP - while I have no experience with WAP I can say that every seminar that I have been to where this topic has been raised the content providers, including Citibank in Australia, make the point that total time-to-market is of the order of 40 to 60% less with i-mode in their experience. In Australia this is most interesting because while Telstra is up there encouraging CPs to say these things to encourage other CPs onto i-mode they also run in parallel their own WAP Active Portal, so I assume that they are unbiased. Now, here's a point - many of these people are talking about free official sites, but not all by any means. So is the backend billing process for i-mode something that takes time and care, yes of course. Would there be a greater comparative saving for non-billing free sites, yes. But also I can say that many of the CPs who have spoken are indeed providing full subscription sites.
Tom, I am not sure why you are worried about the "points for subscription" model as to me this is a creative add-on to the subscription model. Content of different value can be accessed under the one subscription and people can be rewarded in credits for refering friends etc etc etc. It all seems to me to be a great idea. What's at the core of your concern about it being a weakness?
Cheers, Walter
Posted by: Walter Adamson | November 27, 2005 at 04:22 AM
Thanks for the feedback Tom/Walter. I agree, I think this discusssion is quite insightful!
Tom: your question re charging for traffic in Japan and whether this will be replicated in UK. The competitive environment in UK (and Western Europe generally) is to charge for things like voice and data at a certain standard rate - and then offer cheaper effective rates by prepaying (or bulk buying) bundles. In the O2 world - these are called "inclusive allowances" - which can be topped up using "bolt-ons" - i.e. buy more bundles of voice or data or messages etc. So - the £3 per megabyte charge that we have in the UK (for both O2 Active and i-mode) is the undiscounted (out of bundle) rate. The i-mode bolt-ons are £3 for 2Mbytes (50% discount) and £5 for 4Mbytes (60% discount). It might seem hard to believe - but even at the undiscounted rate of £3 per megabyte - it's one of the cheapest undiscounted rates for GPRS/3G data in the world - so in the £5 bolt-on - the rate goes down to £1.25 per megabyte. (By the way - the other thing about the O2 pricing is that it applies to both O2 Active and i-mode, prepay and postpay, and off-portal and on-portal - meaning that this pricing is the same for everything data, no matter what service you using, and what type of paying arrangement you have; - clear, simple, transparent pricing.) Contrast this with the other operators - let's do a quick check on the other UK operator websites:
As of today:
Vodaphone: If prepay then if browsing VLive! then price is approx £1 per megabyte else if browsing off-portal the price is approx £7.30 per megabyte
If postpay then if browsing Vlive! then price is free else if using data for other stuff then price is £2.35 per megabyte
Orange: if browsing Orange World then if postpay starts at around £1 per megabyte - and gets cheaper the bigger bundle you buy - the rates getting cheaper quicker on postpay than on prepay. Cheapest price is 13p per megabyte (400megabyte(!) bundle at £53 per month) - or "all you can eat" at £88 per month - where all you can eat is capped at 1000Mbytes per month. Else if prepay - then price starts at £3 per megabyte - and the cheapest you can get to is £1 per megabyte after consuming 15Megabytes.
T-Mobile: seems to be approx £7.50 per megabyte for access to WAP/T-Zones - can't see whether this applies to both postpay and prepay - but it seems that it does. Web'n'Walk has different pricing structure altogether - is based on 40Mbyte bundles and voice minute bundles for a fixed monthly rent between £30 and £55 per month. If you were only using the PDA for data - then the price per Mbyte starts at 75p per Mb - and becomes £1 per Mbyte after you've eaten through 40Mbytes. If you buy the contracts online and self-serve - then the data pricing is half that - but the out of bundle prices remain the same.
Three: Mobile Web Sites service is only available on postpay and has a fee of £2.50 per month - for which you get 5Mbytes of data. That works out to be 50p a megabyte. If you exceed that - then it becomes £2 per megabyte. You are restricted to sites that 3 deem to work OK (typically existing WAP sites) and it will vary by handset model.
So - as you can see, there's a variety of data prices and variations depending on 1) what you are accessing (on-portal or off-portal) 2) what type of contract you are on (prepay or postpay) 3) and in the case of three and T-Mobile - what type of handset you are using. The price spread seems to be in a range £7.50 per megabyte at the top end - to 13p a megabyte at the bottom end (not counting the all-you-can-eat) - where the monthly spends increase to over 50 pounds per month to get the cheapest rates.
Now - consider the following factor when judging the pricing propositions: just how much data can one realistically consume on a mobile device? The answer depends on several things: what network you are using, the type of device you are using also, and the context within which you are doing it. Let's try to build a very simple model for typical mobile data usage - just to get us started on this thinking. Assume that the consumption of mobile Internet happens whilst a) on the move between home and work or travelling somewhere and b) whilst at home or at work when you are not typically near a PC to get on the Internet. Let's assume furthermore that there is approximately half an hour of such time per day that you are using your phone for browsing. Assume that you are doing this an average of say 20 days per month: browsing on your phone. (Personally I think this is very high in mass market terms - but let's go with it for now.) If you are using a PDA type of device - and you are using it to browse full web pages (as in Web'n'Walk) then assume that the average Internet page size s 100k. If you are using a "phone" style of device then assume you are browsing branded portals or i-mode or WAP sites. Asusme that the average page size for these is 3kbytes. Now consider the amount of data you could possibly consume in a half-hour given that it takes time to download pages (downoload time) - and it take time to absorb the content on the pages (linger time). On a phone - the download time for pages will be pretty much the same whether on 2G or 3G for 3k size pages - let's say less than 5 seconds if you include the click and navigate times. The linger times will vary greatly - so let's just assume you spend an average of 20 seconds lingering on pages. (That will take into account pages that you just speed through - and pages that you spend a lot of time reading - e.g. news articles etc.). For PDA-style devices - let's assume (for GPRS) that the download time per page is around 60 seconds (bigger pages at around 1.4Kbytes per second) and that for 3G - the download time is 15 seconds (6.4Kbytes per sec) - and that the linger time is 3 x that of the phone (richer pages with more detail - and possibility of more time spent navigating a page using scroll bars etc.)
So - for browsing on phones - this very simple/crude model says that you will on average get through 2.4 pages per minute - consuming about 7.2kbytes along the way. Half an hour a day means around 70 pages per day - which means around 1400 pages per month in our model - which is about 4.2 megabytes of data.
For PDAs - our model says that, for GPRS, the page rate per half-hour is about 15 - which means 225 pages per month - which is 22.5 megabytes. On 3G - the page rate goes up to 24 per half hour - which is 480 pages per month - which is 48 megabytes.
Now - it had better be fairly fresh and compelling content in order for anybody to be committed to browse 70 pages per day on a phone 20 days per month and be paying for it. My belief is that this becomes more likely the more content there is available. The i-mode content model is all about having as much content as possible - that is of a quality, choice and freshness that the customer will pay for. More likely than within say closed, walled gardens. (I'm not actually sure that there is enough content in walled gardens to exploit the page rates that can be consumed on mobile phones!) Our model shows that if you had access to great content you would pay for - you would on average consume about 4 megabytes of data per month - which if you go back to the UK price plans for mobile data outside walled gardens - places you firmly at the top end of the scale for pricing with the some operators: Vodafone: £7.30 or £2.35 per meg for prepay and postpay respectively. T-Mob: £7.50 per meg (I think). - but at the much better end of the scale for others - e.g. Orange: £3 or £1 per meg for prepay and postpay respectively. O2: £1.25 per meg (if you buy the 4Mbyte i-mode bolt-on) regardless of prepay or postpay.
So - if you look at it in the way we construct the model - the O2 i-mode pricing is very competitive - and offers a better content line-up than simply going off-portal with the other operator WAP services.
Then you look at the PDA version of the model - and this is one for the pundits who claim that the over-the-air access to the regular Internet pages will kill WAP and i-mode: Today's pricing for data means that despite the fact that you get through pages much slower this way - you end up consuming between 22 and 48 megabytes of data per month - for which you would have to be paying at least £40 per month (Orange) or £30 per month (T-Mob Web'n'Walk after half-price promotional period ends.) Interestingly - the O2 "Mobile Web" 36 Megabyte data bundle is £14.50 per month - and £18.50 will give you 75 Megabytes (with rollovers) - so the O2 pricing is better than the rest if you already have your own PDA. Anyway - the point that I'm really trying to make here is that the PDA-style of browsing the full Internet is still far from being mass-market - not only from a pricing perspective - but also from a user experience perspective. Things like Web'n'Walk might appeal to the likes of me and you - who are comfortable carrying multiple devices - and are comfortable using a stylus and mobile keyboards on devices - that run Windows etc - whilst sitting on the bus or the train - but the majority of people are not there yet. The majority of people are still carrying "normal" handsets - and are just about used to navigating menus on small screens - and are used to typing things on numeric keypads reading and writing messages on small screens - and some maybe are downloading games and ring tones from WAP push type messages. i-mode (or i-mode-like) is a more natural next step than full-blown Internet on a PDA.
And finally - regarding your point 5 Tom: yes the advertising is all about the new things that you can do on your phone - like find a job, book a holiday etc. and this is what we are trying to get mobile phone users to think about. But we won't change people's buying decisions overnight - it will take time. And as we increase the penetration of i-mode devices in our handset portfolio over time - then there is increasing likelihood for things like these new propositions to become greater factors in the buying/upgrading process.
Posted by: Jag | November 27, 2005 at 01:24 PM
A very useful post and comments!
UK customer view - not enough differentiation between standard "Internet on mobile" and "i-mode" leads to confusion of both customers and retailers.
Example: Carphone warehouse shop
me: "I've seen you've got i-mode offers from O2, can you explain me its benefits?"
Sales: "Well, you get to access web sites"
me: "I can do it already on my handset, how is it better?"
Sales: "Well, the ones you get now are not _real_ site, with i-mode its better sites"
me: "Look, I can already access my email and many content providers via my current data connection, why should I buy in to i-mode?"
Sales: "Let me check with someone..."
...after talking to someone he brought up on their sales PC a page provided by O2 to explain i-mode to retailers and we read it together - to be honest from what I read I did not see how the sales assistant could have sold it to any customer...
Example 2: O2 web site, still trying to find the answer to why should anyone bother with i-mode.
There is a flash demo that does anything but demonstrates the technology and its benefits. The only half-real example is Lastminute.com interactive demo that looks exactly as their standard WAP-based site.
Example 3: O2 shop (real world ;)
They had some brochures and "i-mode newspaper". Beyond this also not much information on why should I buy in to this.
Maybe I'm a bit harsh on the current state, but without a single simple selling message I can not see the technology succeeding which would be a real shame... To say that it's about "the new things that you can do on your phone - like find a job, book a holiday etc." is in my opinion simply misleading and weak - most customers have heard that they already can do such things on their phones and some of them are alredy even doing them ;)
Finally, what about developers? OK, i-mode is now here, how can I open my shop in i-mode space? If we follow the strategy of creating attractive content to get customers O2 has to be providing support for merchants/developers. So far I cannot find a single link on O2's web site that will provide a good starting point (can anyone one point me in the right direction?).
Anyway, it's just my silghtly disappointing view on the very low key (not even an article on BBC News) and messy (no clear benefits) launch of the great technology.
Posted by: Igor | November 28, 2005 at 01:37 AM
Igor: good feedback!
You said:
"To say that it's about "the new things that you can do on your phone - like find a job, book a holiday etc." is in my opinion simply misleading and weak - most customers have heard that they already can do such things on their phones and some of them are alredy even doing them ;)"
I'm not so sure that "most" customers would know how to. For such activities to become commonplace - it needs to be made easier to do so. I think the i-mode approach makes it easier. For example - I booked a restuarant using my mobile phone for the first time a few weeks back whilst sitting on the bus on the way home from work. Why? Because it was easy to do so. Even though I could have done this a couple of years ago (and have spent thousands of pounds in Internet shops over the last 7 years) - a combination of things have just never been lined up at just the right moment for me to have done so on my phone; e.g. a few minutes of dead time, one click to get online, fast menu navigation, toptable site already there, well designed pages, easy search and got instant email notification of booking. The opportunity to do this has just never presented itself in the past - not without some effort anyway - and most people treat their mobile phones as an "effortless" device.
Also - expecting this to be seen as some sort of new *technology* is, IMHO, wrong. That's not what it is - and we wouldn't want anyone (including the BBC) to see it that way. Sure, there's a technology element to it - but it's not rocket-science - in fact it's pretty much the same as non-i-mode - but what's different is that an attempt is made to create the right conditions for data services to be consumed end-to-end: from handset, platform through to Content Provider.
Regarding addressing the needs of developers and unnofficial CP community - this is being planned for the new year. OK - if you leave your email address in the comments section of the i-mode article in my personal blog at http://route79.org/journal/?p=57 then I will give it to the right folks and make sure that you are notified when stuff happens.
Sorry to hear about your disappointment Igor: but if you can say to a store assistant "Look, I can already access my email and many content providers via my current data connection, why should I buy in to i-mode?" - then you're probably not going to see why i-mode is any different. But I don't think that most people are like you! ;-)
Posted by: Jag | November 28, 2005 at 12:01 PM