It would be understatement to say that there is a strong sense of expectation surrounding multimedia messaging (MMS). Companies from all parts of the wireless value chain are looking to MMS as the enabler for a new period of growth in the industry. Network operators, in particular, are hoping that the investment they have made in packet-based infrastructure and the expansion of bandwidth will be justified by the additional revenues from multimedia messaging.

This level of expectation contrasts starkly with the introduction of text messaging (SMS) in the early 1990s. SMS established itself as a vital communications tool against all the odds, marred by a terrible user interface, high usage costs and an almost total lack of promotion.

The SMS channel was envisaged as a means for the operator to send brief service messages to subscribers and few carriers attempted to market it as anything else. One of the most popular initial uses was signalling when the user had new voicemail waiting. The peer to peer texting phenomenon was driven by network effects more than the marketing efforts of operators, although smart moves were made at an industry level to ensure seamless compatibility between networks. Today it is as easy to send a text to a colleague in the next room as it is to a friend on the other side of the world.

To give credit where its due, most network operators moved swiftly to build on the success of SMS. New tariffs were introduced to encourage usage, pre-paid packages brought texting within reach of the youth market and content providers were enlisted to drive demand for premium information alerts. Where first generation WAP services were difficult to access, slow and almost impossible to monetise, SMS alerts promised a usage model and pricing subscribers already understood.

The success has been astonishing: SMS now accounts for around 10 – 15% of the average GSM operator’s revenue. In some regions, especially Asia Pacific, SMS represents over 20% of revenues – at Global Telecom in the Philippines, data (primarily SMS) was 27% of 2001 revenues. The GSM Association’s figures for May 2002 show 24 billion messages sent monthly and predict a total of 360 billion in 2002 or nearly 1 billion per day. A quick back-of-the-envelope calculation produces some impressive figures: 1 billion messages, charged at an average of $0.10 each, equates to $100m in SMS revenue every day.

So how will the introduction of multimedia capabilities impact the messaging model? One fundamental unlikely to change is the attraction of messaging over browsing in the mobile environment. The asynchronous, alert-driven nature of messaging is perfectly suited to mobility. In his book, Mobile Commerce (Cambridge Press, 2001), Paul May talks about the importance of time, location and mission to mobile users. Plotted on three dimensional axis, the value of a service increases according to its relevance to the user at a given point in time, dependent on physical location and current mission (i.e. at work, travelling, relaxing at home etc…). Messaging scores much more highly as a delivery mechanism than browsing on all three counts.

The peer-to-peer exchange of short texts is also unlikely to decline. It is now established as a staple of interaction in many countries, supported by stable pricing and a massive installed base of compatible handsets. Indeed, it is tempting to believe that MMS will do little to alter the messaging model. After all, it is already possible to send pictures and ringtones using SMS, so what more could MMS add?

By packaging together and expanding a range of existing capabilities, the multimedia messaging concept represents something far greater than the sum of its parts. A digital camera on its own has a certain value, but integrated into a handset and capable of instantly sending images to any e-mail address, it suddenly becomes much more compelling. The ability to download an MP3 to your phone is exciting, but a service which automatically delivers the latest chart hits every week is something you might well be willing to pay a premium for.

Picture messaging is expected to be one of the most attractive uses of multimedia messaging and the first to be implemented by many operators. The success of proprietary picture messaging services in Japan, most notably J-Phone’s Sha-mail, has prompted operators around the world to accelerate the rollout of MMS and camera handsets in their own markets. User adoption of Sha-mail at J-Phone has been astonishingly swift, rising from under 1m users at April 2001 to over 5m (35% of the entire subscriber base) by March 2002. In the same period, data almost doubled as a percentage of average revenue per user (ARPU), rising from 8.3% to 15.1%.

So successful was Sha-mail that analysts pointed to it as the single most important factor in J-Phone surpassing KDDI as Japan’s second largest operator earlier this year. NTT DoCoMo has been forced to respond by introducing its own picture messaging service, i-Shot, on 1st June 2002. According to figures released on 22nd July, DoCoMo sold 600,000 i-Shot handsets in the six weeks since launch.

Evidence from the Japanese market suggests picture messaging can be a disruptive technology, enabling quick-thinking operators to rapidly gain market share from competitors who act more slowly. It is also worth noting other characteristics of Japanese picture messaging services: they are based on proprietary technology and until recently, images could not be exchanged between networks. There is also a relatively low penetration of desktop internet users, a group which will play an important role in western markets as recipients of ‘my first wireless photo’; almost all of the traffic generated by Japan’s picture messaging services has been handset to handset.

If picture messaging is introduced in Western Europe based on multimedia messaging standards and SMS-style inter-networking agreements, the potential is enormous. However, adoption is likely to follow a different pattern than in Japan. Fewer Western European users have handsets with colour screens and the packet data connectivity required to make receiving a picture message worthwhile. The most likely recipients of a picture message will be the massive installed base of desktop e-mail users.

Developers of MMS infrastructure are addressing this in two ways: firstly by building internet e-mail gateways into their products, allowing pictures to be sent direct to any e-mail address and secondly by enabling users to receive a text alert on their handsets with details of how to retrieve a picture via the web. Neither of these implementations scores highly on the ‘time, location, mission’ graph I mentioned earlier, but both will be important in convincing early adopters to buy expensive new camera phones. To coin an overused phrase, it is a classic ‘chicken and egg’ scenario: why buy a camera phone when there is no one to receive your pictures and why buy an MMS handset when there is no one to send you images?

There is a role for the operators here too. The cost of manufacturing camera phones is likely to remain relatively high for some time and large operator subsidies are a thing of the past. It is likely, therefore, that lower cost handsets capable of sending and receiving multimedia messages but lacking an integrated camera will outnumber camera phones for some time. The Nokia 7210 and 6610 are examples of this category. Users of these devices will want compelling content from day one and network operators are well positioned to facilitate the supply of rich information services, just as they did with SMS.

It will be a basic requirement for operators to offer the standard range of weather, news and sports alerts. Indeed, it would be advisable for such services to be offered at the point of sale, for instance enabling the user to subscribe for daily weather reports when they purchase the handset. The pace of market adoption could also be increased by providing a free trial of such alerts, allowing the user to see the additional value of multimedia messaging first hand. In addition, operators and handset manufacturers must provide a wide selection of pre-formatted content for those handsets without integrated cameras – greeting card templates, amusing photographs, cartoon characters and so on.

The first edition of the MMS specification supports audio, image and text content, packaged together in a miniature slideshow presentation of no more than 50 Kb. This vastly expands the possibilities for developers currently constrained by the 160 character limit of SMS. Handset manufacturers and industry bodies such as the 3GPP and WAP Forum are playing key roles in disseminating development information and tools to prospective MMS innovators, conscious that the availability of exciting services will drive device and infrastructure sales.

Nokia has been particularly active in this area through its Forum Nokia initiative. This provides both technical resources and marketing support to help MMS developers establish commercial relationships with operators and service providers. The most innovative services often come from smaller companies with little experience of the Byzantine operator structures and Nokia, through its developer’s Tradepoint, will shortcut the deal making process. Developers of messaging platforms often have their own programmes, broadening the appeal of their software products by providing a ready-made package of third party applications already optimised for deployment.

Most operators have also established their own programmes, but the extent to which they embrace the developer community varies considerably. Some prefer to rely on intermediaries such as equipment manufacturers or platform developers for introductions to third party developers and use their internal teams solely to manage the commercial relationship and technical implementation.

Others are more pro-active, hosting developer events, enlisting developers to create specific applications and even offering venture financing for interesting projects. In the UK there is the Orange Developer Forum, Vodafone Via, Source O2 and Hutchison 3G actively encourages developer submissions through its web-site. The 3GSM World Congress, the mobile operators’ key annual event, will have a dedicated focus on content providers for the first time at the next meeting in February 2003.

The learning curve for developing MMS services is relatively short. MMS content is specified using a tag structure similar to HTML, immediately opening development opportunities to a huge pool of existing web developers. This is a field in which graphic designers will play alongside telecoms engineers. As the technology matures expect to see simple content creation tools which enable talented content developers to rapidly create multimedia services without in-depth technical knowledge. Indeed, network operators eager to build their MMS traffic would do well to encourage the free distribution of desktop content creation tools ahead of the widespread availability of MMS handsets.

The one key consideration foremost in the minds of all those pinning their hopes on MMS is pricing. No amount of handset innovation, third party applications and operator marketing will establish MMS if a viable price structure is not established. There are several factors which will govern the pricing model – the actual cost to operators of transmitting the data, the level of subsidy used to get MMS handsets into the market, the fees paid to third party content developers and, of course, the consumer’s value perception.

Network operators are hesitant to talk about what it costs them to transmit data across their networks and it varies according to the technology used. Talk to CDMA vendor Qualcomm and they will tell you that data transmission across a CDMA2000 1x network is considerably less expensive than sending the same data over GPRS. Qualcomm’s March 2002 white paper – ‘The Economics of Wireless Data’ – claims 1 Mb over CDMA2000 1x will cost the operator around $0.07, while 1 Mb over GPRS will be approximately $0.42.

Using our back-of-the-envelope calculations again, we can estimate that a 50 Kb message over a GPRS network will cost the operator $0.02. Let’s say the operator has subsidised the handset to the tune of $150 and hopes to recover 50% of that from MMS services over a standard 12 month contract. If a user was to send 20 peer-to-peer messages (i.e. where royalties to third party content providers were not applicable) per month, the operator would have to charge about $0.33 per message just to cover the cost of subsidy and data transmission. Add in a 20% margin and you arrive at a price of $0.40 per message.

This contrasts with an average of $0.10 for text messages. Will the consumer be happy to pay this much to send a photo and audio clip to a friend? I suspect the answer will be ‘yes’, but much will depend on how the pricing is presented to the user. $0.40 is, after all, a relatively small amount to pay for an impressive service – the ability to capture a unique moment and share it with a friend – something which would rank highly on our ‘time, location, mission’ graph. If a set pricing level can be established, just as it has been for text messaging, the individual cost of each message because a non-issue in the mind of the user. The danger lies in uncertainty.

Some operators are talking about billing for MMS just as they have done for early GPRS usage – per kilobyte or as part of a set usage plan (i.e. $10 for 5 Mb per month, $20 for 15 Mb per month). I believe this would be a huge error. Not only will the average user have difficulty understanding the concept of data size, but it will establish a culture of restraint and uncertainty totally at odds with how MMS should integrate with the user’s lifestyle. MMS, especially taking photos and sending them to friends, should be a spontaneous, creative activity. It would cripple MMS usage if people are cautious about how a drunken night of picture messaging at a bar might impact their monthly bill or how much extra it will cost them in data transfer charges to add an audio clip.

In their Mobile Image (IT Press, 2002) study of the social impact of mobile imaging Finnish researchers Ilpo Koskinen, Esko Kurvinen and Turo-Kimmo Lehtonen observe: “The impetus to establish contact with others is often quite simply that one has a bit of time and is able to free oneself from the flow of busy routines. In this sense, sending mobile visual messages is similar to recreational use of a mobile phone while, for instance, waiting for a bus or a train or being on the way and nothing is happening.” If we accept this then peer-to-peer messaging will be extremely sensitive to pricing uncertainty. As an indulgent or ‘luxury’ activity the pricing structure must be such that it encourages spontaneity.

Four pricing models typify operators current thinking: T-Mobile UK has said that it will charge £20 ($31.20) per month and allow users to send 10 Mb worth of MMS. Westel of Hungary splits messages into two categories of small (sub 10 Kb) and medium (10 Kb – 30 Kb) and charges HUF 76 ($0.30) and HUF 160 ($0.64) respectively. Telenor of Norway charges a flat fee of NOK 10 ($1.30) per message. In Hong Kong, CSL charges HKD 0.50 ($0.07) per Kb of usage up to a maximum of 30 Kb. In summary, we have an upfront monthly fee, individual pricing, flat pricing and metered date usage.

Although Telenor’s flat pricing appears most expensive, it is this model I would recommend ahead of the others. Even if operators need to price the flat fee significantly higher than they would if they adopted Westel’s individual pricing, I would argue that an easily understandable, universally accepted flat fee will lead to greater overall usage.

Many operators have also offered free trials of MMS in the first few months after launching. This is sensible, market building activity. The danger lies in the use of special introductory rates or lower pricing to encourage initial usage. Once a user has accepted a particular pricepoint, it will be nigh on impossible to increase that price without damaging the customer relationship. Much better to offer a free trial or free information services, as outlined above, than to attempt an aggressive initial pricing strategy.

Users will pay for capabilities and additional features, not bytes transferred or message size. When the networks become fast enough to support video transmission, it might be possible to charge a premium for video over static content, but again this should be on a flat basis – say $0.40 for a picture message and $1.00 for a video clip.

Similar principles can be applied to the pricing structure for information services, although this is a more complex area. New factors need to be considered, such as revenue sharing agreements with third party content providers, support through sponsorship and subsidy of responses. Uniquely, MMS supports the use of free or cut-price replies. For instance, an information alert about your favourite tennis player might include an advertisement from the player’s sponsor, inviting users to reply and receive details of where to buy the kit the player is using that season. To encourage a response, the sponsor could make it clear that the reply will be free of charge and foot the operator bill itself (albeit at a bulk discount).

Operators and content providers should price information services on the basis of $0.50 for a news alert, $1.00 for a stock price notification, $1.50 for a horoscope etc… The user cannot be expected to sign-up for a service where they will be billed an unspecified amount based on the size of messages over which they have no control. People subscribing to any service – whether traditional or new media – do so on the basis of a simple, subconscious calculation: can I afford to pay X amount to receive Y capability. Pricing structures must work with this thought process rather than trying to alter our instinctive behaviour.

However, further complexity is added into the mix when you consider the financial value to operators of third party content and how it impacts revenue sharing agreements. The fundamental question for operators is: can I afford to pay the content provider a higher percentage of the revenue because their service will attract more users to my network? Few operators and content providers are willing to talk publicly about their revenue share arrangements, but the usual split of the premium is about 15% for the operator and 85% for the content provider.

What if, for instance, the content provider is MTV or Disney? An effective argument could be made for offering these content providers, say, 95% revenue share in return for a period of exclusivity. The hope would be that the lure of a certain type of content would generate sufficient usage to justify the favourable terms of the revenue sharing agreement. Unlike consumer pricing, where flat fees should dominate, I suspect that commercial agreements with content providers will be as flexible as they have to be to attract the big brands.

Handsets too will have a massive impact on the development of MMS. Today we have the Nokia 7650 and SonyEricsson T68i in Western Europe. Further models are planned for release in Q3, including Nokia’s 6610 and 7210, SonyEricsson’s P800 and Sharp’s GX1. The first devices will be expensive, those with cameras will be larger than the handsets we are used to and the colour screens will limit battery life. It is essential, therefore, that they are exceptionally user friendly and so beautifully designed as to overcome our concerns over price and performance.

Manufacturers would be foolish if they pointed to the experience of SMS, which was adopted despite its poor user interface, as an excuse for the sloppy addition of MMS capabilities to existing handset designs. True enough, SMS established itself despite nine key input systems and being hidden away in complex menu systems, but it only really began to explode when predictive text input, pre-formatted messages and decent inbox systems were introduced.

Composing a full multimedia message is actually quite a complex operation. It requires the user to capture a photograph, add an audio clip and insert some text. Imagine trying to do that on your existing handset, with its greyscale screen and fiddly buttons. The user experience can be vastly enhanced with the addition of large colour screens, joy-stick navigation and short-cut buttons (i.e. a hardware button for instantly activating the camera or recording a voice clip).

Here again there are lessons to be learnt from Asia, where proprietary picture messaging services have been operational for some time. Clamshell handsets have swivelling cameras embedded in their hinges and others have a separate colour screen for use as a viewfinder. The easier we make it for the user to compose exciting multimedia messages, the more will be sent. Expect to see basic photo-editing capabilities included with most high-end handsets, allowing you to change the colour of a picture or morph a friend’s face into an amusing shape. Give people the correct tools and their creativity will amaze.

Some operators, such as O2 in the UK, are already trying to build a competitive advantage in MMS by securing exclusive agreements with manufacturers. Sharp’s GX1 camera phone will only be available to O2 subscribers. The Hewlett Packard 928, a PDA with integrated GPRS and the ability to add a digital camera, is available in the UK exclusively on the Vodafone network. Once some of the smaller manufacturers catch up with SonyEricsson, Nokia and Motorola, expect to see handsets being customised to operator’s individual specifications.

MMS is already with us. I could walk to my local mobile phone shop and purchase a T68i with its CommuniCam accessory and start sending picture messages over a GPRS network straight away. If you travel between Finland and Hong Kong, you should even be able to roam with your MMS services on the Sonera and CSL networks. However, it’s early days. The information services are not in place yet, handset prices remain high and operator tariffs are experimental.

Nokia’s 7650, arguably the most anticipated MMS handset, is starting to arrive in commercial volumes and when Europe returns from its August holiday’s most shops will stock at least two different MMS devices. Operators are planning to start heavy MMS marketing campaigns in September, preparing for the run-up to the holiday buying season.

I can’t help but feel optimistic about the prospects for this exciting new service. The reasons for my bullish stance are threefold: messaging is a metaphor people accept and understand, the industry has recognised the importance of MMS and invested huge resources in ensuring successful implementation and the desire for shared visual experiences is fundamental to human nature.