Why the two sided business model does not apply to telecom operators

I have been thinking of this since I attended the LTE world summit in Amsterdam.

The two sided business model has been used by many in the industry, especially @Telco2, to point to a potential gold mine for Operators by comparing to the Web.

While the concept has been popularised recently by Telco2, the two sided business model is a generic idea in macro-economics and in this post, I am going to consider the two sided business model from first principles.

I believe that it is wishful thinking to consider that the two sided business model applies to Telecom Operators.

Two-sided markets , also called two-sided networks, are economic platforms having two distinct user groups that provide each other with network benefits. Credit cards are one such example: Consumers prefer credit cards honoured by more merchants and merchants prefer cards carried by more consumers.

When applied to Telecom Operators, the two sided business model is perceived to have two sets of customers: The downstream customers (millions of consumers) and the smaller set of ‘upstream’ customers – specifically verticals such as healthcare, retail, media etc

However, when we look back at the definition of the Two sided business model, the Operative word is to ‘gain network effects’ i.e. the platform, in this case, the Operator, is supposed to facilitate network effects on both sides

The reason I question the feasibility of the two sided to Telecom Operators is: I do not see how the Operator provides network effects on either side.

A classic example of network effects is the fax machine or the telephone. The more people own a fax machine or a telephone, the more valuable it is to each owner. In other words, every subsequent user adds value to the whole system.

Do we really believe that subsequent users of the Telco platform add value to all other users of the platform?

Now, consider the ‘other’ side.

When considering network effects(fax machine concepts) to two sided business models, the members of each group exhibit a preference regarding the number of users in the other group (cross-side network effects). If firms account for the fact that adoption on one side of the network drives adoption on the other side, they can do better.

We are not seeing any examples of such behaviour (cross side network effects) when the Telecom Operator is a platform

So, not only is the Telecom Operator platform not demonstrating scale (on it’s own side of the platform) but nor is it demonstrating cross side network effects on the ‘other side’.

Let’s consider a concrete example: Network APIs (when applied to Operators)

Yes, network APIs have value to the customer but they do not provide network effects.

Contrast this to Apple and it’s developer ecosystem

Here, we do see a two sided business model ..

By many standards, that ecosystem is also ‘closed’ (ex approval of apps) but, critically, it DOES achieve scale on both sides of the platform (In this case, developers and consumers).

In contrast, the Operator as a platform is (relatively) closed, fragmented and not globally interconnected. The Telco platform is confined in most parts to one country or a group of countries where the customers are not necessarily constrained to using the Telco app (ex they could get the same service from an app from the device maker or from downloading the app from the web on to their phone). The same applies to all such ‘upstream’ customers – ex Retail, health care etc etc in relation to the Operator.

The fallacy lies in applying the ideas of the two sided business model to the Network Operator just because ‘Google / Facebook etc are doing it’. .

Ironically, the Apple developer ecosystem succeeds even when it is closed precisely because it manages to use the principles of the two sided business model to create network effects

Taking a more formal, mathematical approach in the paper Two-Sided Markets: A Progress Report – pdf Jean-Charles Rochet Jean Tirole conclude that

Because all markets involve transactions between two (or more) parties and therefore are potential two-sided markets, it is useful to circumscribe the scope of two-sided-markets theory.
The first objective of the paper has been to propose such a definition: A market is two-sided if the platform can affect the volume of transactions by charging more to one side of the market and reducing the price paid by the other side by an equal amount; in other words, the price structure matters, and platforms must design it so as to bring both sides on board.

Jean-Charles Rochet and Jean Tirole ‘s formal mathematical analysis makes a lot of sense to me because it takes network effects into account and in that context, I cannot see how these ideas apply to Telecom Operators.

To conclude:
The idea of a two sided business model is intertwined with the idea of creating network effects.

The two concepts (two-sided business models and network-effects) cannot be separated.

Hence, the two sided business model does not apply to telecoms because there is no evidence that the Operator platform can nurture scale on both sides of the platform (cross-side network effects)

Image: the roman god Janus – who can see on both sides – but is mythical :)
Image source: flickr


  1. Alex Visser says:

    Hi Ajit,

    I think it is important to make the distinction here between direct and indirect network effects:
    Direct network effects are based on horizontal compatibility between users. A user’s link to a network has no value by itself, except to facilitate interaction with other users.
    Indirect network effects are based on vertical compatibility. Users benefit indirectly from the adoption of that service (e.g. Master Card) by other users because this will eventually allow them to use a wider variety of complementary services (e.g. shops that offer Master Card as a payment method).

    It are these indirect network effects that support the functioning of service platforms and multi-sided markets

    As you stated, there indeed are no direct network effects associated with a telco platform
    However, in my oppinion the notion of indirect network does apply to an operator platform such as carrier billing. This service is more attractive to subscribers as more developers implement it in their services/apps and likewise more attractive to developers as more subscribers are usint it as apayment method. Actually very similar to the often cited credit card example…

    To challenge your final remark, it could indeed be true that there is no evidence yet of a successful implementation of such a service (and it could be very likely that operators will never succeed in implementing a successful and feasible service platform), but that doesn’t entail that the concept of two-sided business models does not apply to them at all

    what’s your take on this?

  2. ajit says:

    Thanks Alex
    Re — However, in my oppinion the notion of indirect network does apply to an operator platform such as carrier billing.

    On first impressions this sounds good – but when you explore more developers have many options and also the audience of a carrier is limited. so the network effects are not seen

    Re – the qs of if such a model could arise in future – I dont think so because I dont see any dimension where you could get scale – partly because of delay from operators but more because there are now many options – (contrast with apple ecosystem which is self contained and has benefits on both sides)
    kind rgds

  3. Telecoms is already full of examples of multi-sided markets (or degenerate forms of them): termination fees, freephone, yellow pages, premium rate services platforms, mobile payment platforms, mobile signature verification services.

    The test of 2-sidedness that matters is whether the platform takes on inventory risk; if it does then it is not 2-sided. The platform pricing (to generate volume) is the other discriminator, but is only a sufficent condition, not a necessary one.

    The opportunity for telcos may be as partners/suppliers into multi-sided markets. They don’t have to be the platform operator. Indeed, sometimes the platform owner is not the profit-taker (compare Visa network).

  4. ajit says:

    Hello Martin

    Thanks for your comments

    1) Re The platform pricing (to generate volume) is the other discriminator,
    Yes, you have captured the crux of my argument. I feel that this aspect is ignored in all the discussions and thats exactly what I want to point out

    I do not agree that it is not the ‘necessary criteria’. IMHO its absolutely critical because failing that the models are reduced to a series of niches – and thats hardly of interest to operators. This is partly why I went back to first principles(see the papers i linked from academia). Volume is ** critical ** and it seems to me that its like saying the ‘emperor has no clothes’ ie the volume is an obvious aspect but yet not talked of .. and all we see is a distant vision of a ‘promised land’ ignoring some very apparant pitfalls on the way!

    2) Re The opportunity for telcos may be as partners/suppliers into multi-sided markets
    Perhaps – but then so could be others ie the telco as a partner(with no scale) is in par with anyone else in the industry ,,.

    contrast with FB and others who have scale, are global and can provide revenue we see that the two sided model is still not applicable IMHO

    many thanks
    kind rgds

  5. There are already marketing in which telcos have huge volume (voice, messaging, and the thing we currently call “broadband” but ought to be “managed packet flows”).

    The opportunity is to retro-fit 2-sided market structures onto these. My upcoming talk at eComm will give some very practical examples of ‘how’ that are being missed. Some of the theory is presented in my “Connect, Interact, Transact” paper at http://www.martingeddes.com/papers.

  6. ajit says:

    thanks for the links Martin. kind rgds Ajit

  7. Tony Fish says:

    Ajit – great piece. You do however have to wonder why operators carry on looking at data and ignoring voice, which does have a natural network effect…..

  8. ajit says:

    thanks Tony. I believe that there is a role for voice but the challenge for operators is to make it non ‘session based’ ie to make it embeddable and usable by third parties. its one of the great areas which telcos have not innovated but could do so rgds Ajit

  9. Thought provoking Ajit. You argue that in a two sided market each side benefits when the numbers increase on the other side (you cited the credit card example: Consumers benefit with more merchants and vice-versa). Then you argue that since this is not the case in the Telco space [I think it is, but let's say it is not as you claimed for the sake of the argument], then the Telco network cannot be two sided.

    Based on this argument, Google is not a two sided platform either. In the Google case you have advertisers on one side and consumers on the other. Advertisers benefit with a larger base of consumers, but consumers don’t really benefit with a larger number of advertisers.

    Based on your argument, would you say that Google is not a 2 sided platform either?

  10. ajit says:

    Thanks Francisco. Quick comment since I am trying to catch a flight soon ..
    My initial impression is: the two sides for Google are people and content ie the search brings people and content and each benefit the other. On top there is a monetization model (through advertising) kind rgds Ajit
    PS – like your work! hopefully we meet one of these days

  11. Thanks Ajit. It would be good to meet. I’m in the Silicon Valley.

    I don’t think of “content” as being one of the two sides in a search platform. Content providers generally do not explicitly choose to join the search platform. Google simply finds their content and indexes it so it can attract consumers and advertisers to its platform. In my mind the two sides of a search platform are consumers and advertisers. The key point of this two sided platform is that advertisers subsidize the service for the consumers. This attracts more consumers, which in turn attracts more advertisers. It’s a little different than the credit card example you gave, but still a two sided platform in my mind.