[This is a repost from an article I published in LinkedIn in Feb. 2016]
Let’s start this post with some boring introduction that states the obvious.
“With the rapid expansion of data services and the explosion of traffic over radio networks, operators face challenges to get all those gigabytes through the airwaves while maintaining or improving the quality of service… blah-blah-blah” yeah, this is certainly something you have heard over and over already.
Squaring the circle of keeping profitable in managing data volumes explosion
Why so? Because MNOs have to solve an equation which is the “squaring the circle” problem of mobile networks, where variables are CAPEX, not exponential and with a tendency to stabilize or reduce, Data Volumes, growing exponentially, and the Wallet of Customer, which is constant at best. In this equation the evolution of revenue per GB will only go down (reverse exponentially) where cost per GB can be optimized a few folds but not at the same pace.
So operators having no sustainable long-term solution, they have to play with a few tricks or short-term moves:
- Implement features to improve radio network efficiency on data
- Use Wi-Fi hotspots to offload RAN on most congested areas
- Deploy 4G to support data volumes demanded by heavy users
- Acquire more spectrum or refarm their legacy voice spectrum for data
- Play with pricing, push usage outside peak hour to reduce the BH ratio
Those tricks can give extra 3x, 5x, 10x extra capacity but won’t create an exponential capacity trend. On the long term, the addition of a fixed network to a MNO will be the only way to deal with this exponential growth. With fiber you can provide this “unlimited capacity” at home, at school and at work where you spend most of your day, while your mobile network will maintain your connectivity on the move.
It would be very interesting to see here the data usage breakdown of an average subscriber depending of his location – home / office / school / on the move – whether it is on his mobile phone or on a computer, a TV, a tablet [Share in the comments below if you have insights on this !].
New models to manage the cost of delivery of data for operators
Recently, two operators announced they will start blocking ads:
- Digicel signed a deal with the Israeli company Shine to test ad-blocking at network level in its operation in Jamaica, before a deployment of the solution in its Caribbean and Latin American markets.
- Three announced the same deal to enable ad-blocking across its network. Initial deployment will be done on Three UK and Italian networks followed by its other markets in a second stage.
The press releases of those 2 operators describes these initiatives as a move to please subscribers and clean their webpages of ads and malwares, but criticisms suspect that those MNOs actually to plan to negotiate revenue shares with mobile advertising companies in the background.
Digicel claims that ads use up as much as 10% of a customer’s data plan allowance, thus implementing this features could be seen as a 10% price reduction of data services for subscribers in markets where data plans are more volume-based than service-based.
This phenomenon highlights a problem today: most data plans are volume-based while consumers are service-conscious: they don’t care how many GBs they need per month but they know how often they want to use Facebook / YouTube / browse some websites and so on.
The Signal to Noise ratio of Internet Applications
In this shift where everyone will have to chip in for mobile network capacity, providers will need to be more conscious of what I want to call their “Application Signal to Noise Ratio”.
Let’s take an example: in Senegal there is a very popular news website which most of people visit daily if not several times per day (I won’t give its name you will understand why). Unfortunately this website is not optimized at all and its home page size varies between 9MB to 25MB (yes, you just read twenty-five megabytes) just because the amount of files it requests – more than 400 scripts, styles, fonts, images – and the invasion of picture and video ads – in the background, on the sides, between articles – make the total page very, very fat.
Compare this website with another well-optimized news website (yahoo.fr) which is also popular in Senegal: for the same type of content, its home page will weigh just above 1 MB and need around 100 requests to load.
From the end-user perspective, reading news on the non-optimized website “costs” him 9 to 25 times more money than on the well-optimized website on a mobile network than on yahoo.fr, but he is probably not aware of this and might even blame his operator for stealing his money and being slow ;-).
While the useful content of both websites – what I call the “Signal” here – might be the same (headlines and summary of articles and illustration picture thumbnails) but the data volume of one is 10 times as big as the second one, probably because:
- Ads are too heavy and invasive
- Thumbnails are not compressed enough
- Page code is not optimized or does not use compression
In this situation: the customer does not know that they pay 10x more from their data plan for the same service, the website publisher does not care it could cost 10x more to their visitors and to the mobile carrier to deliver their content so all the burden lies on the shoulders of the operator.
The customer is only interested by the “Signal” part of the service he is accessing, text of articles and pictures in a resolution optimized to his screen and this is what he is ready to pay for. All the overhead just increases the cost of delivery and potentially the cost for the users who have volume-based plans.
Making content providers or mobile advertising companies pay for the data they send to the network would have a silver lining: improving the “Application Signal to Noise Ratio” and reduce all the junk we encounter on Internet.
That will be a necessary evil to move away from volume-based plans to service-based plans, as each service will have to optimize itself so that the operator can sell it at a time-based price that will look attractive to its customers.
Note: this article is born from the reading of the some articles including:
Note 2: source of the article illustration: http://marketingland.com/ios-ad-blockers-drop-144526