Mobile Video Pricing Challenges: Part II
Following up on part I of my post about mobile video pricing challenges, I asked many content owners what the typical prices are for ingesting big content for delivery to little screens, i.e., many formats in, one or two formats out in lesser quality form? How much are they paying and what percentage of the overall cost of mobile video came from transcoding?
Of all the major content owners I spoke to, none of them really knew an answer. Most content creators are using automated solutions from FlipFactory or Sorenson Media to rip derivatives so it was hard for them to know. Also in some cases like at ESPN the carrier picks up that cost, not the content creator. I’ve quickly found that there is no real industry "standard" for this and the business models vary greatly. Some only do the encoding, others do hosting, others try to do a end-to-end solution. The trend I heard most is that it seems to be going towards end-to-end managed services just because of the complexity.
However, the efficiency on content delivery is important when wireless data access becomes more ubiquitous. We are still dealing with a low percentage of mass market users accessing wireless data in general. This percentage will increase as bandwidth increases and wireless solutions actually give an end user a compelling reason to use it beyond email and web browsing. That application has yet to be found. I don’t doubt that there will be an application or a solution that will stress the wireless network, but this will create opportunities to architect a client/server solution to be more efficient which is what most are doing with their technology today. You can also see this now with Sling Media (unicast) and MobiTV (multicast). These client/server solutions throttle data rates to address network throughput, which is a start, but by no means a total solution.
I also asked content creators how they would respond to a much-simplified pricing model of say 2 cents per minute delivered, regardless of bandwidth? Overall consensus was this would not work, but it depends on the region. The U.S. is all you can eat. Carriers have drilled this concept into the social consciousness of the US consumer. Wireless access is no different. The hotel or airport paid access model where you buy minute/hourly access to wireless is an interim solution to a longer term ubiquitous access to wireless all you can eat usage. Europe may continue its pay as you go or pay per MB for some years to come. I am hearing rumblings that the all you can eat model for home Internet access is changing, but wireless is still an issue with wireless plans not cross country accessible. There are carriers that are attempting to shape a new model, such as "3" in Europe. 3 has shown that it is willing to work with Sling Media and embrace the solution rather than fight them, like others are doing here in the States.
Also, models that adjust Gbps to cpm or pay per minute have not taken off. Most people adjust those number back to Gbps and compare those numbers. The contracts that go along with cpm or per minute pricing are very non standard and have issues making their way past legal. For example A CDN needs to say video encoding rates can not exceed 300kbps. No one wants a clause like this. Is there a similar model out there now? Movies are by ticket. TV is by monthly subscription. iTunes is pay per unit. On Demand is pay per play. Maybe an "EZ-Pass" model might work where you prepay a debit account that slowly depletes but that would be a big shift.
Another question I asked content creators was, are encoding/ingesting costs smaller, larger, or similar to delivery costs? From a mobile perspective, where the actual quantity of bits delivered can be quite small compared to the size of the original asset. This was the only question where everyone agreed on the same answer. Today the encoding/ingestion cost is much bigger because you need to support a lot of handsets, however it should be lower. Once you get to more than a couple of formats, most content owners invest in encoding automation. Formats add storage costs and at least 200k worth of encoding hardware. Delivery is always a much larger cost and sites with no traffic and lots of content are the exception but those tend to go out of business. The costs should be lower due to smaller file size but that is only if the format is a standard format. Most encoding systems have a high sunk cost and then maintenance can be spread across formats. Licensing encoders can raise costs. Farming out encoding should be by size as long as the encoder has the format licenses. But it also depends on the content. For wallpapers and JPGs it is really small. For ringtones is can be small as well, since many carriers limit the payload to a few hundred kilobytes since they deliver this via WAP sites or WAP push or MMS. There are few mobile devices, like Treos, that allow for a full browser to access the Internet and download anything. If video can get to the point where ringtones and wallpaper is from a delivery standpoint, then it can be successful with low production costs.
For my final question to the content owners, I asked if mobile delivery commands a premium due to the more difficult nature of reaching the end user? This was an interesting question. Some of the CDN providers said yes, we charge 10x more for mobile delivery that Internet video delivery which was not a surprise, since most CDNs are not setup to truly deliver mobile video. But on the other hand, the content owners said it should not cost more to deliver video to mobile and that perception is what the carriers want you to believe. Content owners are not buying it.
Some content owners we’re not really sure either way and said that from one perspective this could be right. But there is also content you get access to that you do not get on the Internet and there is also specific content that you get just for mobile use i.e. based on the usage patterns and that you want to get breaking news and info based on presence that a mobile phone/device is well suited for. I think going forward you will see more "premium" content on wireless terminals and specific services that will you will pay a premium for just because it is easier for the content providers to control the deliver as compared to the Internet so they are more compelled to build a "wireless" service.
My take from all I spoke to is that I think it commands a premium now, since the wireless network is not leveraged across as many users. As an alternate example, the cable operator (Comcast) infrastructure can leverage VoIP, On Demand, basic cable, and Internet access. That is a good use of an ongoing cost infrastructure that needs to be updated over time. The wireless infrastructure is unique that it requires broader reach of connectivity and roaming with a consistent bandwidth. Therefore wireless carriers have to provide a broad infrastructure for each customer, rather than a unique home access to each customer with respect to cable operators. However, if you look at the current all you can eat wireless plans from Sprint, the $15/mo is not that bad for EVDO, since they don’t price differently for 1xRTT or EVDO, it is still $15/mo. That may change, but it is not bad for ~300kbps on a wireless network. However, I get ~20Mbps on my Verizon FiOS for $49/mo. The price is still pretty high for mass market adoption, but for that bandwidth I can do a lot, if it was offered.
The more I speak to content owners about their mobile video offerings, the more I realize that there is just so little being done today. When I asked the major content owners what percentage of their inventory is available for mobile or how many videos/streams they are delivering to mobile today, none of them knew or would answer. And I got the sense that they were not trying to hold back the data, but more that they really just did not know, which means that today, it’s not a big part of their business.
On the CDN front, all of the ones I spoke to said that today, mobile video delivery does not take up even 1% of their business from a revenue or bits perspective. When content owners don’t see much traction, and CDNs don’t see much business, it’s clear that mobile video adoption in the U.S. has not changed much in the past few years. Video to mobile has so many business and technology hurdles to get through that you have to really wonder if it will ever get off the ground in the states in any truly mass-market adoption.
While there were just too many to name and quote, many CDNs, major content owners and mobile video vendors contributed to the information in this post.