[majani]

I've always wondered whether VOD is an inherently unsustainable business model. The big daddy in the space charges $8-12 per month yet people regularly watch upwards of 10 movies per month on the service. That's less than $1 per movie?! Also, this forces anyone doing a similar business to charge even less than that to break into the market, so there reaches a point where we're talking fractions of pennies per movie on some services. How is any of this sustainable given the massive cost of acquiring the content in the first place, and is there any VOD service with healthy profits that you can point me to?

[sp332]

Once you've acquired the content, the marginal cost of a view is very low. So it makes sense to get as many customers as possible even if that means charging a lot less per view. And the stability of subscription income is nice too. I'm sure there are lots of people who have a $10/mo subscription going for years who hardly use the service at all.

[toast0]

I haven't seen any data for video content, but I imagine some content is pay $$$ to steam to as many users for a limited time, but much of it is probably pay $$$ to be able to stream it, plus $ each time it is streamed. And I imagine some deals are based on a share of revenue. I wouldn't be surprised if content was significantly more expensive than bandwidth.

[tomphoolery]

It's expensive to do VOD right, because videos are such huge pieces of data. Plus, the "user experience bar" that Netflix, Hulu, and YouTube have set are pretty high and actually require a fair bit of engineering. There's no "15 minute blog demo" for VOD. The last company I worked for started out in that space, but had to pivot to doing linear television because the market just wasn't there. They built some of the same infrastructure that cable companies now use to do their on-demand services.

The only VOD services with healthy profits are actually content creators themselves. All of the services I just mentioned in the other paragraph have their own original content, and at least Hulu & Netflix are pretty successful at that. This original content keeps people on their service, generating more ad revenue, which is how they make money. Basically, if you're a major studio and have billions of dollars to throw away, VOD is pretty easy ;-)

Source: I worked for a failing VOD/broadcast video company for a couple years.

[sandGorgon]

what is the kind of infrastructure that needs to be built ? Very curious to know what are the big tech challenges if someone wants to build a VOD startup in a niche.

How did Viki make it ?

[dylz]

Here are some tech challenges, this is strictly my opinion and shouldn’t be read too deeply, just a casual observation during lunch.

My background is (as a hobby) I’ve built medium-sized >700m PV/m esports analytics apps targetted for Asia, covering CN/most of SEA, and have had to get local connectivity and local hosting as an "outsider"

These countries are fairly insular.

China: GFW (we all know about this), ICP license, network pricing (China Telecom, Unicom, CN2, etc) for international traffic. It’s extremely difficult or impossible to serve Chinese customers from outside of China, especially if you need stable high bandwidth transport (and STUPIDLY EXPENSIVE). Almost all network peering ports in Los Angeles for CN/CU are oversubscribed to death, and CN2 traffic is disgustingly expensive internationally. It’s generally not possible to do anything at all without a “local partnership” type arrangement.

Korea: GFW (Korean censorship is heavy and extreme; there is really no other way putting this, people think SK is very “western”, it is not even close)[0]. You might see news about how Korea has fastest home internet - Korean connectivity is crap as soon as it leaves the country; the country is effectively a LAN, hosting outside of Korea is awful to inside, not possible to compete like that when all your assets are outside the country, and getting hosting is a bitch, and cost is hilarious.

Korea has KYC/“real name verification” laws like China, things like needing to get KSSN via a credit bureau via iPIN/mobile phone verification (since mobile phones also require KSSN to get; and <18s mobile phones are keylogged anyway) collected at registration if you want to do anything social for the most part.

The smaller countries in SEA tended to be a "if you don't speak the language, if you aren't from here, then fuck you, enjoy your highway robbery" to me. I paid bribes or else hardware would "go missing" or things "were impossible to do unfortunately", I paid prices significantly higher than any domestic purchasers would, the companies that had English sites you could literally see the prices have an extra decimal place just switching between the two versions.

I had to do a bunch of logic wrt balancing traffic. I had a server that had multiple physical NICs, each connecting to a different ISP. They don't necessarily like each other, so if I served a PLDT end-user from a Globe connection instead of PLDT, traffic may get forced intentionally through another country. There are incumbent ISPs that control a majority and charge far more that intentionally refuse to peer with any other ISP, and force domestic traffic out of the country to punish those users for not using them.

My site showed user profiles, which obviously includes their usernames. I had physical hardware shut down and removed by police in Thailand because someone had a username mocking the royalty there.

edit: I forgot my [0] - https://www.eff.org/deeplinks/2011/08/south-korea-only-thing... - the person involved is actually on the censorship board

[reaperducer]

I feel ya. You have no idea how much I've paid in bribes to do even basic things in Thailand.

And a company I used to work for frequently had the same problem with Mexico. Traveling two and from on business led to frequent searches and impounding of very expensive equipment unless a bribe was paid. We started writing it off as an "airport tax."

[sandGorgon]

This is mind boggling. Thank you for sharing this. I never knew that internet and traffic insularity was a thing outside of China.

Especially because this is such an alien concept in India.

[TCM]

Maybe crunchyroll? https://en.wikipedia.org/wiki/Crunchyroll http://crunchroll.com

[jetti]

I imagine that one could do well if you break into a niche that isn't really represented in other platforms. If you could make an indie VOD you may be able to get content cheaper and then you can market to those who like indie films.

[shubb]

Or old content.

Spotify provides access to a huge back catalogue and although it probably reduces the value of some content it monetises other old content that was basically dead.

Currently Netflix has a tiny library, when you take into account 50 years of film ripe for rediscovery.

Anyone know what the blocker is?

[tomcam]

The blockers are that old content that people are willing to pay for is more expensive than you think (tons of competition right now), and that rights management is incredibly expensive and Byzantine even for the big companies. To take the simplest example, what do you do when an attractive property is owned by a company that went out of business, or that can't be reached? And that's just the easiest scenario.