One of the flies in the ointment in open or IP based video on demand, is that Netflix has been giving it a moving target. Netflix needs little or no technology, depending upon whether or not you are its customers, or you are Netflix. While broadband delivered films need more than one technology breakthrough.

The Netflix process is simply bulk buying of DVD films from the moment they go to DVD, hold them in shipping centers around any given country (the US in this case although they have said they will hit the UK and Canada this year) and let people order them online, and use the post to ship them both ways, guarantee delivery in one-to-three days. Extra attractions are charging no late fees, unlike most video rental stores, allowing each household to have a set number or DVDs out at any one time, bundling the price for a single monthly subscription, and ship DVD’s to and from the home as part of the price.

If a video on demand service cannot beat that, it cannot take off. At least it can’t take off in the US, where Netflix has done a good job and where it has a head start.

It is conceivable that outside the US, where access to video rental is perhaps less widespread, a broadband service could lead a video on demand service, but it would have to be a download service and why would Hollywood give such a service access to films?

No, it is a simple fact that at the moment Netflix takes between one-to-three days to get any given film to a home in the US, and this is largely due to the fact that the US is such a large geography. If a similar service was launched in parts of Europe, even the local postal service could deliver packets next day.

If we compare this with both Amazon and Kazaa, it compares well. Amazon usually manages to tie, roughly, with this delivery speed, and a file sharing service such as Kazaa, even given the fact that many of its customers swap illegal and therefore free files, would find that speed hard to beat. If someone left their broadband line on all night, or all day while they were at work, perhaps next day is a reasonable speed for a full film download.

Faultline has been asked before can films be delivered to a home, over a broadband line in under three hours. We pointed out that the H.264 codec, and ADSL2 chips due this year are two ways of getting this time way down, and even new versions of the IP protocol are rumored to be set to make an appearance to improve speed. But as things stand, three uninterrupted hours is what you probably need to download a film on a dedicated broadband line.

Why wait any of this time, why not drive two miles to the local video rental store and rent a film? We all know the answer to that is that it won’t have a choice of 15,000 films as Netflix has; and even if it did, can you imagine how long it would take to find just the film you wanted, especially if you insisted on finding one with actors you knew, a director who’s work you admired, in a genre you enjoyed?

All of these obstacles to watching DVDs any other way have propelled Netflix to a $270 million business, in profit, with 1.8 million customers and on a growth curve of 80 per cent. This will see it become a $1 billion organization by 2006 or 2007.

Its penetration of the US market is under 2 per cent, while its original market in the San Francisco Bay area is now running at 5.9 per cent. It reasons that it can reach this penetration, perhaps more, across the US and it is this calculation that initially led it to say at this week’s unveiling of its fourth quarter figures, that it will make the $1 billion mark early.

For 2004 Netflix is expecting to go over $450 million and get its subscriber base up to 2.5 million. CEO Reed Hastings said at the results conference “that it is DVD acceptance that is the rocket propelling studios profits,” and he said he was now buying one million DVDs every quarter from 100 film studios and sending two million packets out every week.

But the driver behind the Netflix business cannot just be a good idea, and Hastings is keen to place the credit with his company’s execution. With at least 100 copycat organizations operating in the US, some of them run by Walmart and Blockbuster, the original Netflix still claims to take 95% of the revenue in this market.

Perhaps he is right, and perhaps those bigger competitors have not yet had a chance to shine, but Netflix is certainly not under threat from a maze of smaller organizations that have copied the original idea. Web advertising and word of mouth drive most of Netflix’ new customer growth, but the heavy TV advertising must give new customers a warm feeling of having a brand behind the business.

Perhaps it is because of this that Netflix has managed to drive its churn down below 5 per cent for the first time in the quarter just gone, and intends to keep it there. For competing businesses, this will be poison, to always be compared in their execution to the market originator, and to have to pay more to acquire customers, only to lose them in a shorter period of time.

Part of Netflix’ future plan is to continue to put more pressure on that window of opportunity for IP delivered video on demand. It will drive up and up the percentage of households in the US that can get films from it next day, and it should achieve this for 80 per cent of customers, sometime during 2004.

But if driving to the shops for a poor selection of DVDs at the rental store is one step too far for most people, then there is something about not even having to remember to post the DVDs back, and being able to select programs on your return home from work, that are watchable that same night – and that’s video on demand.

Netflix would give no more than a few tiny clues that this is in fact the company’s destiny. It said that its business would continue to be DVD based for the forseeable future, but that was working on a film download service that it would test in 2004 and deliver during 2005. The idea would still be ordering films online, but then opting for DVD physical delivery or online, whichever the customer chooses. The company feels this would have a modest uptake
during 2005 and then grow over the next 10 years.

The company frequently points out that the data which Netflix ships on a busy day on plastic discs is about half the total US Internet capacity at around 5,000,000 gigabytes per day. But what it doesn’t point out is that the internet capacity is virtually doubling every year and that with newer compression, in a year or two the entire Netflix shipments might be only a tiny fraction of the US internet capacity, and anyway, we have always expected that video is the killer app for broadband, so it should be a high percentage of what travels over the net. Hastings also pointed out that Netflix is already talking to studios about this and that given his $100 million a year spend, they were listening, but declined to say more about the exact nature of the Netflix future VoD offering and said the company will not discuss it again until it is ready for launch.

There is one irresistible gain for his company in all of this. When a company rents out DVDs, and when it boasts that it has 15,000 films in stock, then it has to lay out cash for a huge number of films just to have them sit idle for much of the time. With 1.8 million customers, most of them holding 3 DVDs at any one time, that’s a minimum of 5.4 million DVDs either on the way to, with, or on the way back from, clients at any one time.

But online downloaded films are copies, and there’s no postage and no up front purchase of all the hard DVDs. Or is there? If Netflix could lose those costs, its yields would sky rocket and the company’s net income of just 2.5 per cent of revenue could double or treble overnight. The image of a re-entrant film copy that sits in a machine’s memory and which can be sent to any number of customers at any one time, is how we all view Video on Demand. It may not have to be viewed that way if you are Netflix.

Could Netflix have in mind a system where the DVD itself is the version of the film that is being played on a remote device, but that it is reliant on the remote device remaining connected to the DVD via Netflix in order to initiate the playing of the film. If such a system was built it might mean that there was no residual online version of the film remaining on the remote machine that could be file shared around the planet. This might mean that Netflix would still have to buy the DVDs, but at least they could sit in server farms, always ready to be used.

But really this is fanciful. Good digital rights management on a server can emulate that effect and really there is no special edge that owning the DVDs themselves can give to Netflix. And anyway, what device at the other end of the broadband line is it that the DVD is downloading to? A PC? Well that’s no good if Netflix customers want to watch on a large TV screen. The numbers of TV households that already have DVD players is high in the US (and elsewhere) but not that have their TVs set up to receive from broadband, at least not yet.

So what special edge can Netflix bring apart from its buying power at the studios?

Let’s see, could it download its films to a set top? Well companies like Motorola and Scientific Atlanta that lead the set top market, have devices that terminate cable TV systems, and they both have models that can act as an advanced Digital Video Recorder, and these can have smart card based conditional access systems in them to protect content.

But neither of these two companies is going to feel too well disposed towards Netflix. It is single handedly denting the numbers of TV programs that its customers are watching, driving down the advertising of their biggest customers, cable companies like Comcast. If Scientific Atlanta befriended Netflix, would it suddenly lose out on the next juicy batch of set top contracts?

Then perhaps it is the independent DVR makers that might make good partners for Netflix, and a good place for its downloaded films to play and even be temporarily stored? And who better than the market leader TiVo?

TiVo has over one million customers in the US, a similar number to Netflix. It has a monthly subscription model just like Netflix. TiVo is supporting movie studios with its new long form, voluntary, DVR advertising. And TiVo is not best of friends with the cable TV companies, given that it is a lead supplier to DirecTV, a company that has come out recently in praise of DVRs. And TiVo technology based DVRs are in the stores already from some of the consumer electronics giants.

And it just so happens that Mike Ramsey, the CEO of TiVo, sits on the board of Netflix.

Just an idea, not so much a conspiracy theory, but perhaps it is the equivalent in the software industry of Bill Gates of Microsoft and Larry Ellison of Oracle sitting on each other’s boards and hatching plots.

Whatever the IP VoD system that Netflix comes up with, it must be confident that it currently sits in pole position in controlling the switch from DVDs to IP TV VoD films, and that gives it the potential to be the iTunes of the film industry, as long as it can find its iPod.

© Copyright 2004 Faultline

Faultline is published by Rethink Research, a London-based publishing and consulting firm. This weekly newsletter is an assessment of the impact of events that have happened each week in the world of digital media. Faultline is where media meets technology. Subscription details here.