Netflix DVD Shipments Vs Instant Watches – revisited October 2009

23 Oct

In our first post to kick off this blog back in February 2009, we looked at the impressive growth of Netflix’s Instant Watch offering. In light of Netflix recent quarter results, let’s re-visit that.  Eight months later, here’s what things look like:

Netflix DVD Vs Instant Watch Oct 2009

Netflix DVD Vs Instant Watch Oct 2009

Growth seems to have accelerated in 2009 with  Streams at about 150% of Shipments this year! It also looks like the number of DVD Shipments in 2009 will not be that much greater than in 2008 but it *will* still exceed 2008 shipments. It will probably take a few years to level off before the curve starts to bend downwards as far as Shipments are concerned.

The Streaming curve on the other hand is quite insane and in 2010, Netflix will likely spend more time and money worrying about ISPs, dark fiber, Net Neutrality and bandwidth caps than they do about the USPS and DVD theft.


Posted by on October 23, 2009 in feedfliks, feedflix, insant watch, shipping, streaming


4 responses to “Netflix DVD Shipments Vs Instant Watches – revisited October 2009

  1. Dan

    October 23, 2009 at 6:50 pm

    Do you normalize these numbers based on how many episodes are contained on the DVDs? I’m not sure how easy that data is to obtain from Netflix. If you don’t adjust for this I believe the numbers are going to be skewed.

    It seems like a significant portion of the streaming content is single episodes of series. I’m assuming that these streams will each be treated as a single viewing, but the corresponding DVDs would be shipped with 3-5 episodes at a time. If that’s the case then the same content streamed counts for as much as 5 times what it does when shipped.

    Perhaps it would be easier to filter out multi-disk and multi-stream content. Then you could build a model that compared things on level playing field.

  2. feedflix

    October 23, 2009 at 6:57 pm

    Dan – no, we don’t do that currently but will look into ways to do that based on the data we have.

    Still, whatever skew that adds doesn’t affect the yr-to-yr streaming curve (same offset in each year) – and that is *some* growth you have to agree!

  3. David

    October 27, 2009 at 5:19 am

    Question along the same lines as Dan – the normalization of the data. Are the Instant Watches indicative of actual content viewed? Meaning, if I start a movie and watch 5 min get bored and switch, does that count as a separate view? What if I stop it and come back?

    There is growth for sure, but it is kind of like reporting views on Hulu (which I hear is going to charge in 2010) vs. YouTube without accounting for the fact that Hulu was much longer content that YouTube. Just a thought.

  4. feedflix

    October 27, 2009 at 6:22 am

    David – perhaps I should do a follow up post on how ‘sticky’ Netflix’s IW titles are (for now see today’s post at but let me say that even if we were to count only those IW views that are non-trivial (i.e. leave out the 5 minute bored/switch events that you mention), the number of titles streamed is still a) greater than DVD shipments in 2009 and b) growing just as sharply.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: