Netflix: Two Weird Facts You Should Know, But Probably Don’t


Download article:

Cable operators, satellite operators, mobile operators, music service providers and most other companies spend a lot of time thinking about APRU, or average revenue per user. And investors are equally focussed on ARPUs, too.


Because it’s generally bad news if your customers start spending progressively less and less with you. This might mean that they are spending more with competitors or it might mean that the whole sector is being affected by a downward trend in expenditure, which might be because the market is maturing or because a recessionary phase has reduced consumer expenditure across the board.

All of this is bad: the idea is to get your customers to spend more, not less, which means that it is really important to keep ARPU’s growing or at least to stop them falling.

While it is important to minimise churn, maximise net additions, boost operational efficiencies and manage fixed costs, company revenues are the first place any investor or analyst will look – and because revenues = (number of customers ) x (ARPU), ARPU will remain a critical indicator.

But something odd has been happening to Netflix’s ARPU.

Netflix’s ARPU crashed by nearly 27% in 2011, and has not recovered but investors couldn’t care less (so it seems)

Something odd has happened to Netflix’s ARPU over the last few years: ARPUs have fallen. A lot.

If we take Netflix’s total revenues and divide that number by the total number of paying subscribers we get a figure for Netflix’s overall ARPU, which we can see has fallen from around USD 12.15 in 3Q10 to around USD 8.40 in 2Q13, a drop of nearly 27% - which occurred over a 6-month period as shown in the figure below:

Title:Netflix: Average Revenue per User (ARPU)
Markets:Television, Internet, OTT Internet TV
Years:2010, 2011, 2012, 2013

This would be an extremely serious matter if you were the CEO of, say Vodafone. Or Comcast.

But not if you’re Reed Hastings, the CEO of Netflix.

In fact, if we now compare Netflix’s stock price with the historic trend in ARPU then it seems the market is very laid back about Netflix’s ARPU crunch. The chart below shows that after a sharp drop around 3Q11 – due to in large measure to the Qwikster fiasco and associated haemorrhaging of US subscribers – Netflix’s stock price recovered strongly.

Even over the 12-month period between July 2010 and June 2011- when ARPUs were steadily falling, Netflix’s stock price was rising steadily.

Title:Netflix Comparison of Stock Price with ARPU
Markets:Television, Internet, OTT Internet TV
Years:2010, 2011, 2012, 2013

We find this strange: if you are operating in a mature market, like mobile communications or cable TV, then it is clear that ARPUs are going to be under pressure: price competition will be fierce and the battle will have long since shifted from acquiring new customers to prising customers away from competitors.

But Netflix is a growth company operating in a growth market segment.

So why has Netflix’s ARPU fallen?

Netflix’s original pricing model for its DVD rental business was based on a flat rate service charge (membership fee) with a usage-based element which was charged on a per DVD basis.

This model still operates in the US (the only market where Netflix offers a DVD rental business): the membership fee is USD 4.99 per month with incremental revenues based on the number of DVDs rented.

This means that the more DVDs you rent, the more you pay.

Netflix’s new pricing model for its streaming business is based on a fixed fee of USD 7.99 per month (US), with the same model being used in all other operating markets.

This means that the company has purposely removed any possibility of earning more revenue from subscribers who use Netflix more – even though the costs of providing the service to those subscribers will be proportional higher.

Why is this important?

In other markets, say Pay-TV, mobile data, or fixed line voice, the early growth phase of the market’s development has always been characterised by a usage-based pricing model.

So, in the case of the Pay-TV market, we originally saw Pay-per-view pricing models. For fixed line voice we paid by the minute for years, decades even.  For mobile communications we had a price per text message, a price per voice minute and, most recently, we had a price per MB for data service.

Generally, as these markets developed, and as competition became more intense the focus shifted from acquiring new subscribers to retaining existing subscribers. In a bid to attract customers away from competitors, market participants were under increasing pressure to get creative.

So they started introducing flat-rate price plans, where you pay a fixed fee per month and then use the service on an ‘unlimited’ basis.

In some markets the first flat-rate plans were introduced by maverick new entrants who wanted to shake things up by offering something new, but in other markets, intense competitive rivalry forced the market into offering flat rate plans.  

But Netflix has leapfrogged the usage-base pricing model and gone straight for a flat-rate pricing model.

We think this is even more remarkable because Netflix is the runaway market leader in OTT internet television – with a total of over 36 million subscribers at the end of March 2013: it’s not as though Netflix is under severe pressure from rival OTT services who have introduced flat-rate pricing models. In fact, you almost need a magnifying glass to see the rivals (for instance Hulu, with barely 5 million subscribers).

We are left with the conclusion that the only credible reason why Netflix has adopted a flat-rate model is because it decided to do so, perhaps thinking that this would maximise the uptake of the service.

This brings us to the second fact about Netflix’s business that we find weird:

Why is Netflix charging USD 7.99 per month when the competition is charging USD 73.00 per month?

Comcast’s SEC filings show that the video ARPU for video customers achieved over the last three years is:

  • 2012:  USD 76.18 per month;
  • 2011:   USD 73.34 per month;
  • 2010:   USD 70.77 per month.

The equivalent figures for Netflix are:

  • 2012:  USD 7.90 per month;
  • 2011:   USD 9.70 per month;
  • 2010:   USD 11.17 per month.
Title:Netflix vs. Comcast Overall ARPU
Companies:Netflix, Comcast
Markets:Television, Internet, OTT Internet TV
Years:2010, 2011, 2012

So not only is Comcast achieving 10x more video revenue per video customer but it has managed to increase video ARPU (in a mature market) while Netflix’s ARPU has fallen (in a growth market).

Seriously, does this make sense to you?

It sure doesn’t to us, and the story makes even less sense we look at what has been happening to subscribers: Netflix’s customer base is still growing strongly, where Comcast’s customer base is falling:

Title:Netflix vs. Comcast Total Video Subscribers
Companies:Netflix, Comcast
Years:2010, 2011, 2012

So even though Comcast lost 800,000 video customers between 31 December 2011 and 31 December 2012 (about 3.5%), the fact that the company has managed to persuade the remaining customers to spend more has allowed the company to actually increase overall video service revenues.

The only sensible conclusion is that Netflix is underpriced.

It could be argued that Netflix is competing with free, that is, with movies that can be downloaded for free from copyright-infringing sites. While that is true, the same is true of all Pay-TV providers who are engaged in a constant battle with such site, so this does not explain Netflix’s throwaway pricing model.

It could also be argued that because Netflix’s fixed costs are less than, say, a cable TV or satellite TV operator then Netflix should be proportionally cheaper: everyone knows that, in the long run, prices will fall to the point where they just cover marginal costs.

But this is wrong.

While it is clearly more expensive to operate a cable TV network the size of Comcast’s than it is it for Netflix to operate its own infrastructure, it is not that much cheaper: we are talking about a cost difference of about USD 30 x (70 – 7) x 12 billion per year, or about 22.7 billion per year – far more than Netflix’s total revenues. So this cannot be the reason either.

We conclude that Netflix has deliberately (perhaps unintentionally) underpriced its service and has, therefore, needlessly destroyed shareholder value. We are reminded of the tobacco executive who once said that “pricing in this market is set by the dumbest competitor.”

Of course, we are not saying that Netflix is dumb: we think the service is generally excellent and that Netflix has done an excellent job of migrating from DVD to digital while becoming a worldwide market leader.

But we do think that the company’s pricing strategy’s needs some work.

It is one thing entering a new market with a competitive price, but it is quite another to pitch a price that is so low that it’s completely suicidal.

At the very least we think that Netflix should think seriously about introducing a tiered pricing model, which at the very least would increase overall company revenues: the reason why most online services use a tiered pricing model is that this is the best way to address different segments of the market which have different price elasticises.  A one-price-for-all model will maximise customers, but not revenues or profits.

However, still smarting after the last pricing mis-strategy, it might be a while before Netflix feels brave enough to change the pricing model again.

Contact Andrew Sheehy at or follow him at