Friday, March 18, 2016

My Take on "Take-Rate"

Passengers are asked more and more to pay for differentiated, itemized services.  The concept of take rate is to price a product or service such as to maximize profit, on the premise that the higher the price, the less the number of purchases.

There is a discontinuity between "free" and "you have to pay something".  It is like buying a lottery ticket.  Your odds to win go up dramatically once you buy your first ticket.  When the credit card comes out, any sale becomes a possibility.

I first heard Symonty refer to this non-linear behavior as a "call to action".  The passenger has to want it so much that they will dig out their credit card (maybe in the overhead), reveal it openly to prying eyes, and trust that the service will not create a pathway to unsavory transactions.

Paying from your seat is more commonplace today, with food and drink purchases.  Food and drink are on a different value system than entertainment.  You see what you are going to get, and there are multiple senses in your body triggering your desire.

There is a precipice in take rate between a free product and a passenger-pay product that begins at about 10%.  No matter how compelling, you won't get more than 10-15% of passengers to buy an entertainment product (on average).  Some flights will see far higher engagement, others will see much lower engagement.

The fundamental concept is delivering a service wirelessly in a Bring Your Own Device (BYOD) framework.  BYOD refers to smartphone, tablet, and laptop.  The service may be hosted primarily on-board (cache) or remotely through a tandem wireless connection between the airplane and the ground.  The connection to the ground may include broadcast (multi-cast) channels shared between subscribers and interactive (uni-cast) channels dedicated to each subscriber.  Multi-cast is a special feature of a given radio network along the lines of satellite direct-to-home television. Passenger uni-cast traffic is comprised of Internet access, to be able to reach public IP addresses and to operate a variety of protocols.

Internet access was originally the path for browsers and for email servers.  Connected applications utilize the same IP network for other services, operating somewhat independently of the user behavior and using sophisticated protocols.

A free product can easily gather more than 10% take rate, but only a truly exciting product can bust through 30% average usage.

With a free product, the number of connected devices could exceed the number of passengers, as each may carry more than one device.  Serving a user with multiple devices creates a greater burden (usage).  Offering to upsell a multi-device service offering is one way to offset the lost margin.  In a free offering, multiple devices simply add more contention.

There is more to life than BYOD wireless entertainment. While take rate will increase over time, it is likely that about 20% of the passengers will never engage.

Having multiple products, a mix of free and paid, will reach the greatest percentage of passengers.  No single product will engage more than half the customers.

Product Portfolio

Bring Your Own Content (BYOC) - titles stored on a passenger's own device itself.  This is a category of Bring Your Own Device (BYOD).  A book or magazine is an example of BYOC.  More relevant are several initiatives to offer content downloads prior to departure.  This category includes on-line content providers such as iTunes or cable providers such as XFINITY.

An emerging subcategory of BYOC is oriented around offering a pre-flight download catalog as part of the ticket.  For example, GEE offers airtime "content to go".

Regardless, BYOC is not accounted for.  The ability to securely carry content onboard remains a viable category of entertainment.  Books, newspapers, magazines, catalogs, and audio entertainment (podcast, music) are other products are vibrant products to a large segment of the population.

Web Connect is a generic title for a sluggish, congested Internet access. It is tolerably useful and popular if free. The same people, and a few more, would use a better Internet access experience.  In my judgment, interactive applications get boring after a short amount of time.  Working in a cramped seat is uncomfortable.  Travel is stressful.  People want to rest, to relax; knowing the world is in abeyance while the door is closed.

Streaming has evolved beyond Netflix.  My personal cable service includes complete remote access to any recorded show and the ability to download most of them.  I have subscriptions to any number of content providers with their own on-line catalogs.  It is nearing the point that any television show can be watched on-demand.  Cutting the cord does not mean abandoning cable providers, but it does mean accessing your content when you want, where you want, and how you want.

Judging by the standard everyone uses, the only exciting application right now is streaming entertainment.  Streaming entertainment can reach 40% or more of the passengers if in combination with a compelling Internet access as a free offering.

Cached audio and video catalogs can be specially attractive if free and built around a special theme; or unique, exclusive or early-window content.

Text refers to managed applications that are mostly character based, such as SMS text messaging. Now this category reflects a managed interface to a walled garden or special "free-ways" to social Internet properties, for example to Facebook or Twitter. The ready access to mobile cameras makes dealing with images a welcome addition to a text-based product.  By-and-large, these limited products are very useful to a specific audience, while the broader population yearns for unfettered access.

Broadcast television is a very popular free product, because it is inherently engaging and contemporaneous.  Broadcast radio still has a niche for the same reasons.

Linear VOD (Mobile Prime Time) is a variation of live television, where the content catalog and commercial inventory are mixed locally on the airplane.  Breaking stories come up in vignette form.

Streaming entertainment has access to the biggest catalogs. Bring-Your-Own-Subscription (BYOS) is the emerging trend.

Streaming phone is 360p or less quality (800 kbps or less).  Looks fine on a tablet.

Streaming Blu-Ray is 1080p or greater (4 Mbps or more) - a service I don't foresee coming to the free world, but would be the most popular service if it did.  There will be steps along the way.

I judged the pay-streaming phone to be more popular than pay-Blu-Ray because I expect it will be cheaper.  I assume a streaming product, pay by the minute, will have only modest engagement. 

After eating, talking, working, reading, sleeping, bio-breaks - comes the reward.


Every Internet access session is a unique experience generating a unique amount of usage (as measured broadly in MBytes). Light, average, heavy and future make up four categories. For each usage scenario, there are three use cases: all streaming, all Internet access, or split evenly between the two. The session duration and the amount of time off-line is another dimension. A short session will have less off-line time than a longer session.

Every customer may represent a smartphone and at least one more device (tablet or laptop), which increases their usage when they purchase the product. 

Judgemental Conclusions

I propose that there are actually two barriers:
  1. Free Versus Pay
  2. Useful Versus Exciting
The "Exciting" category, the thing everyone talks about, is live streaming - the "Netflix" experience.

These two new barriers are judged to be at 10-15% and ~30%.

"Price" is replaced with "Experience" on the Y-axis to convey the capacity of a given service.  You may surmise that all services below a given service are available at each level.

How will these change over time?  My guess the products will evolve and become more capable, but that is in the context of the broad mobile world.  As was stated by more than one speaker at Satellite 2016, the products will be more capable (as in bps), but the ARPU (average revenue per user) won't move.  You have to deliver more and more over the course of time to hold onto your market share.

Products making up the experience will evolve, but their take rates will struggle to cross the upper limit of paying, and only truly exciting, free entertainment can get anything close to a majority of the passengers engaged.

Airline passengers have many different needs and desires, and these change through the course of a journey.  Heading to a critical meeting has a completely different dynamic than returning home from that very same meeting.  Web applications (Email, browsing, file transfer) might be critical on the way out; while a celebration on the way home might call for champagne and a movie or favorite show.  

Satisfaction comes from delivering each service, in a compelling manner, with clear value.  Profit and loyalty, the point of maximizing take rate, come from having the right product available for each and every passenger.

Peter Lemme
peter @

Copyright 2016


The following are excerpts, with some commentary, as references to take rate, starting from the lowest going to the highest.

Gogo Vision (3% Pay)

Working the math, it is calculated that Gogo Vision take rate is less than 3%. /gogo-vision-producing-interesting-dollars-for-content-partner/

Northern Sky Research (7% Pay)

NSR reports that take rate for Ku band is higher than for Gogo ATG, but still around 7%.

Gogo (8% Pay)

Gogo reports take rate is under 8%.

Virgin America - Gogo (2009) (15% Pay)

Virgin America reported breaking through the 10% barrier, to as high as 15% average take rate for its Gogo pay service.

As expected, some routes are very popular, and even today might exhibit as high as 30% take rate for a pay service.  While useful for provisioning an airplane to serve such a demand, as a business you you have to account for the entire route structure when comparing take rate. virgin-america-discloses-gogo-usage-rates 2009

TMF Associates (8% Pay & 30% Free)

Tim Farrar puts it into two categories, with up to 8% take rate for pay and 30%+ for free

Norwegian Airlines (28% Free)

It is calculated that Norwegian free Internet service take rate is 28%.  In their infographic, they state 33% of the passengers use it (social networking).

ViaSat (40% Free)

ViaSat reports 40-50% take rate, with a peak of 148 devices connected on one flight. 
...more than 300 commercial aircraft in service, engagement levels four times the historical rate

More than four times the average number of passengers using the service on each flight compared to the nearest competitor, with take rates on flights more than three hours approaching 40%.

...said Buchman. “On JetBlue, where they give it away for free, we typically see 40 percent to 50 percent take rates on the long haul flights,”

The service has the best passenger take-up rates in the industry; more than four times the historical rate, with as many as 148 simultaneous devices connected during any given flight.

Compared to traditional take-up rates of less than 10 per cent of passengers, the uptake is averaging around 40 per cent on JetBlue flights using the ViaSat service, with a record 148 simultaneous personal electronic devices (PED) connected on a recent flight.

GEE (80% "Free+Pay" Mix)

GEE draws a line between free and pay at about 15%.  They profess a portfolio of products to engage the remaining passengers.   The portfolio aggregates passenger usage as the pleasing the whole cabin is greater than the sum of each individual product or service.  The challenge is to grow the market, and not cannibalize sales as in a zero-sum game.
  • The business traveler needing a premium Wi-Fi experience
  • The leisure traveler just wanting to update social network
  • The bored traveler just wanting something free to take the time
  • The frequent traveler willing to pay to watch that missed show or blockbuster movie
GEE divides the price-elasticity curve into region A, B, and C.  Region A is the area where take rate can be expected, above which a different product mix is brought to bear, and borders 15%.  In Region B, about 60% of the remaining passengers may be enticed to purchase some other product.  The final 20% will not pay for any product, period.

Revenue in Region A is primarily from the user payment.  Region B and Revenue C revenue can be a progressive mix of low cost product sales, ultimately seeking revenue purely through promotion and advertising with a free or heavily discounted price. 

Seatback LiveTV (95% Free)

Seat-back displays are inherently more engaging than any passenger device, given that they literally are "in your face".  Making them free almost guarantees their use.  Bring-Your-Own-Device (BYOD) products can be used without connecting to the airplane networks, and in that context they probably can compete with 95% usage.  When scaling BYOD with connectivity, an upper limit of around 50 percent, or greater than any report so far, seems a lofty goal.
The usage rate on airlines that offer the service for free is over 95 percent.

Chris Moore (6% Pay & 100%+ Free)

EVP and CIO, Sun Country Airlines 
European consumers still expect to pay a premium for WiFi con is slowly changing although Europeans still gleefully expect to pay for using public restrooms, which is unheard of in America. Show me a pay toilet in the USA and I will show you scuff marks (from people climbing over the door).  
the risk suddenly rockets into the stratosphere considering that the average “take rate” for consumer-paid is about six percent, meaning that on an aircraft with 175 passengers only 10.5 of them will pay for the service. These are some sobering numbers ... 
 ....when in-flight WiFi is offered as a complimentary service excess of one hundred percent (this is possible because each passenger can have two or more PEDs they are connecting), and the bandwidth, especially with ground-to-air sys become saturated thus resulting in a poor user experience

1 comment:

  1. This topic has always been one of my favorite subjects to read about. I have found your post to be very rousing and full of good information. I will check your other articles shortly.
    air ticket booking