**Amazon.com** (NASDAQ: AMZN ) is adding tons of Prime members -- in December the company confirmed that it had "tens of millions of members," including adding more than 1 million new members in the third week of December. This raises the question: What are these new Prime members worth to the company? To answer that question, I sharpened my pencil and calculated the lifetime value of a Prime member.

**The customer lifetime value approach**Although it's not a traditional stock valuation technique, customer lifetime value is an interesting way to gauge the economics of subscription businesses. It can be calculated several different ways, but it comes down to how much extra profit a customer will contribute over time. To calculate it for Amazon Prime members, you need to know three things: How much will Prime members spend, at what contribution margin, and over how many years? Below, using a few assumptions, I've roughly answered those questions, and I've calculated an estimated lifetime value.

**1. How much will each Prime member spend per year?**According to a 2013 study by the Consumer Intelligence Research Partners, Amazon Prime members spend $1,340 annually -- more than twice as much as non-Prime shoppers, who spend $650 annually. Those estimates were for 2013, and I expect spending per customer to increase each year. Thus for 2014, I'm assuming that Prime members will spend $1,500 and non-Prime customers will spend $700.

**2. What is the contribution margin Amazon Prime members?**Contribution margin is a managerial account concept, so it may not be familiar to many investors. Essentially, it's a customer revenue less the variable costs necessary to earn that revenue. Amazon generated $75 billion in net revenue in 2013. Subtracting the variable costs associated with earning that revenue, i.e., cost of goods, shipping, fulfillment, estimated content costs and estimated Prime fees (to avoid double counting later), leaves $10 billion of contribution margin, which is 14% of revenue. That is the average across all Amazon customers. Since Prime members generate higher shipping costs, I assumed that the contribution margin for Prime members and non-Prime shoppers are 12% and 16%, respectively.

**3. How long will Prime members stay with Amazon?**The average expected tenure of members can be estimated based on churn rates. For Amazon Prime members, Piper Jaffray estimates annual churn will be 5%; CIRP estimates it will be 6%. Those estimates both look low for me, so I assumed slightly higher churn of 7%.

**Bringing it all together**Using the estimates described above, I calculated the lifetime value of Prime ($2,283) and non-Prime members ($916) using a 10% discount rate. Admittedly, these numbers are imprecise. Since Amazon doesn't disclose all the data necessary, I had to use numerous assumptions and estimates to fill gaps and simplify the model. Often, though, back-of-the-envelope calculations are the most useful for forming big-picture conclusions.

From this model, I concluded three things. First, new Prime members do contribute significant additional value to Amazon's business, and the more Prime members that Amazon adds, the better. Second, Prime members are significantly more valuable than non-Prime shoppers. Third, as Amazon's business mix shifts to include more Prime members, there is potential for operating leverage and margin expansion.

**The biggest thing to come out of Silicon Valley in years**If you thought the iPod, the iPhone, and the iPad were amazing, just wait until you see this. One hundred of Apple's top engineers are busy building one in a secret lab. And an ABI Research report predicts 485 million of them could be sold over the next decade. But you can invest in it right now... for just a fraction of the price of AAPL stock. Click here to get the full story in this eye-opening new report.

## Comments from our Foolish Readers

Help us keep this a respectfully Foolish area!This is a place for our readers to discuss, debate, and learn more about the Foolish investing topic you read about above. Help us keep it clean and safe. If you believe a comment is abusive or otherwise violates our Fool's Rules, please report it via theReport this Commenticon found on every comment.## Report this Comment On April 23, 2014, at 5:23 PM,

luger01wrote:I see one thing missing in your model and that is the simple fact that not all merchandise is "Prime eligible". Factoring this in, the benefit to Amazon Prime members is decreased by anywhere from 10% to 50%. Looking back at all of my past purchases from Amazon and given the fact that membership just increased by 25%/year, I have dropped my Prime membership because the cost is greater then the savings. I originally bought a membership when my wife got her Kindle. But that has since been replaced by an iPad and the value is diminished even more because of that.

## Report this Comment On April 23, 2014, at 6:01 PM,

slickandjakewrote:In a nutshell luger01, you are disputing the 7% churn rate. You dropped your membership because of factors, so you are part of that churn. Nothing else in the calculations would change. Even if not everything is prime eligible, it still does not change the numbers of spending for prime and non-prime members. Perhaps it may change the margins, but that was an estimate anyway. So perhaps you dispute the estimated margins separating the groups.

All in all this article illustrates how much more value a prime member is compared to a non-prime member. In the grand scheme of things I am not sure this really means much to me. That should be intuitive. Therefore, the real question this article needs to answer is how this information can be translated to how it impacts an investor and the decision on valuing AMZN stock.

## Report this Comment On April 23, 2014, at 6:38 PM,

MBAnoobwrote:Please help me understand your breakdown. I am just beginning an MBA, and this looks like a good combination of Time Value of Money and financial statements. Basically, I don't know why you considered churn, as you are figuring the per capita benefit of each Prime member. Your LTV resembles the present value of a perpetuity (PV = PMT/I), which makes sense to me, except that you are using churn rate as the interest rate. I think you would use the weighted average cost of capital for Amazon. Of course, you would have to assume that all the numbers boiling down to LTV remained constant. Thanks!

## Report this Comment On April 23, 2014, at 9:35 PM,

DukeMontrosewrote:AS previously disclosed, this fool had two negative experiences as an Amazon customer in the last 6 month..

"Isolated incidents"?

If not, I would question Amazon's very survival.

Right now, I just question AMZN's 500+ PER's survival.

## Report this Comment On April 23, 2014, at 9:40 PM,

DukeMontrosewrote:(I just posted a foolish comment +

it vanished like the Cheshire cat = try again)

In the last 6 months, as previously disclosed, this fool had 2 unhappy Amazon experiences as a CUSTOMER.

Hopefully "isolated incidents" - otherwise I'd foolishly fear for the company's very survival.

Now this fool only questions AMZN's insane 500+ PER, I mean, its survival.

## Report this Comment On April 23, 2014, at 9:44 PM,

RogerPPMwrote:I have one main problem with these calculations. The problem is the first assumption in 1. on the increased amount spent by Prime members. I don't dispute the difference, but there is an implicit assumption of cause and effect in using this in the calculations. i.e. it assumes that if a prime member were not prime that same person would spend the average amount as a non-prime member. This is not a good assumption. Instead you need to know the incremental spend due to having a prime membership by the same person. I believe many people join prime because they already spend a lot at Amazon so the cause-effect goes the other way, at least partially, although there is most likely some lift due to getting a prime membership. (i.e. a reinforcing cycle). A better way to estimate this would be to look at new members who just finished their first year of Prime membership and compare their spend in the first year to what they spent the previous year. Then, of course, you'd have to follow them for multiple years to get a lifetime value. I also realize this information is not available to the general public so you or I would not be able to do this calculation, but it's probably close to what Amazon does to get their ROI. In any case, this means your numbers are somewhat or very optimistic.

## Report this Comment On April 23, 2014, at 9:51 PM,

RogerPPMwrote:Also, these calculations are not taking into account that Prime members are getting some things for free (in addition to shipping) that non-Prime members do not get for free. Things like movies, TV series, some books, etc. In some cases, at least, the Prime member would have paid for these but instead received them for free. i.e. Amazon is losing at least some amount of revenue from these sources. It's hard to estimate this amount, but I believe it should be considered.

## Report this Comment On April 23, 2014, at 11:41 PM,

WWLADYwrote:I have been a prime member for at least 4 years. I spend considerable more than your calculation. I also plan to buy Amzon stock, but am waiting a bit to see where the price will bottom in the next month.

I also use Prime Streaming, though I also have Netflix, which I already own stock in. Prime has more older movies and TV series than Netflix. I also purchase whole TV series from AMZN, as well as newer release movies. My bigest concern is just how much it will hurt AMZN when they begin to charge state sales taxes. Lets face it, that is one of the big reasons we can save money, while still not having to get in our cars/suv's to go shop at a local store. This will affect the bottom line, can they add enough additional customers to offset the sales tax is anyone's guess.

## Report this Comment On April 24, 2014, at 12:34 AM,

rmarks1wrote:I was going to make the same comment that RogerPPM made and I agree. But one thing he omitted is that joiningthe "Prime Club" makes them more loyal and helps Amazon build a moat against other eCommerce sites.

## Report this Comment On April 25, 2014, at 1:22 PM,

TMFWillSommerswrote:@RogerPPM

Thanks for reading my article and providing comments.

Although I didn't explain it in the article, I did include the content costs associated with serving Prime members.

In my calculation of contribution margin, I assumed that Prime members cost Amazon $35 per year in content (i.e. free books, streaming content, etc). This is a rough estimate that was published in the Wall Street Journal a few years ago (see link below). Assuming an average of 15 million Prime members in 2013, I estimated $525 million in total costs, which reduced the contribution margin.

http://online.wsj.com/news/articles/SB1000142405297020350320...

I hope that helps.

Fool on,

Brendan

## Report this Comment On April 25, 2014, at 1:30 PM,

TMFWillSommerswrote:@MBANoob

Thanks for reading the article and asking a good question.

The inverse of churn is average customer tenure. So if members are churning off at around 7%, that implies average customer tenure is around 14.3 years. So to calculate lifetime value, I multiplied the annual contribution margin ($279) by the customer tenure. So total value is $3,986 (i.e. $279 x 14.3 yrs or $279 / 7% -- mathematically both yield the same result).

To account for the time value of money, those profits, which will accrue over 14 years, I discounted back to the present at 10%.

Of course, this does assume that contribution margin and churn are static, which is probably not realistic. That's one reason why I described my analysis as "rough" and "directional."

I hope that helps. Best of luck in your MBA studies!

Fool on,

Brendan

## Report this Comment On April 26, 2014, at 12:28 AM,

dpjordanwrote:This is fun, however, you cover your tracks every step of the way. Are you introducing a new methodology are did you have a triple shot express today versus a one shot latte? :-)

## Report this Comment On April 28, 2014, at 3:58 AM,

RogerPPMwrote:@ dpjordan - no need for disparaging comments here. If you don't understand or don't want to enter into the conversation you don't have to, but the rest of us want to understand customer LTV for Amazon's Prime program.

@TMFWillSommers

You answered my secondary comment, but my main point still stands. You need to be able to unravel whether (or how much) the Prime program lifts sales for an average Prime customer to do your calculations. Let me give you an example. Suppose I told you that an average frequent flyer member of American Airlines spent 10 times the amount that an average non-member spends on AA tickets. Does that mean their loyalty program lifts revenue an average of 10X for customers who are part of the program? Not at all. Even if customers spent no more when they join the program I would expect the average to be quite a bit larger for members than non-members. Why? Because people who expect to fly a lot are much more likely to join the program since they will benefit more by it.

Same for Amazon Prime. Customers who expect to benefit a lot from the program (probably at least $99 worth a year) are the most likely ones to join, so I believe the "lift" due to the program is a lot less than you assume above. You assume the lift is $800 annually, on average, per Prime customer. I repeated your calculations with some smaller lift numbers and everything else unchanged.

(Please check my work.) If we assume the lift is $300 (i.e. spend of $1000 per year) I get a Present Value LTV of $1792, which is still substantially greater then $912.

Even more extreme (probably too pessimistic), I assumed no lift, so spend of $700 per year. Then I got Present Value LTV of $1497. Again, still much better than $912. Obviously, the difference is the $99 membership fee for this last calculation.

In summary, I believe your calculations were too optimistic regarding the lift from the Prime program and my last calculations too pessimistic. We don't have the data to determine where in the middle the truth lies, but in any case, it seems it is still a good program for Amazon.

## Report this Comment On April 29, 2014, at 10:57 AM,

TMFfinosuswrote:Thanks for this analysis Brendan! Interesting and thought-provoking.

## Report this Comment On April 29, 2014, at 11:04 AM,

TMFNatowrote:I really liked this article, but it left me wondering: How much does Amazon spend per user on Prime costs, including shipping and its deals for content from folks like HBO? I realize that data's probably locked deep within Amazon's black box, but it'd be nice to know whether Prime's turning a profit for Amazon, or just bringing in more revenue to fuel its future growth.

## Report this Comment On April 29, 2014, at 11:31 AM,

Megawrote:20M prime members x $2283

+

100M regular customers x $916

=

$137B (the current market cap)

## Report this Comment On April 29, 2014, at 11:39 AM,

Megawrote:Now check that against your revenue assumptions.

20M prime members x $1340

+

100M regular customers x $650

=

$91.8B

But sales in the past year were only $74B. The model is a little off.

## Report this Comment On April 29, 2014, at 11:57 AM,

anindakumarswrote:luger01 - I'd be very interested to know if you used the streaming services that Prime provides, and if you used Netflix as well. I basically compute that given the amount of items I buy and don't pay shipping for, the streaming does not cost me more than $2/ month after the price increase, ie to say when you cancelled the membership, the streaming was virtually free. Of course it depends on what you buy and how much you buy.

## Report this Comment On April 30, 2014, at 1:36 AM,

8bravowrote:So funny! Fools fooling fools ;-)

Brendan - like every other Fool - is getting paid by the number of page views he generates with his articles. Throw out some numbers and let the fools do the work (i.e. bloat the comment stream). Works every time on every site like this.

## Report this Comment On October 31, 2014, at 4:48 AM,

mkhelwrote:It seems there's a flaw in the calculations -- simple back-of-the-envelope calculation (using margin multiple -- assuming infinite horizong -- which is upper limit for CLV) is CM x [ d / [1+d - retention}

here ;

CM = $279

d = 10%

retention = 93%

so this would mean that CLV of prime

is $279 x [0.93 / (1+0.1 - 0.93)] = $1526

likewise, for regular Amazon customers:

$112 x 93/17 = $612

## Add your comment.