Q408 In-depth Analysis – eBay vs. Amazon – The Giants of Ecommerce Duke it out

February 4, 2009

ChannelAdvisor ChannelAdvisor By ChannelAdvisor

This is the first of a four part series focusing in on eBay and Amazon.  This series comes from many questions we are receiving at ChannelAdvisor from our customers and prospects (and the press/Wall st.)  about what is going on at the two companies and what retaliers strategically should do about it.

Here’s an outline of the series:

  • Episode I – Q408 in-depth analysis
  • Episode II – Introducing the ChannelAdvisor Ecommerce Framework (CEF)
  • Episode III – eBay, Amazon and the CEF
  • Episode IV – How to fix eBay
Many long-time readers have been giving me a hard time for being sparse with posting and this is my attempt to catch up and give some meaty strategies and data vs. looking at smaller news items and covered in other excellent venues.  For those readers that are not even involved in eBay or Amazon, I believe there are some VERY valuable lessons to be learned.  Heck even if you aren’t in ecommerce, this is a very interesting case study playing out that could very likely be simliar to the Google/Yahoo! ‘winner takes all’ kind of example we are seeing in the paid-search tangential market.  We’re going to start at Q4 results and then really drill into what’s going on at these two companies and why their growth rates have started to separate at such an accelerating pace.
I hope you find it useful and thanks for your patience.  The good news is the entire series is fleshed out and I should be able to get it out over the next week or two without interruption.
Now on with Episode I…
Q408 – I think this will be the most important Q in the history of Ecommerce
Given the economic backdrop and what’s going on at the ecommerce giants eBay and Amazon as well as the rest of the industry, I believe that for years we will look back at Q408 as a significant ‘ah-ha’ moment when the entire World finally woke up to what those of us entrenched in the industry have been seeing indications of for the last two years – Amazon is significantly, methodically, and relentlessly eating away at eBay’s market share. I believe Amazon is the largest beneficiary of eBay’s decline, but certainly not alone. Online retailers from Zappos to Newegg to Wal-mart and others are all doing their part in taking share from eBay.  In fact, many of eBay’s changes are actually self-inflicting further declines as I’ll attempt to illustrate later in the series.
Before we go into details on how material this is, I wanted to pull together some highlights from eBay and Amazon’s results.  These results are culled from many areas including:
  • Company announcements and conference calls (transcripts here: eBay , amazon)
  • Analyst notes – I’ve read and dissected information from every analysts and you will see specific references to datapoints from Jeetil Patel @ DB, Scott Devitt @Stifel, Imran Kahn@JPM, Justin Post @ML/BOA, Derek Brown, and others.
  • Industry analysts – I was just at shop.org and talked with some of the top analysts about these trends.
  • Where possible, we enhance this data with data from ChannelAdvisor and our own internal analysis/tracking.
eBay Q408 Highlights
By all but three measures eBay had a terrible quarter.  The focus of this blog is eBay’s marketplace business so I’ll only cover that area and not spend any time on Paypal or Skype (which are both doing well, but still too small to turn the tide of the core decline).  Thus all data and analysis in this section specifically refer to the marketplace business and omit any and all PayPal/Skype information.
The three bright spots for eBay’s quarter were:
  • Items sold were up 3% y/y
  • Active users increased 4% y/y (a small acceleration, well big percentage-wise, but small in absolute terms) from the Q3 3% y/y gain.
  • Fixed price items are now 49% of GMV
Here are some graphs from Jeetil Patel that show the trends in the item sold and active user metrics: (click to see full-size version):



The transaction growth chart shows that while eBay held onto a y/y growth on this metric, the trend is definitely decelerating.  It will be interesting to see if this goes negative in Q1 or if eBay can hold the line here.
In active users, you see a little bit of a bounce y/y growth-rate-wise vs. the lows of Q208.  Jeetil points out that this could largely have been driven by the expensive (contra revenue) coupon and mslive cashback activities in Q408.
These were the bright spots and the negative datapoints were:
  • Overall GMV was down 16% y/y
    • Autos GMV was down 30% y/y
    • International GMV was down 15% y/y
    • Domestic GMV was down 9% y/y
    • Fixed-price GMV was down 2% y/y
    • Auction GMV was down a whopping 26% y/y
    • Total non-vehicle GMV was down 12% 
  • ASPs were down 7%
  • GMV/user was down 8%
I believe the GMV datapoints are the most significant measures of the health of any marketplace and every eBay GMV trend is defnitely very concerning.  Before Amazon announced, it was easy to look at these through the lense of the economic backdrop, but after Amazon, a different picture emerged (more on that later)
Here are some interesting graphs that show these datapoints and the trends that led to them.  First, here’s a breakdown of GMV over the last 5 Q’s with the fixed-price and auction portions called out from JP Morgan.  The second graph shows the growth rates  of fixed-price and auction over the course of the last year:
I think the second chart is very important because it shows both the well known fact that eBay’s auction business is ailing, but it also shows that fixed-price isn’t too far behind.  It also begs the question if eBay’s movement from auction to fixed-price will really help at all.  In other words, are they going from the frying pan (down 26% y/y) into the fire (down 2% and dropping) here?

These two charts also make me wonder about eBay’s search strategy that we believe is really starving off auctions in favor of FP listings.  How much of the trend you see in the second chart is self-inflicted and how much is consumer-inflicted?  Only eBay has the data on that, but we’re seeing such terrible results with auctions from our customers (e.g. pair of shoes at $.99 with zero bids) that I’m quickly moving to the camp where many sellers are that the decline in auction GMV is largely self-inflicted due to the changes made in 08 to fees and finding (more in ep iv).

Amazon Q408 Highlights

As with eBay, for Amazon I’m going to focus on what internally Amazon calls Amazon’s Seller Business (ASB), or what many analysts call their third-party business (abbreviated 3P).

  • Overall revenues (mix of first party/retail GMV and seller business/3P revenues) were up 24%
  • Overall unit/transaction growth at Amazon was 28% y/y
  • Amazon’s Seller Business increased to 27% of all units
    •  International is really accelerating
  • ASB units were up 33% y/y (according to analyst estimates)
  • ASB GMV was up  36% y/y for Q4
  • Active customers grew 16%
  • Customer frequency increased 10%
  • Amazon is forecasting 15% y/y overall growth if you look at their guidance according to analysts.
The most interesting graphic I found for Amazon’s amazing Q4 was this one from Jeetil Patel at DB:
Here you see the unit growth rates of Amazon’s retail (first-party) and seller businesses (third-party).  This graphic illustrates the flexibility that having both models has given Amazon.  If times are tough or pricing competitive, they can rely on partners to compete on price with other retail venues and take their foot off the Amazon retail gas pedal.  In other times, they can change direction (e.g. 2Q07) and surge on the retail part of the business.  This allows them to ultimately manage their unit margin economics at a level unavailable to a pure-retailer and also a pure-marketplace.
Having both models gives Amazon uprecedented and unequaled in ecommerce network effects that I believe are fundamentially unwinding the network effects that built eBay over the last 10 years.  We’ll be spending much more time on drilling into this one concept (in epIII) as it is very very important to understand if you want to take advantage of it in your business.
eBay vs. Amazon
Now that we’ve seen each company’s individual results, let’s compare them side by side.
For brevity, here are the three key measures I think that show the most material differences:
First, y/y GMV growth which for me is the single most important measure and allows us to compare ‘share’ gains and losses:
  • eBay (non-motors): -12%
  • Amazon (seller business/3P): 36%
  • Difference: 48% (In other word’s Amazon’s apples-to-apples business is growing 48% faster than eBay)
Second, y/y active user growth rates:
  • eBay: 4%
  • Amazon: 10%
  • Difference: 6%
Third, y/y unit transaction growth rates:
  • eBay: 3%
  • Amazon (seller business): 33%
  • Difference: 30%
A near 48% growth rate difference is hard to visualize, so what I’ve done is hacked one of Jeetil’s charts to include some extra information:
There’s a lot going on here so let me explain.  First, draw your attention to the industry lines.  The green line is retail sales (offline) and the red line is overall ecommerce.  For anyone in this industry those are the benchmarks you should measure your business against.  If you are growing faster than ecommerce, you are taking share.  If you are growing slower, you are losing share.
Now the top line (blue) is Amazon’s growth rate (entire business – retail and seller business).  I’ve added two other data points to illustrate the 48% difference in Amazon’s seller business and eBay’s non-autos business.  The top square dot is up at the 38% growth rate (I didn’t have historicals to show the trend, but you can imagine it tracks about along the Amazon line, but higher).  The bottom round dot shows eBay’s -12% non-autos growth rate.  You can imagine the historicals by looking at the earlier eBay charts.
This chart clearly illustrates that you have Amazon gaining share significantly vs. eCommerce (21% difference: 18% vs -3%) as well as eBay losing share vs. ecommerce (9% difference: -12% vs. -3%) and then finally the 38% difference between Amazon’s marketplace business and eBay’s.
This startling 48% difference in growth rates got me thinking about what actually had never crossed my mind given eBay’s huge scale – when, if ever, will Amazon’s seller GMV equal or surpass eBay’s?
When will Amazon’s Seller Business surpass and eBay’s non-motors GMV ?
The trick to figuring this out is that while eBay reports their GMV ex-autos (or at least just started to), Amazon does not give this data point.  I reached out to several analysts and their 08 estimate for Amazon’s seller GMV ranges from $5-9b.  This variance is caused by some assumptions you have to make on mix (how much seller activity on media vs. the non-media categories) and average selling prices.  Based on our experience at CA, I settled in on a model that has 08 seller GMV at $7.5m – this jives with several analyst numbers as well.  I filled in the quarterly data based on this estimate and then did a forward quarterly swag, assuming that Amazon continues to grow at 36% and eBay continues to decline at 12%.
While they start Q109 with a large near 6X disparity with eBay at $11b and Amazon at $2b, by Q3 2012, Amazon’s business reaches $6b and is pretty much tied with eBay and exceeds it from that point forward.  There are many arguments you could make either way on this, but I thought it was an interesting exercise and actually happens much faster than I thought it would due to the 48% delta in growth rates.
One argument that supports this assumption is that Amazon and eBay don’t even really compete in the same geographies yet and categories.  As Amazon brings those online it could very well maintain a healthy growth rate difference and continue to chew away.
Here’s a chart of the model.  The blue line is eBay, the red trend-line is Amazon.
Everything to the left of the green line are actuals and everything between green and red is the extension of the 48% growth rate difference into the future. You can see it takes about 3 and a half years for the lines to cross.
You also can see that while Amazon has an enviable growth rate – don’t forget that eBay is at a significantly larger share so does have some time to turn things around and if you’re not on eBay today, you are still missing 20-25% of ecommerce.
Is eBay at a tipping point or caught in a death spiral?
Looking at the historical data (specifically the fourth graphic in this post) and what we’re seeing with sellers (and thus GMV) leaving the platform at an increasing pace (some voluntary, some involuntary), it’s a fact that the spread between Amazon and eBay is increasing and increasingly increasing (for you calculus buffs out there). This raises an interesting question – What if eBay is at a tipping point where they are losing enough GMV and sellers, (and thus buyers) that the model starts to unwind as quickly as it grew in the early days?  Conversely, what if Amazon continues to innovate and add buyers and selection, driving more and more value, helping eBay speed through a death spiral.
We could be looking at a perfect storm here and actually historical data supports this thesis with the fact that Amazon is accelerating while eBay is increasingly decelerating.  Over the last year the difference in their growth rates has increased 20%.
To illustrate, I took my model and experimented with a slight up-tick in Amazon’s growth rate each quarter and a down-tick in eBay’s growth rate – essentially growing the already healthy 48% delta to 68% over time. (this sounds crazy, but remember, the delta has gone from a Q108 delta of 17% (ebay: 14% Amazon: 31%) to 38% in under a year so a 20%+ swing per year isn’t uprecendented and I’ve even spread that out over two years – so you could argue this is conservative in a way based on historicals.  This more aggressive ‘tipping point’ model shows Amazon’s Seller Business GMV surpass eBay’s non-motors GMV by Q3 2011 – a relatively short two and a half years away, which is stunning given the 6X difference at the start.
The ‘tipping point’ model is show in this diagram:
The lines to the left of the green line are the actuals and the data between the green and red line show an acceleration of the growth rates from the current 48% to 68% by 2011.
Which brings us to Episode II….

Amazon’s ability to grow materially faster than ecommerce and eBay’s continued loss of share against ecommerce are very interesting case studies.  Even if you aren’t involved in any of these channels, I believe the case study of what’s going on here can shed light on your ecommerce initiatives and help you become a share gainer vs. a share loser.

In Episode II we’ll introduce a framework for thinking about some of the macro-components of a successful ecommerce site and then look at various Amazon and eBay strategic initiatives against that framework (epiii) and finally we’ll use the framework to see what eBay needs to do to try and turn things around (epiv), before it’s too late.
SeekingAlpha Disclosure: I am long Amazon and Google