Sellify LLC v. Amazon.Com, Inc.

Filing 16

DECLARATION of Robert D. Kaplan in Support re: 11 MOTION for Summary Judgment Notice of Motion.. Document filed by Amazon.Com, Inc.. (Attachments: # 1 Exhibit A, # 2 Exhibit B, # 3 Exhibit C, # 4 Exhibit D (Redacted), # 5 Exhibit E)(Kaplan, Robert)

Sellify LLC v. Amazon.Com, Inc.

Doc. 16 Att. 3

EXHIBIT C
TO THE DECLARATION OF ROBERT D. KAPLAN

(HERRMANN DEPOSITION TRANSCRIPT EXCERPTS)

Dockets.Justia.com

ERIC HERRMANN 5/26/2010
Page 1

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK ______________________________________________________ SELLIFY LLC, Plaintiff, vs. AMAZON.COM INC., Defendant. ) No. 09CV 10268 (JSR) ) ) CONFIDENTIAL ) )

______________________________________________________ DEPOSITION UPON ORAL EXAMINATION OF ERIC HERRMANN ______________________________________________________ 3:47 p.m. to 4:51 p.m. May 26, 2010 925 Fourth Avenue, Suite 2900 Seattle, Washington

Barbara L. Nelson, CCR Court Reporter
CONTINENTAL REPORTING SERVICE, INC. [phone]

ERIC HERRMANN 5/26/2010
2 (Pages 2 to 5)
Page 2 Page 4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

APPEARANCES FOR THE PLAINTIFF: FRANCIS X. DEHN Attorney at Law Smith Dehn, LLP 381 Park Avenue South Suite 713 New York, New York 10016 FOR THE DEFENDANT: JEFFREY WANG Attorney at Law Friedman Kaplan Seiler & Adelman 1633 Broadway New York, New York 10019

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Whereupon, ERIC ALFRED HERRMANN, having been first duly affirmed, was examined and testified as follows: EXAMINATION BY MR. DEHN: Q Sir, could you please state your name for the record? A Eric Alfred Herrmann. Q And Mr. Herrmann, what is it that you do for a living? A I manage the sponsored links business and technical teams for Amazon. Q And would you describe for us what your duties are in that particular job? Actually, before you even do that, can I ask you, do you have an official job title? A Yes. So my job title is Senior Manager of Worldwide Paid Search. Q Okay. And would you go ahead and tell us what your job responsibilities are? A Essentially overseeing the business, working with finance to establish targets for the various Amazon businesses worldwide, interacting with the
Page 5

Page 3

1 2 3 4 5 6 7 8 9 ______________________________________________________ 1 0 INDEX OF EXHIBITS 11 ______________________________________________________ 1 2 EXHIBIT: DESCRIPTION: MARKED: 13 (No Exhibits Were Marked.) 14 15 16 17 18 19 20 21 22 23 24 25
______________________________________________________ INDEX OF EXAMINATION ______________________________________________________ WITNESS: PAGE: ERIC HERRMANN Examination by Mr. Dehn 4 Examination by Mr. Wang 39 Examination by Mr. Dehn 40

various search engine vendors worldwide, overseeing product roadmaps and new technical features, as well as the development team. Q Now, when you talk about paid search, I take it you mean, sir, paid search that is initiated by Amazon; correct? A Yes, so paid search is defined as -sponsored links is another term that's frequently used. Ads that run on external search engines and drive traffic to Amazon. Q Okay. Now, there was a time, sir, was there not, when members of the Amazon Associates Program could engage in paid search and purchase sponsored links and drive traffic to Amazon in that way; correct? A Yes. Q Okay. Did your job at that time have anything to do with that? A No. Q Okay. So your focus has strictly been on sponsored links and paid search originating from Amazon; correct? A Correct. Q Okay. How long have you been in your current job?

CONTINENTAL REPORTING SERVICE, INC. [phone]

ERIC HERRMANN 5/26/2010
3 (Pages 6 to 9)
Page 6 Page 8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A Almost three years. Q Okay. What did you do before that? A For the year prior to that, I worked in a different capacity on the team. I was the development manager of that team. Q Okay. What about before that? A Before that, I spent about five years at EMC and Legato, a company that was actually purchased by EMC, doing backup and recovery software. Q What is your education, sir? A I hold a degree in mathematics from the University of Washington, with a minor in computer science, and a master's degree in software engineering from Seattle U. Q Sir, I just alluded to the fact that there came a time, did there not, when Amazon changed its policy with respect to its associates and informed them that it would no longer pay for their paid search efforts; is that correct? A That is correct. Q Were you part of that decision? A I did a lot of the fundamental research that drove that decision, yes. Q Okay. What generally did that research show? A It showed that -Page 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

you had a 10 percent allowable efficiency and the keyword's running at five percent, that represents an arbitrage opportunity for -- potentially, for an affiliate. An affiliate could jump in, outbid us for that term, and in essence get paid at a ten percent rate for something we were already getting at five percent. So that's the fundamentals of the arbitrage opportunity. Q Okay. My understanding is that when the policy first changed in 2009, that it applied only to North America? A That's correct. Q Is that correct? A Mm-hmm. Q What was the reason for that? A We had two main reasons. One, we had I think more conclusive data in the U.S. at that time, and two, given the size of my team and sort of the complexity of the problem of enacting such a large change, we typically do that country by country, and frequently we'll start with the U.S. in a lot of our technical rollouts, given that we have the natural advantage that my team speaks English and it's a lot easier for us to handle problems and issues and monitor the ads in a country where we're comfortable
Page 9

MR. WANG: I'm sorry, before you answer, again, and I know we've done this -MR. DEHN: Confidential? MR. WANG: -- in the last two depositions, but we would ask, because of the nature of this testimony, that this deposition be marked confidential, as well. MR. DEHN: That's fine. THE WITNESS: I think the main findings were that it showed that in many cases these affiliates were actually arbitraging our existing campaigns and not really adding value to Amazon through that channel. Q Okay. Could you please explain to an outsider like me what you mean by arbitraging? A Certainly. So the Amazon system operates -well, we run our business based upon efficiency targets. So we're willing to spend X dollars to make Y dollars in sales. And our system is designed to optimize to those targets for different product categories. We offered affiliates essentially the same rates. However, in a keyword market, you can be in a situation where a given keyword might be running at a better efficiency than the allowable. So let's say

with the language, as opposed to let's say Japan. So typically, yeah, we'll test in the U.S., move to other locations. Q All right. Now, I understand that there were further rules changes in this year that changed the rules regarding free search, as well as paid search; is that correct? A Could you clarify -Q Yeah, I'll tell you my understanding. A -- on the free search? Q My understanding is that while Amazon instructed its associates sometime in 2009 that it would not pay referral fees based on purchases arising out of paid search, sponsored links, that at that time it continued to pay referral fees on purchases derived from free search. In other words, the free search results on Google and so forth, on search engines. A Mm-hmm. Q My understanding is that in March of 2010, that was changed. You don't pay for results coming from free search, either. Is that your understanding, as well? A That is my understanding. However, I'm not certain that was as a result of a policy change. It was my understanding that -- and again, it was my

CONTINENTAL REPORTING SERVICE, INC. [phone]

ERIC HERRMANN 5/26/2010
4 (Pages 10 to 13)
Page 10 Page 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

understanding that our policy was that that traffic was generally not allowed. I think we just stepped up our enforcement on that point. Q Do you play any role in creating or in modifying the operating agreement with the Amazon Associates? A No. Let me re -- I did have -- I did do some review work over the actual text of the policy changes that went into effect regarding redirectors, but I didn't make any material contribution to that. Q What's your annual budget for paid search at Amazon? A That varies from year to year. Which year are you -Q Let's say 2009. A 2009, I believe we spent close to $250 million. Q Did that exceed spending for paid search in 2008? A Yes. Q By how much? A I would estimate 30 to 40 percent. Q Did that increase have anything to do with the policy change? A The policy change did affect that increase.
Page 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

was accounted for, I take it, by other factors; correct? A Yes, growth being an obvious one. Q Growth in -A In the business. As traffic on the web grows, you know, that proportionately impacts the (inaudible.) THE REPORTER: I'm sorry, impacts the what? THE WITNESS: The sponsored links channel. MR. WANG: I warned the court reporter that you talked faster than the other two witnesses. That's okay. So do the two of us. Q And to be precise about that, how does the growth in traffic impact the sponsored links channel? A Well, it's typically in proportion. So as Amazon's business grows, as Internet usage grows, you know, that -- you know, those are certainly large factors governing the growth of the sponsored links channel. Q Well, I'm looking to see if I can break it down to something even simpler than that. I mean, is it because you're buying more keywords or because the keywords get more expensive or -A I see. Q -- or other factors?
Page 13

Instead of having, you know, the redirectors pay the search engine directly and, you know, have us essentially pay them, the situation changed so that we were paying the search engine directly. So yes, that did impact the -Q Is it fair to say that one of the reasons for the policy change was that you thought that your keyword cost would go down? MR. WANG: Objection to the form. You can answer. THE WITNESS: Yeah, we believed that we would see, by removing the arbitrage opportunity, we believed that, yes, we would see benefits and efficiency. Q Okay. Did that occur? A Yes. Q How can you tell? A Simple time series analysis. Q Maybe to you it's simple. I'm sorry. Go ahead. A Yeah, it's basically taking a snapshot of particular search queries and keywords before and after the event and observing the efficiencies there. Q So that when your spending on paid search went up between 2008 and 2009, the rise in spending

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

MR. WANG: Objection to the form. You can answer. THE WITNESS: So the factors -- the factors impacting us, we tend to see more traffic on the same keywords. As we receive traffic to the Web site, our automated systems generate more keywords, which in turn leads to more traffic. So all of those factors certainly help build growth in the channel. Q I see. Has the price of purchasing those keywords gone up? A Over which period? Year-to-year? Q Well, between -- yeah, between 2008 and 2009, because that's what we're going to be discussing. A Between 2008 and 2009, there was certainly some keyword inflation, although the general cost per click actually suffered a little bit as a result of the economic downturn around this time last year. Q Okay. And what are the indications for 2010? Is your budget increasing? A The budget is certainly increasing. It looks -- so yes, we're growing there. Q How does -- do you have a sense of what the percent increase is? A Worldwide, probably around 30 percent.

CONTINENTAL REPORTING SERVICE, INC. [phone]

ERIC HERRMANN 5/26/2010
5 (Pages 14 to 17)
Page 14 Page 16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Q How does your -- how does the budget for paid search compare to Amazon's budget for conventional advertising, if you know? And by conventional, I would mean both print and digital and other media, but not in the paid search realm. MR. WANG: Objection to the form. THE WITNESS: Yeah, I don't have a good answer for you there. I don't know quite how that budget stacks up against some of the other channels. Q Okay. You wouldn't know what the proportion of your -- of paid search -- I'm sorry, let me say it again. You don't have a sense of what the proportion of the paid search advertising budget is to the rest of Amazon's advertising budget? A I have a sense of the proportion, but I don't think I'd be able to give a very accurate answer. Q Do you have a best estimate or a range? MR. WANG: Objection to the form. THE WITNESS: Thirty percent, something in that range. Q Do you know whether referral fees to associates is considered part of Amazon's advertising budget?
Page 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

other search engines, so paid search. Q Okay. Does Amazon still, to this day, pay referral fees to redirectors? A We do not. Q Okay. And when did that cease? A It's -- well, in the U.S. and North America, it ceased in May 1st of last year. We had a staged shutdown. I believe the UK was February 1st, and March 1st was essentially the rest of the world, so Germany, France, Japan, et cetera. Q Okay. So at this point, Amazon's policy would be not to pay any referral fees to redirectors at all; correct? A Correct. Q How can -- strike that. Following the May 1st, 2009 change in policy, and at that time you stopped paying referral fees to North American redirectors; correct? A Correct. Q I'm glad I understand what that word means, because it's so much easier to say than what I was trying to say previously. What did Amazon do, if anything, to identify whether it was still receiving traffic from redirectors?
Page 17

A Yes, it is. Q Do you have any knowledge as to what the approximate percentage of your advertising budget associate payments comprises? A I think I'd be even less precise there, to be honest. Q Would it be more or less than what you pay in paid search? A It would be more than paid search. Q It would be more? A Yes. Q If you -- I'm sorry. A I think I should offer a clarification. Q Yeah. A Associates spend is derived from a lot of different buckets. So that could include content sites, deal and coupon shops, comp shops, as well as sort of core associates, bloggers and the like, so -as well as redirectors, so they're -- you know, taken as a whole, it's a very large program. Q You used the word redirectors a couple times. Could you provide a working definition of what you understand redirectors to be? A Certainly. A redirector is an associate who drives traffic through placing ads on Google and the

A With regard to the May 1st shutdown? Q Correct. A So we had a couple of things in place. So the payment system, the associates tracking and payment system was modified to allow us to whitelist and blacklist certain affiliate tags. And so that approved -- gave us a way of filtering and spotting sessions that had come in through the redirector channel. We also did a fair amount of manual observation. Unfortunately, we have no automated way of detecting or scraping the actual ad text out on Google and Yahoo and the other search engine vendors' Web sites. It goes against our terms and conditions with those vendors. So as a result, the only way we were able to sort of check to make sure that we were getting compliance, beyond some filtering on the backside, to make sure that anybody that didn't comply wasn't paid, was to break things up and look at things manually. So for about a month or so, my team -everybody on the team took, you know, a hundred keywords and would literally manually go and check them on, you know, a quasi-daily basis to determine, you know, whether or not, you know, whether or not our

CONTINENTAL REPORTING SERVICE, INC. [phone]

ERIC HERRMANN 5/26/2010
9 (Pages 30 to 33)
Page 30 Page 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

who was redirected. Q But, and I guess -- and I guess the -- as we sit here today, though, I guess you don't really need to make that distinction anymore, because whether it comes from free search or paid search, it's all forbidden; correct? MR. WANG: Objection to the form. You can answer. THE WITNESS: Actually, we still have to make that determination, because if it comes from -or it's still difficult to make that determination, because if it comes from free search or paid search and someone has their browser settings set to a position where the referring URL information is null, that's still going to get through and we were going to wind up paying. So it's still very important for us to continue to monitor that. Q You had testified earlier about how, when the rules changed in May, effective May 1 of 2009, to -- so that redirectors would no longer be paid, you and your team did certain work to try to identify redirectors; correct? A Yes. Q Okay. Did you focus on a particular group in trying to find, in trying to target redirectors?
Page 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

A No, it was strictly based on the queries that we witnessed driving the most traffic. MR. DEHN: Can we take a two-minute break? I think I'm almost done. MR. WANG: Sure. (Recess taken from 4:32 to 4:35 p.m.) Q I'll show you, sir, a document that was marked as Tarpey Number 5 in Tarpey's deposition. A Mm-hmm. Q Can you identify that document, sir? A Yes, this was an e-mail I received from my Google account rep after I had notified her about this ad. Q Okay. And how did the -- well, when you say this ad, what ad are you referring to, sir, so we have it for the record? A The Beware the SCAM Artists ad. Q Okay. How did you become aware of that ad in the first place, sir? MR. WANG: You should, in answering this, to the extent there were discussions with Counsel, just don't discuss the content of those discussions. THE WITNESS: Mm-hmm. Yes. So I received an e-mail notification from Kathy Sheehan, another Amazon employee.
Page 33

A Yes. We -- again, we were able to only look at a very limited number of queries, due to the fact that we had to do it manually. And so as a result, we took the queries upon which -- essentially an ordered list of the queries for which the redirectors had gotten the most traffic, as well as some of our top terms, to ensure that redirectors weren't stepping in and arbitraging those terms. I think we did approximately 1,000 or 2,000 queries. Q With respect to those redirectors you looked at who were seeming to drive a lot of traffic, was there a threshold or a cutoff that you used to determine whether you were going to take a look at them or not? A We were less concerned about how much traffic the redirector drove and more concerned about covering the top queries. You know, of course given the hundreds and, you know, hundreds of thousands and millions of queries that come in, you know, on a very consistent, hourly basis, we were -- you know, we were just scratching the surface, but we wanted to get the queries that saw the most traffic. Q So in determining which ones to look at, did you use as a criterion those redirectors who were seeming to drive the most traffic?

Q And without asking you the content of what Kathy may have discussed with you -- Kathy is, in fact, an attorney who works for Amazon; correct? A Yes. Q And having received a communication from Ms. Sheehan, what did you do, if anything? A I believe the first thing I did was to determine whether or not this was indeed an Amazon ad. If our -- you know, it was -- I believe she'd asked me if it was. It clearly wasn't. It's not our ad text. We wouldn't run an ad like this. And you know, I decided at that point, or at that point I believe I -- I believe Anne Tarpey was brought into the conversation. I'm not a hundred percent sure. And you know, that the -- and I was -and I essentially made the call to Google, to the Google account rep, to see if they could do something about bringing down this ad, since we had, you know, been unable to do so at that point. Q What was Google's response? A Essentially what you see in the e-mail. The account manager forwarded this to the policy team. I don't have any direct access to that team. And that was -- that was, in essence, it for my participation. Q Do you know one way or the other whether

CONTINENTAL REPORTING SERVICE, INC. [phone]