Robocast Inc. v. Microsoft Corporation

Filing 449

MEMORANDUM ORDER Denying 331 MOTION to Exclude Expert Testimony of Robert Sherwood. Signed by Judge Richard G. Andrews on 1/24/2014. (nms)

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF DELAWARE

Robocast, Inc.,

Plaintiff,

v.

Civil Action No. I 0-I 055-RGA

Microsoft Corporation,

Defendant.
MEMORANDUM ORDER

Before the Court is Defendant's Motion to Exclude Expert Testimony of Robert
Sherwood (D.I. 33I) and related briefing (D.I. 332, 360, 393). The Court heard oral argument
regarding this motion on January 8, 20I4. Mr. Sherwood is Robocast's expert on the subject of
internet advertising rates. Microsoft contends that Mr. Sherwood's testimony should be excluded
because it fails to meet the fit and reliability requirements for expert testimony.
In his report, Mr. Sherwood concluded that publishers of internet advertising on average
charge a I 00% premium for Automated Browsing System ("ABS") advertising over static
advertising. (D.I. 333-I at I3). Mr. Sherwood bases this conclusion on his analysis of publicly
available rate cards, which disclose the "CPM" (cost per thousand views) that advertisers charge
for static and dynamic ads.
Microsoft contends that Mr. Sherwood's opinions do not fit the case because the
technology included in the "ABS" ads that Mr. Sherwood identified includes technology outside
the scope of the asserted claims. Essentially, because the dynamic ads for which Mr. Sherwood

1

analyzed rate data include technology different from the asserted claims, the research cannot
form a reliable basis to form an opinion of damages for the technology at issue.
Whether expert testimony meets the fit requirement "depends in part on the proffered
connection between the scientific research or test result to be presented and particular disputed
factual issues in the case." In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 743 (3d Cir. 1994)
(internal quotation marks and citations omitted). However, "the standard is not that high." Id at
745. For instance, the Third Circuit held that testimony that PCBs cause liver cancer reasonably
fit the case, even though none of the plaintiffs had liver cancer. !d. The court allowed the
testimony because the "increased risk of liver cancer was probative of increased risk of other
forms of cancer." !d. Here, Mr. Sherwood's analysis of "ABS" broader than the claimed
technology does not render his testimony inadmissible. Knowing the premiums charged for
dynamic advertisements will be helpful to the jury in determining the premiums that could have
been charged for the technology at issue.
Microsoft's other fit argument is that Mr. Sherwood did not consider Microsoft's use of
the asserted technology, nor the fact that Microsoft does not charge a premium over the static
advertisement price. The fact that Microsoft does not charge a premium is not controlling. Just
because Microsoft uses a different pricing model does not mean that Mr. Sherwood's analysis of
the market is irrelevant to the accused infringement. Furthermore, whether or not Mr. Sherwood
considered Microsoft's particular use of the technology does not negate his conclusion about
what other advertisers charge for similar "ABS" advertisements. The extent to which those
advertisements are actually similar can be addressed on cross-examination.
Microsoft's second argument is that Mr. Sherwood's methodology was flawed, and that
he did not disclose the calculations ofhow he came to a 100% average CPM premium. Microsoft
2

argues that Mr. Sherwood only relied on five rate cards, which is too small of a sample. 1
Microsoft also argues that what advertisers say they charge is different from what they actually
charge. The fact that publishers might offer volume discounts or the fact that the sample size was
very small does not negate the findings of the study. Of course, these factors may greatly
diminish the weight of the study, something which can again be addressed on cross-examination,
if not by competing evidence.
As for the transparency of his calculations, there are three sources of confusion. The first
arises from the following statement in Mr. Sherwood's report:
In my examination, Internet advertising publishers, representing more than 20
different websites, charge a CPM premium fee for ABS advertising. Publishers
charge a premium between 30% to 200% for ABS advertising. On average,
publishers typically charge a I 00% CPM premium for ABS advertising when
compared to static ads.
(D.I. 333-I at 13). What is unclear from the report is whether Mr. Sherwood based his average
on the "more than 20 different websites," or the six internet advertising publishers that he
referenced in his report. At deposition, it became clear that Mr. Sherwood based this average on
five ofthe six particular points that were mentioned. (D.I. 333-I at 67-70).
The second source of confusion is how exactly Mr. Sherwood calculated the average
premium when the rates were presented as ranges. At deposition, he testified that he averaged the
upper and lower end of the range. (D.I. 333-I at 65). Additionally, he went through his
calculations again, ending up at a premium of I 09%. (D.I. 363-6 at 2). While this number is
slightly larger than the I 00% Mr. Sherwood arrived at in his expert report, it is certainly within
the error range implicit in such a small sample size.

1

If a larger sample size would lead to a significantly different conclusion, one might expect Microsoft would
present such evidence, either now or at trial.

3

Lastly, there is some confusion as to why Mr. Sherwood chose the particular numbers
that he picked from the rate cards. For instance, Mr. Sherwood testified that he ignored rates for
"rich media," instead using rates for "premium rich media." (D.I. 333-1 at 64). This argument,
however, is really just another attack on the fit of Mr. Sherwood's study to the technology at
issue. While Mr. Sherwood's report was not as clear and precise as it could have been, the
underlying data was present, and the mathematical calculations were readily ascertainable.
Microsoft's expert has had ample opportunity to test Mr. Sherwood's conclusions. In my
opinion, Mr. Sherwood used reliable data and reliable methodology.
Defendant's motion (D.I. 331) is hereby DENIED.

1nrf1

Entered thisVtt day of January, 2014.

4