Xerox’s challenge to all 16 claims of a ticketing patent rested, at its key step, on one expert paragraph that repeated the petition word for word. The Board gave it little weight and denied institution. On Director Review, Xerox offered the reasoning it hadn’t filed. It came too late, and the denial became precedential.
Before you read
The petition’s only support for a key claim limitation is an expert paragraph that restates the petition. What weight does the Board give it?
Little weight. Testimony that does not disclose its underlying facts or data is entitled to little or no weight (37 C.F.R. § 42.65(a)), and here it decided all 16 claims.
Part 1 of 7
Xerox and four co-petitioners challenged all 16 claims of a Bytemark ticketing patent. The case came down to one limitation: storing a data value that flags fraudulent activity in the user’s account record. To show the prior art did that, the petition cited one thing, paragraph 54 of its expert’s declaration.
The petition, as the decision summarizes it
Blocking a purchaser would require recording the blocking in a data record associated with the user’s account.
Expert declaration, paragraph 54
“[a] POSITA would understand that such a blocking would require recording the blocking in a data record associated with that user’s account.”
Supporting facts or technical reasoning
none cited
Source
The Board found that paragraph “merely repeats, verbatim, the conclusory assertion for which it is offered to support.” A POSITA is a person of ordinary skill in the art, the hypothetical skilled worker patent law measures obviousness against.
The PTAB’s evidence rule addresses this directly: “Expert testimony that does not disclose the underlying facts or data on which the opinion is based is entitled to little or no weight.” An expert declaration exists to supply what lawyers can’t: technical reasoning and facts from someone skilled in the field. Repeating the brief supplies neither.
Part 2 of 7
The main reference, Terrell, taught that a purchaser suspected of fraud “could be blocked from further use of the system.” The petition argued that blocking would require recording the block in the user’s account record. The Board saw a gap. Terrell blocks the purchaser, not the purchaser’s account, and it was “far from clear” that one requires the other.
Blocking a purchaser means recording the block in the account. Support: paragraph 54.
The account could simply be deleted. One alternative was enough to show the reference didn’t compel the petition’s conclusion.
So the petition needed evidence of how a skilled person would read Terrell. Its only evidence was the expert, and the Board reasoned in three steps. The expert “does not cite to any additional supporting evidence or provide any technical reasoning.” The testimony is therefore “conclusory and unsupported, adds little to the conclusory assertion for which it is offered to support, and is entitled to little weight.” A second assertion, that obviousness would follow, got the same treatment.
Part 3 of 7
The limitation sits in claim 1. Every other ground ran through it.
16 challenged claims · 16 depended on claim 1’s limitation · 0 independent of it
The Board wasn’t persuaded that claim 1, or dependent claims 2 to 6, 8 to 10 and 12 to 16, would have been obvious over Terrell and Ritter. The grounds for claims 7 and 11 added other references, Masahiro and Laudermilch, but depended on the same showing about claim 1’s limitation. With no claim reasonably likely to fall, the Board declined to institute review of any of the 16.
Part 4 of 7
On Director Review, Xerox tried new ways to describe the inference: common sense, a finite set of predictable options, the efficiency of flagging an account. The Director found those arguments untimely and still unsupported by the expert’s testimony, affirmed the denial, and designated the Board’s decision precedential on February 10, 2023.
The tempting reading
If the Board misses the point, we can explain the inference on rehearing.
What the Director said
The explanations were new, so they were untimely, and the expert’s testimony still didn’t support them. The record that counts is the one filed with the petition.
Source
Part 5 of 7
Not every missing feature defeats a petition. In Philips v. Google, the reference didn’t disclose every claimed feature, and Google relied on a skilled person’s general knowledge of pipelining: downloading upcoming content while earlier content plays. Expert evidence, corroborated by a publication, supported that knowledge and the reason to use it. The Federal Circuit upheld obviousness on that basis.
| Xerox v. Bytemark | Philips v. Google | |
|---|---|---|
| What was missing | Recording the block in the account | A pipelining feature |
| What filled it | One expert paragraph repeating the petition | Expert testimony on general knowledge, corroborated by a publication |
| Result | Little weight; institution denied on all 16 claims | Obviousness upheld |
The Federal Circuit separately found error in the Board’s addition of a combination the petition hadn’t raised. Supported general knowledge doesn’t let the Board invent a new ground.
Part 6 of 7
Choose your side.
Part 7 of 7
This is one designated decision, though it is precedential, and the rule targets unsupported repetition, not overlapping words. An expert may say the same thing as the brief if the expert also supplies the facts or reasoning behind it.
It holds
It doesn’t hold
The denial also turned on a genuine gap in the reference, which the expert didn’t fill. The decision doesn’t say how the Board would treat a repeated paragraph that had other support.
The takeaway
For every step the art doesn’t state, ask what the expert proves that the petition doesn’t already say.
Precedential decisions bind the Board within their designated scope. Informative decisions show the Office’s reasoning but do not bind the Board. Court opinions are cited for what they decided; later history was checked on the date shown above.