Questions we’ve been asked, and how we answered them

A question we’ve answered · Uber, ahead of its IPO

“Hundreds of patent portfolios are for sale, our IP team is small, and the IPO is coming. Which patents are actually worth buying?”

Uber needed a defensive portfolio in a crowded market, fast. Thousands of patents were on offer, and a lean team couldn’t read them all while also finding assets that weren’t yet on the broader market.

1. 1Ran every offered portfolio through ClaimScore before anyone read a word
2. 2Grouped the patents into families and ranked them by claim breadth
3. 3Put the best claim from the highest-ranked patent in each family in front of an attorney

What we told them. Read the top of the list, not the whole pile. ClaimScore narrowed thousands of patents to the families worth an attorney’s time, and Uber built its portfolio while others were still sorting.

“Prior to manual review of any patents, we run each offered portfolio through analysis software…” Kurt Brasch, Head of Patent Transactions, Uber

A question we’ve answered · A patent owner considering assertion

“We think someone is infringing. Out of hundreds of claims, which ones should we lead with?”

Picking the wrong claim wastes years and money. The owner had a large portfolio, a suspected infringer and a sense of which patents were strongest, but no objective way to check it.

1. 1Scored every claim for breadth: fewer, less restrictive limitations score stronger
2. 2Benchmarked those scores against patents that have been litigated, and their outcomes
3. 3Reviewed the shortlist against the accused product and the business goal

What we told them. A short, ranked list of assertion candidates with infringement estimates, and an honest view of where their instinct was right and where it wasn’t.

A representative engagement, with the details anonymized

A question we’ve answered · A technology company entering a new market

“Who else is patenting in our space, and where is nobody looking yet?”

Keyword searches kept returning too much noise or too little. The team needed to see the whole field, organized the way their engineers think about the technology.

1. 1Turned a plain-English description of the technology into a classifier
2. 2Curated training examples with a practitioner in the loop, then scored the full corpus
3. 3Clustered the results into a custom taxonomy and mapped competitors across it

What we told them. A landscape built on their own taxonomy: who owns what, how fast each area is growing, and the open ground where nobody has filed yet.

A representative engagement, with the details anonymized

A question we’ve answered · An in-house IP team facing a budget review

“We’re paying to maintain hundreds of patents. Which ones can we stop paying for?”

Maintenance fees add up, but nobody wants to be the person who let go of the patent that mattered. The team needed a way to decide that they could defend.

1. 1Scored claim strength across the whole portfolio
2. 2Compared patents within each family to find overlapping coverage
3. 3Weighed business relevance with the client, then reviewed the borderline cases by hand

What we told them. A keep, review or let-go call on every patent, with the reasoning behind each, so the decision could be explained to leadership, not just made.

A representative engagement, with the details anonymized

A question we’ve answered · A software company with a pending claim

“Before we spend more on this application, will the claim survive an Alice challenge?”

Software claims live under the shadow of §101. Before spending more, the client wanted to know how a court would likely view the claim on an early eligibility motion.

1. 1Analyzed whether the claim is directed to an abstract idea (Alice step one)
2. 2Looked for an inventive concept beyond that idea (Alice step two)
3. 3Rated its survivability on a 1–5 scale, with the reasoning set out

What we told them. A clear rating, the reasoning behind it, and where the claim is most vulnerable, so they could decide whether to amend, continue or move on.

A representative engagement, with the details anonymized