How to Do a Better Trademark Search: Getting it Right the First Time


Artificial intelligence (AI) has become standard in healthcare, finance, and countless other industries. Polydimensional algorithms are used to assess a customer's qualification for a credit card, mortgage, or optimal dosages for an antibiotic prescription. However, the legal technology field has lagged significantly in its adoption of AI for risk assessment and decision making.

The result is trademark search technology that may seem to be stuck in the 1990s. In many cases, this perception is reality. Trademark attorneys are familiar with the challenge of having to repeat searches many times in order to perform a single risk analysis. In this blog, you'll learn about the challenges with trademark search and how AI could be the answer to slow, costly, and ineffective search engines.

Challenges with Trademark Search: When Wildcard Search isn't Sufficient

Most attorneys are reliant on legacy trademark search technologies which only permit "wildcard" searches. The wildcard search concept is based on character matching patterns that connect search queries to documents surveyed by the database. While this offers some slight advantages over the even more bare bones concept of exact string matches, it's still not adequate for the trademark landscape.

Wildcard search-based technology won't inform you if your candidate has offensive or negative meaning when translated into another language—which companies like Ford and General Motors experienced after products were launched. It also won't reveal very important risk factors like visual or phonetic similarities of search terms or the aggressiveness of existing trademark holders.

An effective trademark search requires lawyers or their technology to think multidimensionally and explore multiple avenues of risk. Matching characters are one form of risk, but it doesn't reveal the full picture. This means that with traditional resources, searches may need to be repeated several times in order to explore all possibilities sufficiently—or else run the risk of missing out on the full picture.

Why Trademark Search Technology is Often the Problem

Classic search engine technology is engineered to survey documents, which isn't ideal for the crowded brand space of today. Attorneys need access to a polydimensional search in order to quickly and accurately assess risks without having to reach back into their resources repeatedly.

The right kind of technology is equipped with the legal data sets and algorithms and has the built-in artificial intelligence to think with the same broad, contextually-aware approach that an experienced attorney does.

Risk is More Complex than Exact Matches

Risk isn't as simple as a 100% match. When reviewing results, attorneys often take a more conservative approach to considering how existing registrations and trademarks are similar to their candidate. This may involve a mental set-point of a 50% match as an acceptable risk threshold. However, risk acceptance may vary based on a match's history of litigation or other factors uncovered in search.

Trademark law can be highly subjective and difficult to capture in any AI system, which is why a polydimensional approach to risk analysis is crucial. Trademark search technology should take a similar approach to determining the risks associated with results that appear for a variety of reasons that are more complex than character string matches.

How to Get a Trademark Search Right the First Time

Trademark attorneys need search technology that's comprehensive and fast. These factors do not need to be mutually exclusive. The recipe for "getting it right the first time" is a search technology that employs smarter artificial intelligence and richer resources. This means a search tool that's polydimensional, or has the built-in capabilities to integrate hundreds of decision criteria across a wide array of sources in many languages.

Exact match for character strings is the first step, but it's nowhere near the level of assessment necessary.

Your search tool should also consider:

  • Similar matches, with a conservative risk threshold.
  • Aggressiveness of trademark holders.
  • Phonetic, visual, and root word similarity.
  • Word meaning in a wide array of global languages.
  • Unregistered trademarks being used in apps marketplaces and on eCommerce websites.

The right approach to search doesn't require that attorneys repeat their processes over and over again, or switch tools countless times. A tool that can analyze the relevancy of information on multiple levels can provide access to ranked results in seconds.

Faster Trademark Clearance with Confidence

Balancing speed with quality during trademark candidate assessments is among the most pervasive challenges for contemporary trademark professionals. The solution, in many cases, lies in technology that thinks as multidimensionally as you do.

To learn how trademark counsel teams for IBM and Trust Tree have shaved days off their trademark search processes while lowering risk, we recommend the free webcast Tips from the Field: Enhancing Speed and Quality in Trademark Search.

New Call-to-action

By TrademarkNow
The most advanced #trademark protection technology in the world – Trademark search, risk analysis and watch, powered by artificial intelligence