Patent Rating Systems

There is a saying:  “An intelligent person learns from his own experiences; a wise person, from the experience of others.”  As a patent researcher, you gain lots of experience researching your field of patents and learn to apply it intelligently.  Sadly (and expensively), your organization is often none the wiser.

The insight and information you glean from project to project is used for its purpose then often forgotten, filed on a chaotic hard drive, or simply tossed out.

With the launch of our custom fields feature, you can use to capture information to benefit you and your organization now and well into the future.

By no means do I claim to have all the answers.  But I think I have a pretty good understanding of what most organizations require to effectively rate their patents, establish efficient workflows, and retain data for future use.

Patent Ratings vs. Value Indicators

Don’t be confused between a Patent Rating and a Patent Value Indicator.  Patent value indicators are exceedingly helpful when doing patent research, but they tell you absolutely nothing about the value or quality of a specific patent.

patent-value-indicators-vs-patent-ratings

Value indicators include things like the number of forward citations, or a patent’s family size.  They are predictive of valuable patents, but will not give you actionable data.  Value indicators have their limits because for every rule, there are many exceptions.

A patent rating, on the other hand, is derived by human intelligence, and tells you everything you need to know about a patent’s quality, its usefulness to your organization, and its monetization potential.  A patent rating uses the insight and experience of an expert in the field to score the patent along specific, well-defined dimensions.  You can trust a patent rating (for the most part); a value indicator only points you to potentially valuable/important/strategic patents.

Strategic vs. Tactical Patent Rating Systems

I believe in keeping rating systems as simple as possible as long as the information you gather is useful.  Once your organization has mastered a relatively simple system, it can evolve into more complex rating methodologies.  I like to think of patent rating systems in two ways:  Strategic and Tactical.

A strategic rating system will necessarily be more complex because you have to consider all the use cases for all the potential stakeholders of the patent rating information.

However rating systems are often ad hoc and tactical where you have a specific goal in mind, such as finding patents that read on your competitor’s products.  In this case, you might only need two fields, “Reads On Competitor (yes, no, maybe)” and “Detectability (easy, medium, difficult)” along with a comment field to annotate which products the patent may read on.

In the strategic rating system, you have to consider all your stakeholders, and develop scenarios of how they might use the information once it is captured in your system.  In the tactical rating system, the data you collect from your subject matter experts will likely to constrained because of limited time and budget.

Designing a Patent Rating System

The patent rating system you develop will be designed for your stakeholders’ need for information.

patent-rating-system-stakeholders

  • A patent annuities manager may only need to know one simple question.  Should we renew or abandon the patent?
  • A patent licensing manager may need a ranked list of your best patents further annotated by features they enable and devices they affect.
  • A patent litigation attorney needs a sorted list of patents that read on a competitor’s products where claim charts might be developed with some insight into the cost involved in developing claim charts.
  • R&D engineers need to understand the major landmines and competitive landscape when designing new products.
  • Executive management needs high-level consolidated information about the relative strength of your portfolio compared to competitors.

When you design your patent rating system, think about specific use cases for each of the major stakeholders.

This first article is about the characteristics of ratings systems.  I have seen lots of them, and you would be surprised how they get cobbled together over the years in big organizations, and how confusing they become.  Rather than, getting into specific elements of a rating system, (we’ll save that for the next article), let’s talk about characteristics of good, easy to manage, and useful patent rating systems.

Characteristics of Well Designed Patent Rating Systems

There are many ratings systems out there, but the well-designed ones have the following in common:

detectability-ratings2

Well designed patent rating systems use a 1 to 5 scale and have clearly defined parameters for both the rater and the consumer of the patent ratings data.

  1. They are numeric.
Never use A-F to rate patents.  There is no way to sum up letters, or weight them using a formula.  Even your letter grades in school are transposed to numbers before providing summary and statistical data.  If you use letter grades you’ll add an unnecessary step when trying to manipulate your ratings later.
  2. They use a 1 to 5 scale.
Scales that are too granular, like from 1 to 100, can cause paralysis when trying to rate a patent, and nobody will clearly define what it means to be a 67 vs. a 68 on a rating scale.  1 to 5 scales seem to work best and is by far the most accepted patent rating scale.
  3. Bigger numbers are always better.
5s should be indicative of the favorable condition in your rating system.  This is cultural.  We naturally think of more as being better.  You wouldn’t say that a restaurant was terrific and give it 1 on a scale of 1 to 10.  Don’t confuse the ultimate consumer by making them transpose every score in their head.
  4. They are multidimensional.
Patents are complex properties.  Patent value and usefulness has many faces. A patent’s overall rating comes from it possessing a sum of specific important characteristics.  In the next article, I discuss the dimensions you will want to consider when developing your patent rating system.
  5. Simpler is better.
I’ve seen rating systems that are complex and powerful but are also expensive, time consuming and as a result, not implemented well (or at all).  The best ratings systems tend to be relatively simple, straightforward and actually put into practice.  You have to balance your need for information with the practical reality of rating patents.  Define what is truly important, and rate your patents by the fewest useful dimensions as possible.
  6. Rating scale is defined.
You’re not evaluating a pizza. When you give a patent a rating, make sure that the parameters are well defined for both the rater and the consumer of the ratings.  One person’s 3 should be another person’s 3.  Two reviewers may not agree on the rating, but they should always agree on the definitions of the rating scale.
  7. Defend your ratings with comments. 
Your consumers of patent ratings data should trust your rating system.  Don’t tell me that you think a patent is a 5 without telling me why.  Have your experts justify their rating by giving short comments along with each numeric rating.  Even if your experts are wrong, at least you know their way of thinking when they rated the patents.
  8. Don’t paralyze your experts.
Some ratings systems require so much and time that your experts can’t economically rate patents.  It is a balancing act, but if you require perfection, you won’t be able to rate your patents efficiently.  Your experts should feel comfortable providing their best guess and give their justification for their thoughts.  The ratings can always be refined later as new information comes to light.
  9. Make your system dynamic.
As the market changes, so does your picture of a patent’s quality.  You may have had some amazing CRT patents back in the day, but as the market moved to flat screens, their use, and their quality plummeted.  Re-rate your patents from time to time to see if the ratings you applied in the past apply today.   In a perfect world, US patents will be reviewed and re-rated before each of their renewal periods.
  10. Capture your ratings for future use.
It is time consuming and requires an investment to rate and review your patents (or your competitor’s patents). Store your data so all the stakeholders across your organization can access the data and benefit from the investment you made.  An amazing rating system, implemented to perfection, which nobody has access to does the organization no good.

Summary

A patent requires major investment. By most estimates, the average patent costs on the order of $25,000 to prosecute, $13,000 to maintain and $10,000 to manage.  If you look at an overall R&D budget and divide by the number of patents, companies like Microsoft spend about 4.5 million dollars of R&D for every patent they are awarded.  Applying only a small fraction of the R&D expenditures to the patent group, it is easy to justify a 100K per patent lifetime expense, but to what end?

Patents are supposed to be useful and help your organization by protecting you from copycat competitors (who didn’t make the big R&D investment), by giving leverage for cross licensing with legitimate competitors, or by making money from outbound patent licensing and sales.

These goals are well established and well understood, but your organization is hobbled if you don’t have a deep understanding of the strengths and weaknesses of every patent in your portfolio, and a quick way to recall the relevant patents for every possible offensive, defensive or monetization scenario.  Companies that have a profound knowledge of their patent portfolios have a strategic advantage over their competitors who waste time and money every time a new problem arises.

Clearly there is an investment required in developing and implementing a patent rating system.  But you don’t have to eat the elephant in one bite.  Get started by capturing ratings data on tactical projects; develop internal workflows and rules before starting an enterprise-wide patent ratings project.

In the next article, I’ll discuss specific dimensions which patents are often reviewed and rated and give you a list from which you can pick and choose to meet your company’s long-term strategic, and short-term tactical goals.

About AcclaimIP’s Custom Fields

AcclaimIP’s custom fields feature is a breakthrough in IP knowledge management.   Patents can be tagged with rating systems, extensive comments, internal taxonomies, and any data point that helps you better manage your portfolio.

Unlike other systems, your custom data is fully integrated and immediately available through the same indexing technology that powers the rest of AcclaimIP.  Apply custom data to tens of thousands of patents and search, sort, filter and export with no noticeable reduction in performance—something unique to AcclaimIP.

Your custom data is protected by a robust access and security model and managed in a US based data center by US based personnel.  Custom Fields is part of our “Enterprise Toolbox” and requires an additional subscription.  Call us for a demonstration of these exciting capabilities.

A view of custom data showing exposed fields in the search results grid.  The facets or filters to the left work like other filters in AcclaimIP.  (All data was imported using a randomize function and does not represent expert ratings of the patents shown)

A view of custom data showing exposed fields in the search results grid. The facets or filters to the left work like other filters in AcclaimIP. (All data was imported using a randomize function and does not represent expert ratings of the patents shown)

custom-fields-document-details

In the Document Details view, users can set their “East” panel to show the custom fields that are currently of interest.  Permissions can be set so specific users and groups of users can view (or not), edit and export custom data.

TTFN,

Matt Troyer

Post navigation

LATEST ARTICLES