“It is often said that you shouldn’t buy a car made on a Friday. Well, what should we say about a patent based on an application filed in December!?” – R. Millien
Many a Chief IP Counsel (“CIPC”) annually determine, at the end of the fourth quarter or the beginning of the first quarter, patent application filing targets for their enterprise. This exercise is most usually dictated by a C-suite executive (e.g., GC, CFO, CTO, COO, or CEO) in order to allocate an IP budget. After all, the commonly-held belief in the corporate world is that if you do not set a budget, every dollar spent is over budget. Right!? Thus, we ask: “Should we be setting annual (‘hard’ or ‘soft’) filing targets in the first place and, if so, how do we go about setting them?”
The answer to the above questions no doubt varies among enterprises of varying number of employees, revenue, industry, and corporate culture. Those CIPC who are forced, or otherwise feel compelled, to set a patent application filing target must inform their C-suite of an exact number. The calculus of that number is essentially determining how many patent applications should their enterprise be filing of the 2.9 million patent applications filed globally (including the 1.1M filed in China and the 589K filed in the United States last year). Well, there is no doubt that such calculus may not only be more art than science, but may very well be placing the cart before the horse (i.e., predicating an enterprise’s level of innovation on what may be an arbitrary, legal department created fiction).
Hypothetical Companies A & B
In order to discuss the wisdom of setting annual patent application filing targets, we look at two hypothetical U.S.-based enterprises – Company A and Company B. Let us assume both companies are large entities in similar industries and organized, for example, into three divisions (or strategic business units) and nine total profit-and-loss (“P&L”) centers (i.e., individually-managed product lines) as shown in Figure1. Each company has a CIPC, a managing IP counsel responsible for each division, and possibly an IP counsel at each of the individual P&L levels. Each company also uses outside law firms to perform patent application preparation and prosecution on a flat-fee (i.e., “rate card”) basis. Further, both companies have a set list of (for example, three to seven) jurisdictions in which they typically consider the international filing of their priority cases.
We may even further assume that Company A and Company B each have the reported average number of 13.6 IP personnel (i.e., lawyers, agents, paralegals, administrative assistants, etc.) globally and file U.S. and Patent Cooperation Treaty (PCT) priority applications. Now, the two enterprises diverge as Company A sets annual “hard” filing targets and Company B does not.
The Setting of “Hard” Filing Targets by Company A
“People with targets and jobs dependent upon meeting them will probably meet the targets – even if they have to destroy the enterprise to do it.” – W. Edwards Deming
How does Company A set their annual patent application filing target? Maybe they look at the number of priority patent applications filed in the previous year – let’s call that number X – and then perform the following calculus:
|Has it been a “good” year financially for the company?||X + 5%|
|Has it been a “great” year financially for the company?||X + 10%|
|Has it been a “bad” year financially for the company?||X – 5%|
|Has it been a “really bad” year financially for the company?||X – 10%|
Having arrived at a filing target – let’s call that number Y – Company A can then use their outside counsels’ rate cards (along with the relevant jurisdictional patent offices’ most current fee schedules) to come up with a filing budget. This exercise may be done at the HQ, divisional, and/or P&L level. In any event, what invariably happens is that Y haunts the CIPC and their entire department during the following year. Let us examine why.
In almost all enterprises, whatever flavor of performance management that is implemented, it consists of setting goals and then measuring employee performance against these goals. That is, it is standard human resource practice to encourage managers to set specific, measurable goals against which employees are measured. Consequently, people steer their performance towards the metric by which their job performance is measured and judged, and which ultimately determines their compensation and prospects for promotion.
Now, returning to our Y annual patent application filing target, and keeping in mind the basis of performance management, the haunting of Company A’s IP department personnel goes as follows at various levels:
- If I don’t meet Y, will I get my full cash and stock bonuses?
- What if I don’t file Y cases this year, will I lose my budget next year!?
- What happens if by October or November I am well short of Y, do I file (or convince my Patent Committee(s) to file) “junk” invention disclosures in December to meet Y!
- What happens if I filed so many cases by October or November that I will pass Y, so I need to “table” some really good invention disclosures until next year!?
From the above, one can easily see how the proverbial cart (i.e., filing targets) gets placed before the proverbial horse (i.e., an enterprise’s natural level of (important) innovation). So, let us now turn to Company B.
The Lack of Setting Hard Filing Targets by Company B
In contrast to Company A, Company B does not set a “hard” annual patent filing target. The CIPC at such an enterprise may push back on setting such a false number based on the belief that she cannot predict the level of her enterprise’s innovation for the next 12 months. In other words, such a CIPC reasons that, “how can I predict how innovative (and prolific) my company’s engineers and/or scientist will be next year?” Thus, she may attempt to convince her C-suite that no target is necessary or possible, or alternatively, inform them of only a “soft” filing target. In the latter case, how could such a soft target (or any “hard” target for that matter) be derived in light of the above-described Y calculus that results in an arbitrary number? The answer is the use of some form of metrics that are based on objective (and mostly publically-available) data.
Metrics for Informing Enterprise Patent Filing Strategy
“You get what you measure.” – Jack Welch
Once an enterprise decides to set any type of annual patent filing target, the question that next presents itself is how should that target be set?
The weakness in the traditional setting of targets for the number of patent applications to be filed by an enterprise, by simply modifying last year’s production targets according to how well the company is (or perceived to be) doing, is that this merely extends a series of annual guesses. Such targets are responsive to aggregate internal budgets and progress in top-line revenue, but are not responsive to strategic business goals of the enterprise or the benchmark performance rankings vis-à-vis its competitors. Rather, an enterprise should set (preferably soft) patent filing targets (and other goals) using certain metrics for its overall strategic goals, and a target ranking, and then allocate resulting patent productivity targets to individuals in the IP department to achieve those goals. Thus, the metrics discussed below may be driven by the following inputs:
- What is each P&L’s technology roadmap for the following year?
- How many planned new product introductions are planned for next year?
- How many brainstorming sessions do I have planned for next year?
- What are my competitors doing?
- What rankings among our competitors do we care about and where do we want to be in such rankings?
- What is my companies R&D budget next year and what should be my “research dollar-to-filing” ratio?
Once these inputs are collected, one or more of the following metrics may prove useful to the setting of annual patent application filing targets.
Competitive Benchmarking and Rankings
“When in doubt, count.” – Charles Babbage
One metric-based approach to determining patent filing targets is to compare the performance of an enterprise’s patent portfolio against its industry competitors. The enterprise may then objectively rank its patent portfolio performance. This competitive ranking determines what metric values are needed (and hence what patent productivity is needed) to reach its preferred ranking in the industry (e.g., top half, top quartile, top 10, or number one in its field). If the enterprise is a publically-traded company, where patent performance metrics presumably correlate with future stock price movement, patent filing targets should pursue the ultimate enterprise strategic goal – to increase market capitalization, competitive to the company’s peer group.
What remains to be determined, however, is what specific, patent-related metrics correlate best with future stock price movement or other corporate goals. One such example metric is simply the number of PCT applications filed by like enterprises. In 2015, the top 10 PCT applicants ranged from 3,898 filed applications (Huawei Technologies) to 1,378 (Philips) filed applications. Or, looking at domestic U.S. patent applications shown in Figure 2, the top 25 filers in 2014 averaged 2,761 filings and the top 250 filers averaged 700 filings. Again, this is simply one metric to guide the filing target setting process. We recognize that no single patent metric will always absolutely correlate with future business performance rankings because of the large variations in corporate size.
Another example benchmarking metric is “domestic patent intensity for R&D” (i.e., the number of U.S. patent application filings per $1M in domestic R&D expenditure) for certain industries and company sizes, which we calculated using data available from the National Science Foundation as shown in the two tables below:
Yet another example benchmarking metric (and frame of reference) is the “European patent intensity for R&D” (i.e., the number of European patent application filings per €1M in global R&D expenditures). The top 10 EU patent filling firms for the year 2011, for example, are shown in the table below:
Telling the Story with Caution
The use of the above-described comparative metrics for ranking peer groups can then be used to objectively measure and explain the patent budget and the patent story to management. For example, in the annual budgeting process, management can be shown that to raise the company’s peer ranking from, say, the third quartile to the second quartile, patent production must be raised from X to Y, which requires a budget increase from $W to $Z.
We now offer three cautions with using comparative benchmarking.
- For publically-traded companies, while the above-described metrics are insightful, patent metrics normalized to the size of R&D budgets may not have the best correlation with the future growth of market capitalization across multiple industries and countries, although they are of interest regarding the productivity of R&D.
- A requirement of these comparative metrics is that in order to be comparative across an industry sector, they must be based entirely on publicly-available data (even though the data may be obscure, esoteric, and awkward to work with for this purpose). Obviously, other better metrics based upon confidential data may be unavailable for peer group rankings, but useful for internal management.
- In the case of a large diversified conglomerate that is active in many industries, it is useful to calculate patent statistics and targets by each division/tier one business/strategic business unit, rather than for the company as a whole. This way the competitive benchmarks and targets for each division can be set in comparison with the competitors of each division. This can be particularly important because patent practices, patent intensity across revenue streams, and the sensitivity of revenue growth to patent activity, margins, and earnings tend to vary in different industries. That is, the coefficient of elasticity, or sensitivity, of stock price movement to various patent portfolio metrics tends to be particular to each industry. Hence, for example, it may be more useful to compare patent metrics for an equipment division with other competing equipment manufacturers, rather than comparing the patent metrics for the equipment division with the software or finance divisions of the same company.
Admittedly, the comparative metrics described above may be more useful for larger, publicly-traded companies with hundreds or thousands of active patent assets but are difficult to apply to smaller, younger, and/or private companies. Thus, we present some more metrics below that are more generally applicable across all enterprise types.
Protection of Markets
A useful metric for setting an annual patent filing target is the percentage of an enterprise’s sales or profits that are actually covered by one or more issued patents. This is useful to set a target to increase that coverage percentage over the next year to a select, targeted amount. In order to accomplish this, a matrix of product lines versus patent coverage must be created to indicate where sales are patent protected or unprotected. Then, patent applications needed to target specific opportunities for new patent coverage in the next year can be identified. Where products are not patent–protected, targets for patentable, second-generation product improvements may be scheduled (e.g., targeted “brainstorming” or “patent harvesting” sessions over the next twelve-month period). Where products are covered by patents, the date of expiration of such relevant patents can be verified in each case, and a plan should be developed to upgrade that product line with patentable second-generation product improvements prior to the expiration of such first-generation patents.
Is it always possible to develop patentable second-generation product improvements, when needed? Probably not. But in many industries there is an opportunity for second-generation improvement patents that extend the patent monopoly and competitive advantage by adding functionality using, for example, wireless communication, GPS assists, artificial intelligence, on-board equipment, cloud computing, IoT remote sensing, and data collection and analytics from a large population of equipment installations.
Protecting New Products and Technologies
Most mature enterprises use some form of phased design review – with formal tollgates and checkpoints – to introduce new technology, products, and services into the marketplace. Thus, another metric useful for setting an annual patent filing target is the percentage of such new product (or technology) introduction projects that are covered by patent applications or patents. Where projects have no associated patenting activity, targets for one or more patent applications may be scheduled to increase that coverage percentage over the next year to a select, targeted amount. Again, this may be accomplished by targeted “brainstorming” or “patent harvesting” sessions over the next twelve-month period to pursue patent filings prior to commercialization. This same process may be repeated for major internal IT, software and system projects to bolster the enterprise-wide target (and thus budget).
Another metric useful in setting an annual patent filing target is “revenue” (or losses) from patent licensing and litigation. In this sense, we mean revenue actually received from (or disbursed to) non-affiliated, third-parties (less broker fees), in the form of: (i) up-front or back-end cash fees; (ii) quarterly, annual or other periodic royalty cash payments; (iii) debt and/or equity securities; and/or (iv) any real or personal property; from IP-based transactions.
This begs the question of what are “IP-based transactions”? As much as CIPCs like to think that all of an enterprise’s sales and revenue are “IP-based,” we beg to differ. Here we mean transactions involving outbound, patent license agreements (even if copyrights, source code, know-how and/or trade secrets are included in the deal) and not revenue from: (i) the sale or leasing of patented products/services; (ii) savings realized from patent-pruning activities; (iii) revenue from government sole-source contracts awarded due to having patented products; or (iv) the like. Similarly, related metrics useful in setting an annual patent filing target also include the number of patent enforcement litigations and the number of patent injunctions obtained, and the sales and market share covered by such injunctions.
“Data is the sword of the 21st century, those who wield it well, the samurai.” – Jonathan Rosenberg, former SVP of products at Google
In today’s technological environment where data is abundant, it may be time for Chief IP Counsels across all enterprises to alter the annual exercise of “let’s set a filing target x% higher/lower for next year” exercise. Arbitrarily setting patent application filing targets detached from any business goals and metrics lead to arbitrary performance incentives for in-house IP personnel that are not directed by corporate goals and peer group benchmarking. Such a result is a failure of an IP counsel’s essential job function to help create or sustain their client’s profitability and to facilitate long-term enterprise growth. Rather, setting annual priority patent application filing targets that are driven by the enterprise’s overall goals, and guided by objective and relevant metrics, should be the norm.
 WIPO Facts and Figures 2016 at 12.
 Alternatively, the companies may be organized not by product lines or P&L’s, but by geographic regions (e.g., EMEA, APAC, LATAM, NA, ANZ, etc.).
 This is based on a 2014 survey of 109 CIPCs. See AIPLA, Report on the Economic Survey (2015) at 25.
 See American National Standards Institute, Inc./Society for Human Resource Management, Performance Management Standard (Nov. 2012) at 7.
 Id. at 16.
 This is not a trivial question given that in 2014, the reported median income of CIPCs with no subordinates was $240,000, while those with 1-5 subordinates reported a median income of $285,000. AIPLA, Report on the Economic Survey (2015) at 26.
 We note that there is a small, but growing cottage industry of trading and hedge fund activity which seeks to capitalize on the correlation of IP portfolios to financial market performance that is beyond the scope of this article. See generally S. Glazier, New Patent Metrics for Management and Investors, Law360 (Oct. 28, 2014).
 WIPO Facts and Figures 2016 at 17.
 See National Center for Science and Engineering Statistics, and U.S. Census Bureau, Business R&D and Innovation Survey (2014). A special thank you goes to Raymond Wolfe, Economist & Senior Analyst at NSF, for pointing us to the right data tables to use for our calculations. The NFS data represents a sampling of 44,162 companies representing the 1,998,858 companies located in the U.S. that perform or fund R&D and employ at least five employees.
 European Commission, Joint Research Centre Institute for Prospective Technological Studies, The Capability of the EU R&D Scoreboard Companies to Develop Advanced Manufacturing Technologies, Rep. EUR 27176 EN (2015) at 10.