Trend(time series) analysis contents scheme
Using augmented reality as an example of keyword of interest, and Apple's augmented reality as an example of Apple's keyword of interest, we provide trend analysis contents utilizing PatentPia GoldenCompass. If you click on the chain (link) marks that are attached to each item in the above utilization map, the example page will open in a new window.
Trends(time series) measurement time units
Trend (time series) analysis data in GoldenCompass is presented as i) yearly, ii) quarterly, and iii) percent change data.
Yearly data is provided for the most recent 11 years (excluding the current year for charts). For mobile, reports, etc., 6 years are provided due to space limitations. Quarterly data is provided for the 13 most recent quarters including the current quarter (excluding the current quarter for charts).
Rate of change data is available for: i) recent 10-year compound annual growth rate (CAGR), ii) recent 5-year CAGR, iii) recent 2-year growth rate, and iv) recent 2-quarter growth rate.
Trend data is available by nation. For each item (single items such as company, keyword, technology field, patent classification, etc., and compound items, which are combinations of two single items), we create a nation-specific patent set related to the item, and generate various trend data for the nation-specific patent set.
Kinds of trend(time series) analysis
There are 10 common types of trend (time series) analysis that exist regardless of single and composite items, and share analysis and concentration analysis that exist only for composite items. This is because share and concentration only exist when there is a subject (company/researchers, etc.) and a target field (keyword, technical field, patent classification, etc.).
We provide trend analysis contents using PatentPia GoldenCompass, using augmented reality as a single item example and Apple's augmented reality as a combined item example. If you click on the chain (link) display that is combined with each item in the above utilization map, the example page will open in a new window.
Context of the trend(time series) analysis (application location/utilization environment)
The context (application/utilization environment) of trend (time series) analysis is: i) the trend of the item itself, ii) the comparison of trends between items. The trend of the item itself has the form of a "1-line time series", while the comparison of trends between items has the form of an "n-line time series". Narrower trend analysis refers to trends within the item itself. This page mainly describes trend analysis of items themselves.
Time reference for trend(time series) analysis
Kind of time reference
The types of time that PatentPia GoldenCompass uses for the purpose of analysis are: i) first publication date, ii) most recent publication date, iii) date of patent issued, and iv) date of events.
Through open patent publications, external third parties can see/recognize something/someone/something in the patent. Accordingly, the date used by GoldenCompass for the purpose of analysis is the point of open cognizance of the something/somebody/somewhat.
After a patent is applied for, it is common for it to go through both an open and a registered process. If it is open, a patent publication is issued, and if it is registered, a registered patent publication is issued. However, in special cases, a patent may be registered directly without being open. In such cases, a patent publication is not published.
GoldenCompass (for the time being) does not use application date as a basis for analysis (aggregation/quantification); however, there is a roadmap to provide supplemental analysis based on application date in the future.
One of the patent systems adopted by almost all countries is the patent publication system. This system forces everything (something/someone/something) about a patent to be kept private, except for voluntary disclosure, before the patent publication period (usually 18 months after the application date). Because of this, the SOMETHING/SOMEBODY/SOMEWHAT for patents applied for within 18 months of the most recent filing date is almost always close to zero. This may be natural for patent professionals who know the system of open patents, but for the vast majority of people unfamiliar with the system, it makes it seem like patent activity is down/absent.
(For the record, any date can be used for analysis in PatentPia Analytics).
Day of first publication
The first publication date for a particular subject is the date that a patent publication related to that subject was first published. If both a publication and a patent publication are published for a particular patent, the first publication date for that patent will be the open date. If a patent was issued without the issuance of an open publication, or, in a very special case, if a patent publication was issued before an open publication, then the open date of the patent publication, the registration date, will be the first publication date for that patent.
Date of publication of the most recent publication
The most recent publication date for a particular subject is the date of the most recent issued patent publication related to that subject. If both open and patent publications are issued for a particular patent, the most recent publication date for that patent will be the date of patent issuance, which is the date of registration. If a patent was issued without the publication of a patent publication or, in very special cases, a patent publication was issued before the publication of a patent publication, the open date of the patent publication, the registration date, will be the most recent publication date for that patent.
As a note, it is possible that a correction publication may be issued after a particular patent has been issued. However, PatentPia's recent publication date does not use the correction publication date.
Date of patent registration
The date listed in the issued patent publication as the date of patent is the date of registration.
Day of Events
The events provided by GoldenCompass include: i) patent transactions/purchased patents through M&A, ii) patent litigation, iii) patent trials (contentious trials with opposing parties, such as invalidation trials and scope determination trials), iv) standards, v) FDA approvals, vi) government-funded R&D, and vii) financial events such as M&A or investments.
Patent transactions are based on the transaction recording date (not the execution date) in the US. Outside of the US, they are based on the issued date.
Litigation is based on the date of filing in court, and trial is based on the date of judgment.
Standards are based on the standard declaration date.
FDA approvals are based on the approval date of the request for FDA approval.
Government-funded R&D is based on the day of the first official publication of a national R&D patent.
M&A or investments are based on the announcement date.
Analysis by time reference
First patent publication date
First publication date is the most used time base. Typical trend analyses include: i) patent publication and something per patent during portfolio analysis ii) patent families, and iii) convergence.
Date of recent publication
The recent publication day is only used for forward citations analysis. A citation analysis is an analysis of the occurrence of forward citations (cited patents) that, given a patent set related to a particular item (its own (input/selected) item), cite the individual patents that comprise the patent set. Thus, the time base is the forward cited patents (not the patents related to the particular item (its own (input/selected) item)).
However, among the publications of a late-filed patent, there are many cases where the issued patent publication records the first-party patents (patents related to a specific item (its own (input/selected) item) that the late-filed patent cites. Some late-filed patents may also record the first-party patents that the late-filed patent cites in the open publication of the late-filed patent, but such cases are relatively rare. Therefore, the analysis of the citation of patents related to a specific item (its own (input/selected) item) is made based on the recent publication date of the late-filed patent.
Date of patent
Everything related to registration is based on the date of patent: i) No. of registrations/something per issued patents, ii) Issuance rate/time to register, iii) Independent claim scope retention ratio/claim count retention ratio, etc.
Days of events
The analysis of events is based on the day the event occurred.
Portfolio trend analysis
GoldenCompass' portfolio trends include i) Application publication, ii) Registration, and iii) No. of claims per published/registered patents. For patent sets related to its own (input/selected) items, it aggregates i) Total patents, ii) No. of issued patents, and iii) No. of claims per patent (filed), which aggregates based on patent publication, or No. of claims per patent registered, which aggregates based on patent registration publication, respectively.
Event trend analysis
2 families of events trend analysis
There are two families of event trend analysis: i) event trends, which aggregate the events themselves/event counts by day (year/quarter) of event occurrence, and ii) event (used) patent trends, which aggregate the patents used in the event by day (year/quarter) of event occurrence. The former, event trends, aggregates events in which a patentset related to its own (input/selected) item was used; the latter, event (use) patent trends, aggregates the No. of patents involved in events for only those patentsets related to its own (input/selected) item in which the event occurred.
1 patent vs. n events
One patent can be related to multiple events (a relationship of "patent 1 : event n"). One patent may be used in multiple litigations, and it may be assigned multiple times.
Accordingly, when there are N patents, the sum of the No. of assigned patents per yearly event may be greater than n if some patents are assigned in two or more different years.
On the other hand, among the events, trials occur on a per-patent basis (a "patent 1 : trial 1" relationship). For all events except trials, one or multiple patents can be the subject of the same event (the "Event 1 : Patent N" relationship).
Varieties of litigation/trial events analysis
1.
the existence of plaintiffs/defendants, trials as petitioner, and claims/claims as respondent
For events with parties such as lawsuits or judgments, a company is either sued (defendant) or sued by patents in the patent set related to its (input/selected) items; in the former case, it is the defendant, and in the latter case, it is the plaintiff.
2.
Unrelatedness of the litigation/trial to the current patentee (patent owner)
Patent litigations are not only filed using your own patents in possession. If you have an exclusive license to practice someone else's patent, you can also sue them for patent infringement based on your exclusive license. A special form of patent litigation is a lawsuit seeking a declaratory judgment. This is a lawsuit in which a person sues another person (the defendant) and claims that he or she has not infringed a patent in possession of that other person. Thus, it is also a lawsuit in which the defendant in the lawsuit (not the plaintiff/complainant) is the holder of the patent.
Accordingly, the No. of litigations/trials is measured as the number of times a patent set related to its own (input/selected) item has been used in a lawsuit/trial, regardless of the current owner of the patent. The No. of patents used in litigation is measured as the No. of patents in the patent set related to its own (input/selected) item that have been used in litigation/trial, based on the year in which the litigation/trial occurred.
3.
No. of plaintiff litigations vs. defendant litigations
The number of plaintiff litigations is measured by the No. of litigations filed by a specific company/organization in its capacity as a plaintiff using patents in the patent set related to its (enter/select) item. The number of defendant litigations is measured by the No. of litigations filed by a specific company/organization as a defendant using patents included in the patent set related to its own (input/selected) items. No. of plaintiff litigations vs. No. of defendant litigations In this analysis, the current owner of the patent is not considered.
4.
No. of patent trials as petitioner vs. No. of trials as respondent
The No. of trials claimed measures the No. of trials filed by a particular company/organization in its capacity as petitioners in trials using patents included in the patent set related to its (input/selected) items. The Number of Trial Respondents measures the No. of trials in which a particular company/organization has been used in trials as petitioners, using patents included in the patent set related to its own (input/selected) items. In the analysis of No. of trials as petitioner and No. of trials as respondent, the current owner of the patent is not considered.
Using Apple as an example of an interested company, we provide trend analysis contents using PatentPia GoldenCompass. If you click on the chain (link) marks that are attached to each item in the above utilization map, the example page will open in a new window.
Forward citation trend analysis
Definition of forward citation trend analysis
Forward citations trend analysis is arguably the most important analysis in patent analysis. There are many different detailed analyses of forward citations, in proportion to the importance of the analysis.
When a particular subsequent patent references a preceding patent, from the perspective of the referenced patent, the preceding patent is in a citation relationship with the particular patent. From the perspective of the preceding patents, the No. of subsequent patents that cite themselves can be aggregated by the year in which the citations by the subsequent patents occurred or by quarter.
Patents in the patentset related to the own (input/selected) item may also include patents that have been cited by subsequent patents. Accordingly, the No. of patents that cite patents in the patentset related to the own (input/selected) item (forward citations) can be aggregated by year or quarterly by the year in which the citation by the subsequent patent occurred. This aggregate analysis is called forward citations analysis for its own (input/selected) item (and related patent sets).
(Note) The opposite direction of forward citation analysis, i.e., analyzing the preceding patents cited by the patentsets related to the own (input/selected) item by the day the reference occurs, is called reference analysis. A reference analysis is everything exactly like a forward citations analysis, except for the direction and the date it is based on. The reference day for a reference analysis is the year of the most recent publication of a patent in the patent set related to its own (input/selected) item.
Sources of forward citations
Forward citations occur by three main routes. i) by examiners referring to preceding patents during the examination of a particular patent, ii) by applicants referring to preceding patents related to their own patent by IDS, etc., and iii) by applicants referring to preceding patents related to their own patent in their background art, etc.
In particular, in the case of i), the preceding patents that are referenced by examiners in office action (OA) for the purpose of negating the patentability of a particular patent are called "referenced patents by examiners in office action." In this case, from the perspective of the referenced patents, the subsequent specific patents become "forward citations by examiners in office action."
Patent vs. right holder relationship between referenced patents : full forward citations vs. self-forward citations
1.
Definition of self citation
The right holders of the referenced patents and the forward citation patents that cite the referenced patents may be the same at a certain point in time, but in most cases, they are different. In this case, when the right holders of both patents are the same, it is said that the preceding patents in possession are cited by the subsequent patents in possession, and "self forward citation" has occurred by the subsequent patents from the perspective of the preceding patent.
2.
Two criteria for determining self forward citation
GoldenCompass determines whether or not self-forward citations are self-forward citations based on the same right holder as the current right holder. If the patents belong to different right holders as applicants, but the right holders become the same at the current time through assignments/transactions, etc., it is treated as a "self-forward citations" relationship.
Of course, it is possible to judge self-forward citations based on the applicant as the right holder at the time of application, but GoldenCompass believes that it is relatively more valid to judge self-forward citations based on the current right holder for several reasons, one of which is that self-forward citations are key indices of R&D continuity.
If a particular set of patents related to a particular self (input/selected) item held by a particular company is heavily cited by subsequent patents held by that company, it can be said that the item is being continuously researched and developed (R&D). Continuous R&D will result in a large number of patent applications for related technology, which will reference their own preceding patents. From the perspective of the preceding patents, there will be many self-forward citations. Therefore, if a particular item is not cited by its subsequent patents, it can be considered that there is no or little additional relevant R&D.
Therefore, when judging R&D continuity at the right holder level, it may be more useful/relevant to judge self-forward citations based on "current right holder vs. current assignee" who currently holds the patent, rather than "applicant (past right holder) vs. applicant" who may have sold/abandoned the patent.
Criteria for forward citations : forward citations of individual patents vs. forward citations of patent families
1.
Need to expand the scope of reference patent families and forward citations
When a particular subsequent patent references a preceding patent, it does not necessarily list all related preceding patents without omission: it may not be aware of the related preceding patents, it may accidentally omit them, and it may not be necessary to cite all patents as precedents just because they are related. Because of this, relevant preceding patents may be omitted from the entire family of preceding reference patents, and from the point of view of the preceding reference patents, relevant patents may be omitted from the family of subsequent forward citations.
2.
Expanding the scope of reference patent families and forward citations using patent families concept (coming soon)
At this point, the No. of patents in reference and forward citations can be supplemented in a rational way by using the concept of patent families.
A patent family is a No. of domestic and international patents that share one or more priorities and have a high degree of commonality in the technical content of the patents. Therefore, when a particular subsequent patent refers to a preceding patent, it can be assumed that it is also referring to the patent family of the preceding patent, which has a high degree of commonality in the technical content. Similarly, when a particular subsequent patent refers to a preceding patent, it is safe to assume that the patent families of the subsequent patents that have a high degree of commonality in technology are also referring to the preceding patent. In other words, when a particular subsequent patent refers to a preceding patent, it is safe to assume that the entire patent families of the subsequent patent refer to the entire patent families of the preceding patent.
Forward citations are typically measured on an individual patent basis. The concept of patent families and expanding the scope of reference patent families is to expand the scope of forward citations to patent families rather than individual patents, thereby measuring forward citations at an expanded scope.
Nation(intellectual property office) relationship between patents vs. forward citations : domestic forward citations vs. overseas patents
If the subsequent patents that cite a particular patent occur within the same nation, the forward citations are said to be domestic. On the other hand, if the subsequent patent is issued overseas, the patent is called a forward citation by overseas patents. In the past, except for patents in English, patents from non-English speaking nations were difficult to reference or forward citations due to language barriers. However, nowadays, the language barrier has been greatly reduced due to machine translation. As a result, the number of references and forward citations by overseas patents is increasing rapidly.
Citation metrics: total citations vs. No. of forward citations per patent
Forward citations weighing criteria can be broadly categorized into i) analysis from a total volume view, and ii) analysis from a density (per patent) view.
Analysis of citations from a total volume view refers to the analysis of all subsequent forward citations patent sets from the perspective of the own (input/selected) item (and related patent sets) as a whole, quantified as a whole. In this case, the larger the own (input/selected) item (and related patentsets), the larger the number of forward citations. In this case, it may happen that an individual patent does not receive many forward citations, but its own (input/selected) item (and the patent set related to it) becomes so large that the overall No. of patents forwarded becomes a large value. Therefore, there may be an optical illusion when looking at the No. of patents forwarded alone.
On the other hand, each patent in the composition of its own (input/selected) item (and its related patent set) may receive many citations, resulting in a high value of the density (No. of forward citations per patent). In other words, the No. of forward citations per patent is independent of the size of its own (input/selected) item (and its related patent set). Thus, if the citation count represents the quantitative citation view of the self (input/selected) item (and the related patent set), the No. of forward citations per patent represents the qualitative citation view of the self (input/selected) item (and the related patent set).
Means of forward citation analysis
Forward citations analysis can have a variety of purposes/meanings. Some of them are i) reputation, ii) impact , iii) competitiveness, iv) business opportunity, v) technology diffusion, etc.
Reputation
As in a paper, if a particular patent is being cited by subsequent patents more than others, it can be said that the patent has a higher reputation than others. Similarly, if a self (input/selected item) (and its related patent set) receives more forward citations than other self (input/selected item) (and its related patent set), it can be said to have a relatively higher reputation.
Reputation can be further enhanced by the following factors
•
i) Temporal recency (if the temporal distribution of the No. of forward citations is more concentrated in the recent past than in the past) ,
•
ii) Qualitative view (if the index of citations from a qualitative view, such as No. of forward citations per patent, is higher),
•
iii) refined citations (forward citations by examiners or examiner citations in OA are better than general citations or citations by IDS),
•
iv) forward citations by important patents.
◦
Purchased/transaction-generated patents,
◦
Patents used in litigated patent litigations/examinations/etc,
◦
Standard-essential patents/Patents used in FDA approvals
◦
Patents with many international patent families (high investments in building overseas patent networks)
◦
Patents with many self-forward citations (large R&D continuity)
◦
Patents that have received many forward citations
In particular, the self (input/selected) item (and the related patent set) may be i) related to a specific right holder, ii) related to a specific researchers, or iii) a specific field of the right holder/researchers. In this case, the (input/selected) item itself (and the related patent set) may be highly evaluated from a reputation view.
Impact (impact)
If a patent is being cited by subsequent patents, it can be said that the patent has a technology impact on the subsequent patents. Similarly, a self (input/selected) item can be considered to have a technology impact on subsequent patents that cite a patent set related to the self (input/selected) item. In particular, the cited subsequent patents may be i) related to a specific right holder, ii) related to a specific researcher, or iii) related to a specific field of the right holder/researchers. In this case, the (input/selected) item itself (and the set of patents related to it) can be considered to be impacting a specific right holder/researcher or a specific field of that right holder/researcher. Conversely, a particular right holder/researcher or a particular field of the right holder/researcher may be considered to be impacted by the input item (and the selected item).
In particular, the impact can be even higher if the subsequent patents being cited are
•
i) Temporal recency (the temporal distribution of the No. of forward citations is more concentrated in the recent past than in the past) ,
•
ii) Qualitative view (if it has a higher index of citations from a qualitative view, such as No. of forward citations per patent),
•
iii) refined citations (forward citations by examiners or examiner citations in OA are better than general citations or citations by IDS),
•
iv) forward citations by important patents.
◦
Purchased/transaction-generated patents,
◦
Patents used in litigated patent litigations/examinations/etc,
◦
Standard-essential patents/Patents used in FDA approvals
◦
Patents with many international patent families (high investments in building overseas patent networks)
◦
Patents with many self-forward citations (large R&D continuity)
◦
Patents with many forward citations
Competitiveness
Comparison analysis of citations to multiple self (input/selected) items can be utilized to evaluate the relative patent competitiveness of self (input/selected) items. In particular, comparison analysis after removing the impact of the size of self (input/selected) items (patent size), such as No. of forward citations per patent, is more useful for analyzing patent competitiveness among self (input/selected) items.
Comparative analysis of competitiveness is more useful for analyzing patent competitiveness between own (input/selected) items, such as when the subject of competitiveness analysis is own (input/selected) item I and own (input/selected) item J, etc.
•
Analysis of comparison with companies
◦
Yourself vs. a group of competing companies
◦
Self vs. competing companies in a specific field
◦
Multiple companies in your supply chain in a specific field
◦
Between potential M&A/investing companies
◦
Between companies as potential collaboration/co-research targets
•
Analysis of comparisons with researchers
◦
Between researchers within an organization
◦
Between researchers within an organization in a specific field
◦
Between researchers targeted for recruitment (recruitment/HR)
Business opportunity
If the own (input/selected) item (and the related patent set) is receiving many forward citations, various business opportunities may arise for the right holders (companies) of the subsequent patents. Business opportunities may include i) patent licensing, ii) transfer of technology/patents, iii) settlement/damages from warning letters/infringement lawsuits due to possible patent infringement, etc. In particular, if the own (input/selected) item (and the related patent set) is a specific researcher or a specific field of a specific researcher, the researcher may be a key target for the human resource (HR) business.
The business opportunity is likely to be even more opportunistic if the right holder of the subsequent patent is
•
The right holder (company) has the ability/culture/willingness to make investments in IP
◦
has recently made patent purchases or licensed practice rights
◦
Recent M&A/investments in related companies in the technology field
◦
The right holder has improved solvency
▪
Received investments, or,
▪
Had an IPO,
▪
stock price has skyrocketed
▪
Financial statements (profit/sales, etc.) have improved
•
The right holder (company) is in trouble
◦
Has been sued as a defendant
◦
Has petitioned for a trial
◦
Received a warning letter
•
The right holder (company) is actively enforcing its rights
◦
Litigation has been filed (reinforcement of ongoing litigations)
◦
Has been petitioned for trial
◦
A warning letter has been sent
•
Changes in the competitive relationship/competitive landscape in which the right holder (company) is engaged
◦
Competitor's patent investments (patent purchases, etc.)
◦
Competitors' M&A/investments in technology companies
◦
Competitor's M&As, investments, or IPOs
•
Changes in related industries
◦
Market leader's patent investments (patent purchases, etc.)
◦
Market leader's M&A/investments in technology companies
◦
Market leader's share changes
•
Business opportunities are likely to be more opportunistic if the subsequent patent is
•
i) Temporal recency (the temporal distribution of the No. of forward citations is more concentrated in the recent past than in the past) ,
•
ii) Qualitative view (if the index of citations from a qualitative view, such as No. of forward citations per patent, is high),
•
iii) refined citations (forward citations by examiners or examiner citations in OA are better than general citations or citations by IDS),
•
iv) forward citations by important patents.
◦
Purchased/transaction-generated patents,
◦
Patents used in litigated patent litigations/examinations/etc,
◦
Standard-essential patents/Patents used in FDA approvals
◦
Patents with many international patent families (high investments in building overseas patent networks)
◦
Patents with many self-forward citations (large R&D continuity)
◦
Patents with many forward citations.
Technology diffusion
If a particular patent is being cited (referenced) by a subsequent patent, the technology of that particular patent can be said to be diffusing into the technology field of the subsequent patent. Similarly, the technology contained in its own (input/selected) item (and its related patent set) can be said to be diffusing into the technology field of the subsequent patent.
Technology diffusion is more important in the following cases.
•
Diffusion to patents related to field that is completely new (diffusion to new/heterogeneous areas)
◦
If the technology/product-part/products-materials field, patent classification (CPC/IPC), etc. of the recent cited patent is not found in the patent classification of the past cited patent
◦
Keywords included in the recent forward citations are not found in the keyword families of past forward citations.
•
Diffusion to patents related to field that is rapidly growing/hot
•
If it is a field of rapid growth/focus of new investments by market leaders
Patent family trend analysis
Definition of patent families
A patent family is a group of domestic and international patents that share one or more priorities and have a high degree of commonality in the technology content of the patents. Patent families can be categorized into domestic patent families that occur within the same nation and international families that occur in different countries.
The main tracks on which patent families arise are priorities. The priorities are detailed into domestic priority and treaty/foreign priority which arises from overseas applications. Related applications, such as i) divisional applications, ii) continuation applications, iii) continuation-in-part applications, etc., which usually represent the original application, are also the main tracks in the composition of patent families. Reissue patents, application conversions, etc. are also major tracks from which patent families arise.
Definition of patent family trend analysis
Patent family trend analysis for own (input/selected) items (and related patent sets) is an analysis of patent families (patent network) construction by i) domestic, ii) overseas, and iii) nation, analyzing i) patent (network construction) strategies and ii) investments by nation.
On the other hand, for own (input/selected) items (and related patent sets), i) No. of domestic patent families per patent and ii) No. of international patent families per patent can be analyzed. Density/qualitative analysis such as per patent analysis is very helpful to understand the global patent strategy of a company or a company's specific field because it excludes the effect of the size of own (input/selected) items (and related patent sets).
Means of patent family analysis
Applying for, obtaining, and maintaining patents and rights in multiple countries is costly. Despite the high cost, establishing No. of domestic patent families, especially in multiple countries overseas, is a strategic choice on the part of a company/organization. From the view of rationality of expenditure and efficiency of resource allocation, the amount of money spent at a particular point in time can be viewed as representing the importance of a particular action/choice at that point in time.
Since patents are a product of R&D, they are applied for and opened at a much earlier stage than products/services, i.e., patent information is future intelligence.
Selecting strategic market nations/locations
Before a company enters a particular nation with its product/service, it tries to secure a patent in that country first to minimize the patent legal risk of implementing the product/service. Except in very special cases (e.g., pharmaceutical substances), no company takes the inefficient strategy of securing sufficient patents in all countries in all fields, but rather focuses its available resources (money) selectively. Thus, if a company has a diverse portfolio of technologies/products/services, its go-to-market strategy can be analyzed in terms of which technologies/products/services it intends to focus on in which countries, through the No. of patent families it forms in those countries.
Fields of strategic choice
A company's technology/product/service portfolio is often diverse, so analyzing whether a company forms a large number of patent families for patents related to field can reveal the company's strategic choice field. In particular, by utilizing the essence of patent information as future intelligence, it is possible to analyze the future technology strategies of competing companies or industries. In particular, by analyzing keywords or detailed technology/patent classification (e.g., CPC end node) included in a specific patent portfolio with a large number of patent families, you can discover key points of the company's strategic choice field.
Key researchers within the company
By analyzing the total No. of international patent families or No. of international patent families per patent by researchers, you can discover key researchers within a company. The formation of international patent families is costly; therefore, researchers with a higher No. of international patent families per patent than the average for that company or researchers (well above the average of the distribution) are the ones who are chosen by the organization from the view of cost expenditure. These researchers are likely to be the key researchers.
Critical/Strategic Technologies
The ways in which patent families are formed within a nation are i) divisional applications, ii) continuation-in-part applications, iii) continuation applications, etc. If the content of the patented technology is rich and the technical ideas do not have diverse views, there will be nothing to divide, nothing to upgrade/add/add/extend/connect through continuation-in-part applications, etc. Therefore, the (detailed) fields in which more than the average number of patent families (far above the average of the distribution) have been formed within a nation are likely to be important/strategic fields for the company.
Patent family trend analysis
The key points of patent families trend (time series) analysis are i) recent spikes, ii) recent rapidly falling, iii) steady accumulation, etc.
Analysis & monitoring of fields of recent surge in patent families
Patent family formation involves a lot of expenditure; therefore, specific (detailed) fields with rapidly growing patent families, specific detailed fields of a particular company, are highly worth monitoring. Detailed analysis of keywords included in patents belonging to these spike fields, detailed technology/patent classification (e.g., CPC end node), etc.
Recent Rapidly Falling
There may be a rapidly falling No. of patent families for a particular company (e.g., competing companies) in a particular nation for a particular field. Such rapidly falling signals can be interpreted as i) a cessation of further R&D, ii) an inward-looking decision to exit the market (management's decision/direction), iii) a reduction in spending due to financial vulnerability, etc. As a side note, rapidly falling signals are just as important and useful as soaring signals.
Steady accumulation
There is no more relentless and powerful market domination strategy than the steady accumulation of patent families. Steady accumulation of patent families refers to i) steady investments in R&D expenditures, ii) the generation of R&D outputs leading to patent creation, iii) the increasing capabilities of researchers through steady R&D, iv) the accumulation of research outputs, and v) the resulting comprehensive patent network.
Quality trend analysis
Quality measurement Items
Patent quality can be quantified by its own (input/selected) items (and related patent sets). Quality measurement items include i) registration related items, ii) claims related items, iii) content richness related items, iv) time related items, v) forward citations related items, etc. It makes sense to utilize/interpret items related to quality measures in a comprehensive manner rather than utilizing/interpreting them in isolation. Statistically, "convergence to the average/larger the size of the group, relatively less variability" occurs. Therefore, it is better to use items related to quality measurement on a patent set basis, rather than for a single patent.
Registration related items
Registration-related quality measures include i) issuance rate, ii) time to issue, etc.
Patent applications are registered after a rigorous examination of patentability requirements (novelty, non-obviousness, etc.) by examiners of the Intellectual Property Office. Since patents that are not patentable, that do not fulfill patentability requirements, are not registered, the average issuance rate can be used as a quality measure from the view of patentability of its own (input/selected) items (and related patent sets).
The Intellectual Property Office examines all the applied claims of a patent application to determine whether the claims meet the patentability requirements; if they do not, the examiner will not grant the patent issuance and will send an opinion of unpatentability to the applicant, along with supporting prior art, and provide an opportunity for the applicant to submit a rebuttal. Upon receipt of the applicant's written opinion and/or amendments (corrections) to the claims, the examiner re-examines the patent claims and issues a decision of allowance, or a decision to issue an opinion or rejection.
All these processes of examination of patentability/patentability requirements are time consuming. Therefore, if the average time to registration is high, it is likely to be due to i) lack of patentability requirements or existence of related prior art, ii) reduction in scope of patent rights or deletion of claims for registration. These have a negative impact on patent quality. Therefore, the average issuance time can be used as a measure of quality.
Claims related items
Claim related quality measures include i) independent claim scope retention ratio, ii) claim count retention ratio, etc.
Patent applicants expect their patented inventions to have broad scope of rights. Accordingly, in patent applications, applicants or their agents request as wide a scope of rights for independent claims (the most important of patent claims) as possible. Broadly claimed independent claims are characterized by i) fewer elements (because the scope of rights is an AND (intersection, not union) condition of the elements), and ii) fewer limitations/limiting elements. In simple words, broadly claimed independent claims are shorter in length and have fewer words.
Broadly claimed independent claims are more likely to have a large amount of related prior art, and are therefore more likely to receive an initial rejection notice (including an office action notification) from the PTO examiner. Upon receipt of a rejection notice (including an office action notification), the applicant or agent will make changes to the independent claim. This act of alteration is manifested in i) addition of elements, addition of limitations/limiting factors, etc., i.e., the length of the independent claim becomes longer and the number of words increases. The examiners of the Intellectual Property Office examine the altered independent claim and, if it meets the requirements of the patent, issue an allowance notice to the altered independent claim.
Therefore, if the scope of the rights in an application requested to be broad is not narrowed, or is registered with a relatively less narrowed scope, it can be viewed that the scope of the rights has been well maintained. Therefore, the independent claim scope retention ratio, which is measured by the rate of change in the length/word count of the independent claim, etc. can be used to measure quality from the viewpoint of rightfulness.
In the process of examination, opinion submission, and amendment (revision) of a patent application, claims that are not recognized as patentable are deleted. On the other hand, the claims that are certified as patentable are non-exclusively retained at the time of registration. Therefore, the No. of claims at the time of registration is often smaller than the total No. of claims at the time of application. Therefore, the claim count retention ratio, which is measured as the rate of change between the No. of claims at the time of application and the No. of claims at the time of registration, can be used to measure quality from the view of rightfulness.
Items related to the richness of technology content
Items related to the richness of technical content for the own (input/selected) item (and the related patent set) Related quality measures include: i) convergence.
The convergence of the own (input/selected) item (and the related patent set) is measured by the average number of patent classifications (CPCs) per patent.
Every patent has one main classification CPC and 0 to n sub classification CPCs. Depending on the technology content of the applied patent, the patent examiner assigns a patent classification selected from a standardized patent classification scheme used by intellectual property offices worldwide. The patent classification assigned corresponds to the technical viewpoints/features/elements/attributes, etc. contained in the patent. Thus, if a patent is assigned multiple patent classifications, it can be viewed that the patent contains various technical viewpoints/features/elements/attributes within it.
The standard patent classification schemes are the international patent classification (IPC) and the cooperative patent classification (CPC). For more information on IPC and CPC, see the page "Patent Classification IPC/CPC"
If a patent contains various technical views/features/points/attributes within one patent, it can be considered as having high technology diversity/richness. If there is high technology diversity/richness, it is non-exclusive, and the utilization/application/applications are likely to be diverse.
Therefore, we can evaluate the convergence of its own (input/selected) items (and related patent sets) by the average No. of patent classifications (CPCs) per patent for its own (input/selected) items (and related patent sets).
Time related items
Time-related quality measures include i) precedence.
Intellectual property offices around the world adopt a first-to-file system, which means that patents are granted to those who apply first. Therefore, it is important to apply for a patent first. Even when multiple patents are applied for in a given field, the patent that is applied for first will have a relative advantage. The extreme form of a patent with an advantageous position is the so-called "original patent." An original patent is the first patent filed in the field and is widely used in the technology/product/service in the field.
Every patent has a patent application date, so it is possible to calculate the distribution/statistics (average, standard deviation, histogram, etc.) of the application dates of all patents in the population under comparison. Similarly, the distribution/statistics of patent application dates of patents of a particular entity (company/researchers, etc.) can also be calculated.
In this case, if the average value of the application dates of the patents of a particular entity is less than the average value of the application dates of the patents of the population under comparison (temporally preceding), then the patent portfolio of the particular entity has a high precedence (at least 50.1%) on average.
The population for comparison can be technology/products-parts/products-materials/patent classification (CPC, etc.)/keywords, etc. In some cases, it can also be n companies in a particular field. The subjects of the precedence calculation can be companies/researchers, etc.
Forward citations related metrics
The quality measures related to citations include i) No. of forward citations per patent.
Citations are measured by various categories and combinations of categories, such as i) overall vs. examiner vs. examiners in oa, ii) overall vs. self, iii) domestic vs. international, iv) all patents vs. special patents, etc. Therefore, citations per patent can also be measured by combinations of these categories and categories.
If a particular self (input/selected) item (and related patent set) has a higher citation-related quality index, such as No. of forward citations by examiners in office actions per patent, than another self (input/selected) item (and related patent set), then the particular self (input/selected) item (and related patent set) can be said to have high quality.
Related contents
•
.