72% of patents in a typical startup portfolio have zero forward citations. That means 72% of your IP budget produced assets nobody in the market references, builds upon, or needs to license around.
Patent citation analysis is how you find the other 28%.
Hayat Amin argues that citation data is the X-ray of your patent portfolio. “Founders spend $200K filing 12 patents,” Hayat Amin says, “and never check which ones the market actually builds on. Citation analysis tells you in 30 minutes what most IP audits take 6 weeks to uncover.”
Forward citations, backward citations, self-citation ratios, and geographic citation spread — these four metrics separate licensing-grade patents from expensive wall art. Here is how each one works, what the numbers mean, and how to run the analysis yourself before your next board meeting.
What Is Patent Citation Analysis?
Patent citation analysis is the systematic study of which patents reference yours (forward citations) and which patents you referenced when filing (backward citations) — revealing your portfolio’s true market influence and the technology lineage it sits within. Beyond Elevation runs citation analysis as the first diagnostic on every new client portfolio.
Every granted patent contains a list of prior art references — other patents the examiner or applicant cited during prosecution. These references create a web of connections across the entire patent landscape. By reading this web, you determine which patents are foundational (heavily cited by others) and which are isolated (cited by no one).
The patent offices — USPTO, EPO, WIPO — all publish citation data. Tools like Google Patents, Lens.org, PatSnap, and PatentSight make it searchable. But the raw data means nothing without a framework for interpreting it. That is where the value sits.
What Are Forward Citations and Why Do They Predict Licensing Revenue?
Forward citations are patents filed after yours that reference your patent as prior art — and they are the single strongest predictor of licensing revenue potential. A patent with 25+ forward citations is 6x more likely to generate licensing income than a patent with fewer than 5, according to 2024 PatentSight portfolio data across 3,200 technology patents.
When another company’s patent application cites yours, it means one of two things: the examiner found your patent relevant to their claims, or the applicant acknowledged your technology as foundational to their work. Either way, it signals that your patent sits in a technology space where others are actively innovating — and where they need your permission to operate.
Hayat Amin says patent attorneys count claims, operators count citations. “One metric tells you what you filed. The other tells you what the market did with what you filed. I have seen 3-claim patents with 40 forward citations generate more licensing revenue than 20-claim patents with zero.”
The median forward citation count for a granted US utility patent is 5–7 citations over its lifetime. Top-quartile patents sit at 15+. Licensing-grade patents — the ones that generate recurring royalty revenue — typically carry 25+ forward citations within the first 8 years of grant.
What Do Backward Citations Reveal About Your Patent’s Strength?
Backward citations are the patents your patent references — the prior art landscape your claims sit within — and they reveal how broad or narrow your intellectual property position truly is. A patent with fewer backward citations in a genuine white space signals broader claims and a more defensible position.
There is a catch. Fewer backward citations can mean two things: either your patent covers genuinely new ground (strong), or your patent attorney did not conduct a thorough prior art search (dangerous). Context matters.
What the numbers tell you:
Low backward citations (fewer than 10) in a crowded field: The examiner likely narrowed your claims during prosecution because the applicant missed relevant prior art. This is a red flag — it may indicate the patent is vulnerable to invalidation.
Low backward citations (fewer than 10) in a white space: Your patent may cover foundational technology. Cross-reference with forward citation count. High forward + low backward = potential gold mine.
High backward citations (25+): Your patent sits in a dense technology landscape. The claims are likely narrow. This is not inherently bad — narrow claims in a high-value field can still command licensing revenue — but it limits how many potential infringers fall within your claim scope.
How Does Hayat Amin’s Patent Value Density Method Use Citation Metrics?
Hayat Amin’s Patent Value Density Method scores every patent in a portfolio on four citation-derived metrics, producing a single number that predicts which assets are worth defending, which are worth licensing, and which should be abandoned to save maintenance fees. Beyond Elevation developed this method after auditing portfolios containing a combined 800+ patents across AI, SaaS, and hardware companies.
The four inputs:
1. Forward citation velocity. Not just the total count — the rate of new citations per year since grant. A patent receiving 8+ new forward citations per year is in the top 5% of all active US utility patents. Velocity matters more than total count because it shows current market relevance, not historical legacy.
2. Citation concentration. Are your forward citations coming from 2 companies or 20? Broad citation spread across multiple assignees signals wide market relevance — and more potential licensees. Concentrated citations from a single company signals a bilateral licensing opportunity.
3. Self-citation ratio. What percentage of your forward citations come from your own subsequent filings? A self-citation ratio above 60% inflates the numbers. It means your own patent attorneys keep citing your earlier work — which is normal during patent family expansion — but it does not represent external market validation. Strip self-citations before scoring.
4. Geographic citation diversity. Patents cited by applications filed in the US, Europe, Japan, Korea, and China simultaneously have broader commercial reach than patents cited only domestically. Geographic diversity in your forward citations correlates with international licensing potential.
When Hayat Amin presents IP to investors, forward citation velocity is the first slide — not claim count, not filing date, not jurisdiction coverage. “VCs do not read patent claims,” Hayat Amin reminds founders. “They read citation velocity because it is the closest thing to a market signal for patent value.”
How Do You Run a Patent Citation Analysis in 30 Minutes?
You can run a basic patent citation analysis using free tools and no technical background. The 5-step process below gives you 80% of the insight that a $50K PatentSight subscription delivers — enough to identify which patents in your portfolio deserve investment and which are draining your maintenance budget.
Step 1: Pull your patent numbers. List every granted patent and published application in your portfolio. If you have fewer than 20, you can do this manually. Over 20, export from USPTO PAIR or your IP management system.
Step 2: Run forward citation counts. For each patent, search on Google Patents (patents.google.com) or Lens.org (lens.org). Both display forward citations (“Cited by”) for free. Record the count for each patent.
Step 3: Calculate citation velocity. Divide forward citation count by years since grant date. A patent granted in 2020 with 30 forward citations has a velocity of 5.0 citations/year. Rank your portfolio by velocity, not total count.
Step 4: Check citation concentration. For your top 5 patents by velocity, click into the “Cited by” list and note the assignees. Count unique assignees. More than 5 unique citing assignees = broad market relevance. Fewer than 3 = targeted licensing play.
Step 5: Score and triage. Sort your portfolio into three buckets: (A) licensing-grade (velocity above 5.0, broad assignee spread), (B) hold and monitor (velocity 2–5, moderate spread), (C) consider abandonment (velocity below 1.0, narrow or zero external citations). Bucket C is where most founders waste $5K–$15K per year in maintenance fees on patents nobody cares about.
Why Patent Citation Analysis Matters More in 2026 Than Ever Before
Patent citation analysis has become the default valuation input for IP-heavy M&A deals in 2026. Acquirers and investors use citation metrics as a quick filter before committing to full technical due diligence — and if your citation numbers do not clear the bar, the deal stalls before it starts.
Three trends driving this shift:
AI-generated patent filings are flooding the system. USPTO applications rose 8.3% year-over-year in 2025, driven partly by AI-assisted drafting tools. More patents means more noise. Citation analysis cuts through the volume to identify signal — the patents the market actually references.
Intangibles now represent 90% of S&P 500 market capitalisation. Ocean Tomo’s 2025 study confirmed the number. When intangible assets dominate enterprise value, investors need quantitative tools to assess which intangibles are real. Citation metrics provide that quantitative layer for patents.
Companies with patents are 10.2x more likely to secure early-stage funding. But not all patents are equal. A 5-patent portfolio with high citation velocity can outperform a 50-patent portfolio with zero market traction. Citation analysis proves which scenario you are in.
What Should You Do After Running Your Citation Analysis?
Running the analysis is step one. Acting on it is where the value multiplies. Here is the decision tree:
If you have 3+ licensing-grade patents: You have a monetisable portfolio. Model the licensing revenue using the patent licensing revenue model framework and approach potential licensees among your citing assignees — they have already acknowledged your technology in their own filings.
If your portfolio is concentrated in hold-and-monitor territory: You need strategic filing to strengthen the cluster. Patent clustering around your highest-velocity patents can shift 2–3 assets from bucket B to bucket A within 18–24 months.
If most of your portfolio falls in bucket C: Stop paying maintenance fees on dead patents and redirect the budget toward filing in areas where your forward citation data shows market traction. Beyond Elevation runs this triage as part of every IP portfolio assessment — the 30-minute version above gives you the direction, but a full audit prices the licensing opportunity and builds the outreach strategy.
FAQ
How many forward citations does a valuable patent typically have?
A licensing-grade patent typically accumulates 25+ forward citations within 8 years of grant. The median for all US utility patents is 5–7 over the patent’s lifetime. Patents in AI and semiconductor technology tend to accumulate citations faster — averaging 8–12 within the first 5 years — while pharmaceutical patents accumulate slower but with higher per-citation licensing value.
Can I run patent citation analysis for free?
Yes. Google Patents and Lens.org both display forward and backward citation data at no cost. For portfolios under 20 patents, manual analysis on these platforms takes 30–60 minutes. For larger portfolios, tools like PatSnap, Derwent Innovation, or PatentSight offer batch analysis — but they cost $15K–$80K annually.
What is the difference between examiner citations and applicant citations?
Examiner citations are prior art references added by the patent examiner during prosecution — they indicate the examiner found your patent relevant to another application’s patentability assessment. Applicant citations are references added by the filing party. Examiner citations carry more weight in citation analysis because they represent an independent, third-party assessment of relevance.
Does patent citation analysis work for design patents?
Design patents accumulate far fewer citations than utility patents because the citation network for ornamental design claims is thinner. Citation analysis is most valuable for utility patents in technology-heavy sectors: software, AI, semiconductors, biotech, and clean energy.
How does patent citation analysis affect IP valuation during M&A?
Forward citation metrics directly influence the income approach to patent valuation — the most common method in M&A. Higher citation counts correlate with broader claim relevance, more potential licensees, and stronger negotiating position. In Beyond Elevation’s portfolio assessments, patents with citation velocity above 5.0 consistently value at 3–5x the floor estimate from the income approach alone.