Patent Citation Analysis for Design-Around: Why I Follow the Citation Chain

In my 20 years of practice, citation analysis has been less an academic exercise in "measuring patent influence" and more a practical search tool. I use it in two situations: when I coach companies through design-around work, and when I help them with patent mining — identifying what in their own R&D is worth protecting. In both, the goal is the same: find the documents that are genuinely close to a specific technology.

What are patent citations, in practical terms?

Every patent document sits in a web of references.

• Backward citations are the earlier documents a patent refers to. Some are supplied by the applicant; many are cited by the examiner during search and examination because they were judged relevant to the claims.
• Forward citations are later documents that cite the patent. They show who came afterwards and built on, worked around, or was examined against the same technology.

Patent offices publish these links. The European Patent Office's search reports, for example, also mark how relevant each cited document is judged to be. The practical point is that these links were made by people — often examiners — who read the documents and decided they belong together. That is a different kind of evidence from a keyword match.

Why do keyword-only design-arounds run into trouble?

Many of my Chinese clients come to me to design around patents held by leading overseas companies in their industry. I have seen the same pattern many times.

The team searches the relevant documents using keywords, studies what it finds, and designs around those patents. It builds a sample. Then, at the freedom-to-operate stage, the FTO analysis turns up other patents with a high infringement risk — patents that the keyword search never surfaced.

At that point the cost is real. The design has been settled, the sample has been made, and the team has to go back and rework a solution it thought was finished.

The reason is not carelessness. Keyword searching depends on the words the searcher chooses and the words the authors happened to use. A leading company's portfolio around a product is written by different drafters, over different years, often in different languages and then translated. The same technical idea can appear under very different wording. A keyword search finds the documents that share your vocabulary and misses the ones that do not.

The patents a keyword search misses do not go away — they reappear in the FTO review, after the sample is built. Find them before you design, not after.

How does following the citation chain help?

When I start from the key patent we need to design around and follow its citation links — backward to what the examiner considered relevant, forward to what later documents connect to it — I reach documents that a keyword search would not have found, because the link does not depend on wording.

Based on many experiences of finding key documents this way, the patent documents found through the citation chain are, from my experience, clearly more accurate, and the results are much better than relying on keyword searching alone.

That does not mean keywords are useless. They are a good way to find the starting points. But for design-around work, where the cost of a missed document is a redesign, I do not consider the search finished until I have walked the citation chain around the key patents.

For design-around, keywords find the starting points; the citation chain finds the neighbours. From my experience, the documents it surfaces are clearly more accurate than keyword results alone.

How does citation analysis fit into patent mining?

Patent mining is the work of finding what in a company's R&D can and should be protected. Citation links help here in the same way: starting from the closest known patents, the chain shows what the surrounding prior art looks like. That gives the R&D team and the patent professional a clearer picture of where the company's own contribution really lies, and where it does not.

How should you combine citation analysis with FTO?

A practical sequence for a design-around project looks like this:

1. Identify the key patents that stand between your product and the market.
2. Walk the citation chain around each key patent, backward and forward, and read the documents you find — not only their titles.
3. Design around with that full picture, so that avoiding one patent does not lead you into another you have not seen.
4. Run an FTO analysis on the final design before committing to production, mapping your product's features against the claims of patents in force in your target markets.

Citation analysis does not replace step 4. An FTO conclusion depends on how claims are interpreted in each jurisdiction, including questions such as equivalents, and that is a legal judgement. This is not legal advice; consult a qualified patent attorney for your situation.

What citation analysis does is move the discovery earlier. In my experience, that is where most of its value lies: finding the difficult documents while the design can still change.