Preexamination is not better keyword search. It is reading a body of prior art, mapping what it teaches, and adapting the application while you still control the response — before you file.
“Preexamination” here is not the wait to first action, a pre-docketing notice, or a post-filing AI search. It is the practitioner’s synthetic examination of claim scope against the current constellation of art, performed before the application is filed.
Landscape first: context-based campaigns, cononym clusters, functional filtering, and limiting invention context until the claim has somewhere to live. Then adapt the disclosure, claims, and IDS while correction is still cheap.
The instrument reads and charts at scale. The practitioner turns the dial, checks the figures, and decides what to file.
On the Office side after filing, the same analytical work is examination on demand. Two roles stay distinct.
Coming next (practitioner track): disclosures and claims from the landscape · IDS and § 1.56 · aggregating examination findings · claim and disclosure harmonization · portfolio white space · and more. Then a composite series on how the Office can use the same instruments.
The search variable is context. Call a contextual synonym a cononym. A group of those neighbors is a cononym cluster.
When you limit invention context to escape anticipatory art, you are discovering what limitations the claims must carry. The landscape and the claim scope may be the same variable read from two directions.
Figures for cononym clusters and rescoping are in Part II.
Not a filing obligation. A practice choice — and an early answer to whether the space supports a patent worth having.
Want the next Preexamination post when it drops?