USPTO show-cause orders for zero-term filings
The USPTO is mailing orders to show cause in pending applications that would issue with no patent term. What the notice says, and what it leaves open.
How to compare intellectual property management systems
Most IPMS evaluations are decided by demo quality, which is a poor predictor of how a system behaves in year three. These are the dimensions that actually separate them, in the order they tend to cause trouble.
Why there is no vendor table here
A feature-and-price grid for other vendors is stale within weeks, cannot be verified from public marketing material, and ages into a page that misinforms. What follows are criteria you can apply to any system on your shortlist, including this one. Ask each vendor the same questions and compare the answers you get, not the answers someone else recorded a year ago.
Does the data model respect the differences between rights?
This is the first thing to test, because everything downstream inherits it. A trademark carries goods and services under the Nice Classification, a design carries a Locarno class and a representation, a patent carries claims, priority and family relationships. A system that models all of these as one generic "case" with custom fields bolted on will handle the common path and fail at the edges — and the edges are where portfolios get lost. Ask to see the same portfolio entered as all four right types.
Where do the deadline rules come from, and are they versioned?
Ask who maintains the jurisdiction rules, how quickly a change in law reaches the system, and — the question most often skipped — whether the system can still show which version of a rule produced a date calculated three years ago. A system that silently recalculates historic dates when a rule changes has destroyed the evidence that the original date was correct at the time. That distinction is invisible in a demo and decisive in a dispute.
Can it separate a fact from a calculation from a verification?
These are three different things and many systems collapse them into one status field. What the office actually published, what the system derived from it, and what a human confirmed are separately meaningful, and only the combination is defensible. Test it directly: ask the system to show, for one deadline, the raw source value, the rule applied, and who signed off — on one screen, without a support request.
How does it reconcile against the official registers?
Offices change status continuously, so the real question is not whether a system imports data but what it does when the register disagrees with the record. Ask what happens to a manually entered value when an import contradicts it, whether conflicts are surfaced or silently overwritten, and how a user sees when a record was last confirmed against the source. A portfolio nobody has reconciled is a portfolio whose status is an assumption.
What does the deployment and data-residency picture look like?
Establish where data is held, under which jurisdiction, and whether that is configurable per client or per matter. For teams with confidentiality obligations this is often decided before features are discussed, and discovering the constraint late invalidates an evaluation. Ask specifically whether an on-premise or single-tenant option exists, and what it costs in practice — not whether it is theoretically available.
What does migration actually involve?
Ask for the shape of the import, not a reassurance. The question is whether history migrates or only current state: if only current dates arrive, the new system starts with no provenance for anything that happened before, and every legacy record becomes an assertion. Ask who does the mapping, how exceptions are reported, and what a reconciliation between old and new looks like. The exit side of the same question is covered under independent intellectual property management.
Which questions can be deferred, and which cannot?
Reporting layouts, interface preferences and most integrations can be changed later and should not decide a purchase. The data model, the rule provenance and the export format cannot be changed later, because they determine what evidence exists about everything entered in the meantime. Weight the evaluation accordingly: the dimensions that are cheap to fix later are the ones demos are best at showing.
Applying this to Iprelia
These criteria are the ones Iprelia is built around, so it is fair to hold it to them: rights modelled on a common core with their own specialist fields, a jurisdiction-aware deadline engine with versioned legal rules, raw source and rule version and verification kept as separate fields, dual-control confirmation, and data residency in the United States, the European Union or Switzerland with an on-premise option. It is in testing now — a good moment to ask the harder questions above.
See the platform →What is an IPMS? →Read our intellectual property law analysis →
General information about evaluating intellectual property management software. Not legal advice.