IP Docketing Software: 9 Critical Tests Before You Buy

Read any three buying guides for IP docketing software and you will get the same list: automation, deadline calculation, reminders, document storage, reporting. Not one of them tells you which deadline rules the engine has to get right. That is the whole question. A docket that reminds you reliably about the wrong date is worse than a spreadsheet, because it is trusted. So here are nine tests built from the statutory windows themselves. Run them against a demo instance and you will learn more in an afternoon than a feature matrix will tell you in a month.

What “Automated Deadline Calculation” Has to Mean in IP Docketing Software

Here is the problem with how this category is sold. “Deadline calculation” is written as if it were one feature, like full-text search. It is not. It is a rules engine, and a trademark portfolio asks it to model at least four different response clocks, three different renewal base dates, one conditional five-year dependency that lives at WIPO, and a set of windows that open and close rather than falling due on an anniversary.

Patent docketing is, in this one narrow respect, easier. Maintenance fees in the United States fall at predictable intervals from issue, and the arithmetic rarely surprises anyone. Trademarks are where generic engines quietly go wrong, because trademark deadlines are mostly windows, several of them are conditional on facts the software cannot see, and the base date changes by jurisdiction.

What the feature list says	What you actually have to test
“Automated deadline calculation”	Does it open the Section 8 window at the fifth anniversary, or does it drop a single date at the sixth?
“Renewal tracking”	Does it compute the renewal base date from registration, filing, or the international registration date, per jurisdiction?
“Office action alerts”	Does it hold all four response clocks separately, including the one with no extension?
“Global coverage”	Does it carry the Madrid five-year dependency as a live date with a cancellation consequence?
“Audit trail”	Can it show who entered a date, when, from which document, and what the value was before?
“Easy data migration”	Can it export every date field with its source and its history, not just a matter list?

Everything below is written so you can hand it to a vendor. None of it requires access to your real data: each test is a fact pattern you can type into a sandbox in under a minute. If you want the companion piece for the platforms themselves, we maintain separate notes on trademark docketing software and on patent docketing software, and a cross-type view in patent and trademark docketing software.

Test 1 – Does It Open the Section 8 Window at Year Five?

This is the single most revealing test, and it is the one the published buying guides get loosest about. The USPTO states the requirement plainly: a Declaration of Continued Use under Section 8 is filed “between the fifth and sixth years after the registration date”, with a six-month grace period afterwards for an additional fee. See the USPTO’s own page on keeping your registration alive and 15 U.S.C. 1058.

So this is a twelve-month window with a six-month tail, not a due date. A docket that stores one date at the sixth anniversary has thrown away the first twelve months of the period in which you were allowed to file. In practice that is how firms end up filing in the grace period and paying the surcharge on registrations that were never actually late.

The test: create a registration with a registration date of 12 March 2021 and ask the software what it is showing you.

• Pass: a window opening 12 March 2026 and closing 12 March 2027, plus a grace deadline of 12 September 2027 flagged as surcharge-bearing.
• Partial: a single date of 12 March 2027 with the grace period derived correctly on request.
• Fail: one date, no window, no grace, or a date computed from the application filing date rather than the registration date.

Ask a second question while you are in there: what happens at the next cycle? The second Section 8 is not due at year eleven or twelve. It is filed together with the Section 9 renewal between the ninth and tenth years, and every ten years after that. An engine that models Section 8 as a repeating six-year event will put that filing in the wrong decade.

Test 2 – Does It Treat Section 9 Renewal as a Window, Not an Anniversary?

Section 9 renewal runs on ten-year cycles, and the USPTO describes the filing period as “between the ninth and 10th years after the registration date”, then “every 10 years after that (between the 19th and 20th years, 29th and 30th years, etc.)”. Equivalently: within the one-year period before the end of each ten-year term, with a six-month grace period after it. The statute is 15 U.S.C. 1059.

The common modelling error is to store the tenth anniversary as the deadline. That is not wrong about the outer edge, but it hides the fact that the filing window has already been open for a year, and it usually means the combined Section 8 and Section 9 filing is surfaced as two unrelated tasks instead of one.

The test: using the same 12 March 2021 registration, ask for the renewal dates.

• Pass: a combined Section 8 and Section 9 window from 12 March 2030 to 12 March 2031, grace to 12 September 2031, and the next cycle at 2040-2041.
• Fail: a renewal date of 12 March 2031 only, with the Section 8 obligation at that cycle missing entirely.

If you want the long-form version of these cycles, including the arithmetic for leap years and the rollover rules when a deadline falls on a weekend or federal holiday, that is in our notes on trademark renewal deadlines and trademark deadline calculation.

Test 3 – Does It Know Section 15 Is Conditional, Not Calendar-Driven?

A Section 15 Declaration of Incontestability is optional, and it is the first test of whether a vendor understands the difference between a deadline and an eligibility condition. Eligibility requires at least five consecutive years of continuous use in commerce after registration, with no adverse decision and no pending proceeding. See 15 U.S.C. 1065.

Because the five-year use period and the Section 8 window overlap, Section 15 is usually filed together with the Section 8 declaration. But it is not due then, and it is not due ever. There is no penalty for never filing it; you simply do not get the benefit.

The test: ask the software to show you Section 15 on a registration where use has been continuous, then on one where the mark went unused for eight months in year three.

• Pass: Section 15 appears as an optional, condition-gated opportunity tied to the Section 8 window, and the second registration is flagged for review rather than auto-scheduled.
• Acceptable: it is modelled as an optional task requiring human confirmation of the use facts.
• Fail: it is auto-scheduled as a hard deadline at the five-year anniversary regardless of use, or it is absent from the rule set entirely.

A platform that auto-schedules Section 15 as a statutory deadline is telling you something useful: its rule engine cannot represent conditionality. That matters well beyond Section 15, because the Statement of Use, excusable non-use, and post-registration audit responses all turn on facts no date arithmetic can reach. We cover two of those in statement of use deadline and the post-registration audit.

Test 4 – Does It Model Section 71 Separately for Madrid Registrations?

If any of your United States registrations arrived through the Madrid Protocol as a Section 66(a) extension of protection, they do not maintain under Section 8 and they do not renew under Section 9. They maintain under Section 71, and the renewal of the underlying right happens at WIPO on the international registration, not at the USPTO.

The windows look familiar, which is exactly why this gets collapsed. Per the USPTO: a Section 71 affidavit is filed “between the fifth and sixth years after the U.S. registration date”, then “between the ninth and 10th years”, then every ten years, each with a six-month grace period. Same shape as Section 8. Different statute, different form, and no Section 9 filing at all.

Registration basis	Maintenance filing	Renewal filing	Base date
Section 1(a) or 1(b) use-based	Section 8, years 5-6 then 9-10 and each decade	Section 9, years 9-10 and each decade	U.S. registration date
Section 44(e) foreign registration	Section 8, same windows	Section 9, same windows	U.S. registration date
Section 66(a) Madrid extension of protection	Section 71, years 5-6 then 9-10 and each decade	None at the USPTO — the international registration is renewed at WIPO	U.S. registration date for Section 71; international registration date at WIPO

The test: add a Section 66(a) registration and see what the docket schedules. A platform that produces a Section 9 task for it has merged the two regimes, which means someone will eventually file the wrong form or, worse, treat the WIPO renewal as covered because the United States row looks green. Our piece on trademark docketing integration goes into how the USPTO and WIPO records have to be reconciled rather than merged.

Test 5 – Does It Hold All Four Office-Action Clocks?

This is the test most worth your time, because it is where the 2022 Trademark Modernization Act left behind a genuinely messy rule set, and because a single wrong assumption here abandons an application. There is no one trademark office-action deadline. There are four, and they differ in length, in whether an extension exists, and in what the extension costs.

For pre-registration office actions on applications filed under Sections 1 or 44, the USPTO states you must respond within three months of the issue date, with the option to request one three-month extension for a fee, currently $125. The extension request has to arrive on or before the original three-month deadline, and the total available is six months from the issue date. The rule came from the Trademark Modernization Act final rule at 86 FR 64300 (17 November 2021) and took effect for applications on 3 December 2022. Sources: the USPTO response time period page and the delay-of-effective-date rule that set that date.

Section 66(a) Madrid applications are the exception, and it is a trap because the deadline is longer rather than shorter. The USPTO is explicit: “Madrid applicants must respond within six months from issue date, with no option to extend the deadline.” A docket that applies the three-plus-three pattern universally will show these as due far too early, which is survivable, and will also offer an extension that does not exist, which is not.

Post-registration office actions are the third clock, and this is where stale guidance is still circulating. The same 2021 final rule did extend the three-plus-three pattern to post-registration office actions, but those provisions were stayed before they ever took effect, and in July 2024 the USPTO proposed withdrawing them outright (89 FR 58660, RIN 0651-AD81). They have never been in force. The response period for a post-registration office action remains six months from the issue date, or the end of the statutory filing period if that is later.

The fourth clock belongs to expungement and reexamination proceedings, which the TMA created. An office action in one of those runs three months from the issue date, extendable once by a single month for a fee. One month, not three. Expungement may be requested between three and ten years after the registration date; reexamination within the first five years.

Office action type	Response period	Extension available	Total
Pre-registration, Section 1 or 44 application	3 months from issue date	Yes — one 3-month extension, $125 fee, requested on or before the deadline	6 months
Pre-registration, Section 66(a) Madrid application	6 months from issue date	No	6 months
Post-registration (e.g. on a Section 8, 9 or 71 filing)	6 months from issue date, or the end of the statutory filing period if later	Not applicable — the 3+3 regime was withdrawn before taking effect	6 months or the window end
Expungement or reexamination proceeding	3 months from issue date	Yes — one 1-month extension for a fee	4 months

The test: raise one office action of each type in the sandbox and read back the computed due dates and the extension options offered. Four distinct answers is a pass. One answer applied four times is a fail, and it is the kind of fail that does not announce itself until an application has gone abandoned. The expungement side is covered in detail in trademark expungement docket dates.

Test 6 – Does It Carry the Madrid Five-Year Dependency as a Live Date?

An international registration under the Madrid Protocol stays tied to the basic application or registration it came from for five years from the date of the international registration. If the basic mark ceases to have effect in that period – refused, withdrawn, cancelled, not renewed, wholly or partly – the international registration is cancelled to the same extent. That is Article 6(3) of the Madrid Protocol, and it is the mechanism usually called central attack.

After five years the international registration becomes independent and the exposure closes. So this is a date with a real consequence attached and an expiry of its own, which is precisely what a docket is for. Most platforms do not carry it, because nothing is due on it. Nothing being due is not the same as nothing being at risk.

There is a second date hiding behind it. Where an international registration is cancelled at the request of the office of origin, Article 9quinquies lets the holder refile the same mark nationally or regionally in the territories the registration covered, keeping the international registration date and any priority – but only if the application is filed within three months from the date the international registration was cancelled. See WIPO’s guide to the Madrid System.

• Pass: a dependency-expiry date five years from the international registration date, linked to the basic mark’s record, plus a transformation task template with a three-month clock.
• Partial: the dependency date exists but is not linked to the basic mark, so a refusal of the basic application triggers nothing.
• Fail: no dependency date at all, which is the common case.

The three-month transformation window is unforgiving, and it starts on a date you learn about by notice rather than by calendar. If your docket does not have a template waiting for it, you are relying on whoever opens the mail. More on the full set of Madrid dates in Madrid Protocol deadlines.

Test 7 – Does It Get the Renewal Base Date Right per Jurisdiction?

This one is short to test and expensive to get wrong. Renewal terms are ten years almost everywhere, which lulls people into thinking the arithmetic transfers. It does not, because the date the ten years run from changes.

Right	Renewal term	Counted from	Filing window
U.S. registration (Section 8 and 9)	10 years	Registration date	Years 9-10, then each decade; 6-month grace
U.S. Section 66(a) extension (Section 71)	10 years	U.S. registration date	Years 5-6 and 9-10, then each decade; 6-month grace
European Union trade mark	10 years	Application filing date	6 months before expiry; further 6 months late renewal after expiry
International registration at WIPO	10 years	International registration date	Renewed every 10 years from the IR date

The European Union trade mark is the outlier that catches United States-centric engines. An EUTM is registered for ten years from the date of filing of the application, not from registration, and because examination and any opposition can take well over a year, those two dates can sit a long way apart. The renewal request and fee are due in the six months before expiry, with a further six-month late period afterwards. See the EUIPO renewals FAQ.

The test: enter an EUTM with a filing date of 4 June 2019 and a registration date of 20 November 2020, and ask for the renewal date. If the software says 2030, it is counting from registration and every EUTM in your portfolio is docketed roughly a year and a half late. The correct answer is a term expiring 4 June 2029. We go deeper on the base-date problem across jurisdictions in our international renewal notes, and on how these errors cluster in trademark docketing errors.

Test 8 – Can It Prove Who Entered a Date, and What It Was Before?

Docketing is one of the most common roots of professional liability claims in IP practice, and the reason is structural rather than careless: the harm surfaces years after the entry that caused it. By the time a registration lapses, the person who typed the date has often moved on, and the question is not whether someone was negligent but whether anyone can reconstruct what happened.

“Audit trail” appears on nearly every IP docketing software feature list. The useful version records four things, and most of the thin implementations record one.

• Who entered or changed the date, as a named user rather than a service account.
• When, to the timestamp, and in which time zone.
• From what source – the office correspondence or document the date was read off, ideally linked.
• What the previous value was, because an overwritten date with no history is the one you will need.

The test: enter a Section 8 window, change it, then change it back, and export the history for that one field. If the export shows three entries with users, timestamps and prior values, the audit trail is real. If it shows “modified” with today’s date, it is a last-touched flag wearing an audit trail’s name.

Ask one more question: can the audit history leave the system? An audit trail you cannot export is not evidence you can use, and it disappears the day you change platforms. This is also the practical core of a docket review – see trademark docket audit for what a reconciliation against the register actually involves, and litigation docketing for why contested-proceeding dates need the same treatment.

Test 9 – What Does It Cost You to Get Data In, and Out?

Two of the most-asked questions about IP docketing software are really the same question asked from opposite ends: how much manual entry does onboarding take, and what can you take with you if you leave. Both are answered by the same artefact – a full field-level export – and both are routinely discovered after signature.

On the way in, the honest version of “automated capture” is specific about scope. Direct feeds from the USPTO can populate status and many statutory dates for United States matters. They cannot populate your internal dates, your client reporting rules, your non-United States matters in offices without a feed, or any date that depends on facts only you hold. Ask which fields the feed fills and which a human must type.

• Which offices does it pull from directly, and which are manual?
• On a migration, which fields arrive populated and which are recomputed by the new engine’s rules?
• If the new engine recomputes, are the old and new dates reconciled and the differences reported, or is one silently overwritten?
• Can you export every date field, its history, and its linked document, in a format you can read without the vendor?
• Is the export self-service, or a chargeable professional-services request?

That third question is the one that matters most and gets asked least. A migration where the new platform recalculates every deadline is not a data transfer, it is a re-derivation – and if its rules differ from the old ones even slightly, you have changed dates across the whole portfolio without a review. Any difference between the two should be surfaced as a reconciliation report, not resolved by whichever system wrote last. Our trademark docket migration checklist is built around exactly that step, and the cost side is broken down in trademark docketing software cost.

How to Run These Tests in a One-Hour Vendor Demo

You do not need a pilot or a trial dataset. Every test above is a fact pattern you can read out loud, and the whole battery fits inside a normal demo call if you send it ahead and ask for a sandbox rather than a slide deck. Most IP docketing software vendors will open one if you ask two days in advance; a refusal is itself a useful answer.

1. Send the nine tests to the vendor two days before the call, and ask that a sandbox be open and shared.
2. Create one registration dated 12 March 2021 and read back the Section 8 window, the grace date, and the year 9-10 combined filing.
3. Add a Section 66(a) registration and confirm Section 71 appears and Section 9 does not.
4. Raise all four office-action types and read back four different due dates and extension options.
5. Add an international registration and ask to see the five-year dependency date and a transformation template.
6. Enter the EUTM with filing and registration dates eighteen months apart, and check which one drives renewal.
7. Change one date twice, then export that field’s history and read it without help.
8. Ask for a full field-level export of the sandbox, and note whether that is a click or a quote.
9. Write down every answer. The answers are the comparison table the published guides do not have.

One caution on scoring. A platform that fails Test 3 or Test 6 is not necessarily wrong for you – conditional dates and dependency exposure can be handled by procedure if you know they are missing. A platform that fails Test 1, Test 5 or Test 7 is computing statutory dates incorrectly, and no amount of procedure around it is safe. Separate the “we do this outside the system” answers from the “the system says a wrong date” answers, because only one of those is a workaround.

Where the Market Sits, and What That Means for the Tests

A brief and deliberately narrow orientation, based only on what each vendor publishes about itself. We are not ranking these and we are not quoting prices, because docketing pricing is almost always quoted per matter or per user against a specific portfolio and any number printed here would be fiction.

The category sorts into roughly three shapes, and the shape predicts which of the nine tests is worth pressing hardest.

Shape	Examples (per their own descriptions)	Press hardest on
Enterprise IP management suites, docketing as one module	Anaqua, which also offers Pattsy Wave after acquiring it in June 2020 and describes it as purpose-built for patent and trademark docketing teams; Clarivate’s FoundationIP, described as IP lifecycle management with docketing calendars and annuities	Tests 5 and 7 — breadth sometimes comes with one generalised response clock and a registration-based renewal assumption
Docketing-first platforms	DocketTrak, which describes auto-docketing of statutory deadlines from the PTO and customisable reporting; CPI	Tests 3 and 6 — strong statutory arithmetic, conditional and WIPO-side dates worth confirming
Trademark-first, built on direct office feeds	Alt Legal, which describes pulling data directly from the USPTO, WIPO and EUIPO to generate docketing calendars	Tests 8 and 9 — confirm what the feed does not fill and what the export contains

Treat that table as a starting hypothesis, not a finding. Every one of these vendors updates its rule engine, and the only evidence that counts is what the sandbox computes on the day you ask. We keep longer per-platform notes in Pattsy Wave alternatives, Alt Legal alternatives, and DocketTrak pricing, each written the same way: characteristics the vendor publishes, tested against the statutory rules rather than against each other.

And there is a fourth option the software comparisons structurally cannot include, which is not buying software at all. If the rule engine is the thing you actually need, a managed service puts the rules and the liability for applying them with a team rather than a licence. The trade-offs are laid out in outsource trademark docketing, and the service itself is trademark docketing.

How PerspireIP Can Help

If the answer to “who owns the rule engine” turns out to be “nobody here”, that is an argument for buying the outcome rather than the tool. PerspireIP runs trademark docketing as a managed service: we hold the statutory rules, we enter and verify the dates, and the audit trail is part of the deliverable rather than a feature tier. You keep the register of truth; we keep it correct.

Already have a platform and just want to know whether it is computing the right dates? That is a scoped exercise, not a migration. Talk to our team about running these nine tests against your live docket and reconciling the output against the USPTO and WIPO records.

Frequently Asked Questions