Patent prosecution software is docketing software with additional features attached, and the order matters when choosing one.
Docketing is the core job. Calculating and tracking every deadline in a matter, and surfacing it before it passes.
Everything else is secondary, because a missed deadline is the one failure that cannot be undone. An application abandoned for a missed response, or a patent lapsed for an unpaid fee, is gone.
Which sets the evaluation order. Test date calculation first, on matters you already know. Everything else can be assessed afterwards.
What the software has to track
| Deadline | Frequency | Consequence of missing |
|---|---|---|
| Office action response | Per action | Abandonment at 6 months |
| Extension windows | Per action | Fees, and lost PTA |
| US maintenance fees | 3 per patent | Permanent expiry |
| Foreign annuities | Annual, per country | Permanent expiry, per country |
| Continuation window | Before the parent issues | Family closed permanently |
| PCT national phase entry | 30 months from priority | Foreign rights lost |
| Issue fee | 3 months, no extension | Abandonment |
| PTE application | 60 days from FDA approval | Extension forfeited |
| Priority claim deadlines | 12 months from provisional | Priority lost |
Foreign annuities dominate by volume. Four patents in five countries is twenty deadlines a year, each in a different jurisdiction with its own rules.
The continuation window is the one with no external prompt. Nothing arrives in the post; the parent simply grants and the option closes. See patent application process.
Testing date calculation
This is the evaluation that matters and it takes an afternoon.
| Test | Why |
|---|---|
| Run matters you already know | Discrepancies surface immediately |
| Include a continuation chain | Term calculation from the earliest parent |
| Include a patent with PTA | Adjustment applied correctly |
| Include a terminal disclaimer | Caps the term |
| Include foreign family members | Annuity schedules differ |
| Include an entity status change | Fee amounts differ |
| Check a six-month grace period | The second silent deadline |
Continuation chains are where systems most often fail. A patent filed in 2019 as a continuation of a 2013 application has a 2013 term start, and software that uses the later date is wrong by six years.
Terminal disclaimers are the second. They override the arithmetic entirely and must be flagged manually in many systems.
USPTO integration
| Data | Value of automatic updating |
|---|---|
| Application status | High — reduces manual entry |
| Office actions and correspondence | High |
| Issue and grant dates | High |
| Maintenance fee payment records | High |
| Assignment records | Moderate |
| PTA calculations | High |
Manual entry is where most docketing errors originate. Every field a person types is a field that can be typed wrong, and automatic population removes that class of failure.
Ask what updates automatically and how often. A system that pulls status weekly behaves differently from one requiring someone to check.
Foreign offices vary widely. US integration is generally good across products; EPO and national office coverage is where they differ.
Getting deadlines to a person
A deadline stored in a system nobody opens is not a reminder.
| Delivery | Effectiveness |
|---|---|
| Calendar integration | Highest — reaches where people work |
| Email to a named individual | High |
| Email to a distribution list | Low — nobody's responsibility |
| In-app dashboard only | Low |
| Weekly digest | Moderate |
| Escalation on non-acknowledgement | High |
Ask how a deadline reaches a person, not how it is stored. That question predicts whether the system prevents a missed date.
Escalation matters more than reminders. A deadline that has been surfaced three times without acknowledgement should reach someone else.
Double-entry is the common failure in practice. A deadline recorded in the platform but not in the calendar the fee-earner actually uses depends on someone remembering to look.
Where docketing errors originate
| Source | Share of risk |
|---|---|
| Manual data entry | Highest |
| Wrong priority date entered | High |
| Continuation chain not linked | High |
| Deadline not delivered to a person | High |
| Entity status stale | Moderate |
| Jurisdiction rules misapplied | Moderate |
| System outage at a deadline | Low |
Manual entry dominates, which is why USPTO integration is worth more than it appears on a feature list — every automatically populated field is one that cannot be mistyped.
Priority date errors propagate. One wrong date generates a whole schedule of wrong deadlines, and nothing flags it.
Beyond docketing
| Feature | Genuinely useful? |
|---|---|
| Deadline calculation and alerts | Core |
| USPTO data integration | High |
| Document management | High |
| Fee payment or annuity integration | High |
| Reporting to clients | Moderate |
| Drafting assistance | Useful for specifications, weak on claims |
| Office action response assistance | Emerging |
| Analytics on examiner behaviour | Moderate, and interesting |
Drafting assistance helps most where coverage is the task. Generating alternative embodiments, materials and parameter ranges expands what can later be claimed, and that is a volume exercise.
Claim drafting remains the weak point, because claim scope is a prediction about how an examiner and a competitor will behave. A claim reciting a result rather than a mechanism reads well and fails eligibility. See ai patent drafting tools.
Examiner analytics are genuinely interesting — allowance rates, interview receptiveness, typical rejection patterns — and they inform strategy rather than replacing it.
Integration with the rest of the practice
| Integration | Why it matters |
|---|---|
| Calendar | Deadlines must reach the person, not sit in a tab |
| Document management | Where correspondence and drafts live |
| Time and billing | Capture without double entry |
| Correspondence filed automatically | |
| Annuity service | Foreign renewals actually paid |
| Client reporting | Status without manual assembly |
Double entry is the failure that erodes any system. If a deadline has to be recorded twice, it eventually gets recorded once.
Annuity service integration is the one most specific to patents. A system that surfaces a foreign renewal without connecting to whoever pays it leaves a manual handoff at the riskiest point.
Scale thresholds
| Portfolio size | Approach |
|---|---|
| Under ~20 matters | Spreadsheet plus shared calendar |
| 20–50 | Spreadsheet plus annuity service for foreign renewals |
| 50–200 | Docketing system |
| 200+ | Docketing system plus annuity service, named owner |
Below fifty matters the discipline beats the software. A maintained spreadsheet with every deadline, checked weekly, prevents more missed dates than an unused system.
Foreign portfolios move the threshold down. Twenty US patents generate sixty deadlines across a decade; twenty patents in five countries generate hundreds.
Worked example: a switching decision
A practice with 340 matters across four jurisdictions evaluating two products.
| Requirement | Product A | Product B |
|---|---|---|
| Date calculation on 20 known matters | Matched all 20 | Differed on 3 |
| Continuation chain handling | Correct | Used the child's filing date |
| PTA applied | Yes | Yes |
| Terminal disclaimer flagged | Yes | Manual only |
| Foreign annuities, all four jurisdictions | Yes | US and EP only |
| USPTO auto-update | Weekly | Daily |
| Calendar integration | Yes | Add-on tier |
| Drafting assistance | Basic | Strong |
| Analytics | Basic | Strong |
The decision
Product A, despite weaker drafting and analytics.
The three discrepancies decided it. Product B calculated term from the continuation's own filing date rather than the earliest parent, which was wrong by several years on three matters and would have been wrong silently on more.
Foreign annuity coverage was the second factor. Two of the four jurisdictions unsupported means either a second system or a gap.
Calendar integration behind a paywall was the third. A deadline that does not reach the fee-earner's calendar depends on habit.
The test that produced this took one afternoon and twenty matters the team already knew.
Migration risk
| Task | Risk |
|---|---|
| Importing the existing docket | Highest single risk in adoption |
| Verifying every imported deadline | Essential |
| Re-creating alerts and recipients | Moderate |
| Historical matter linkage | Moderate |
| Training | Low |
| Running both in parallel | Recommended for one full cycle |
A silent mis-map is the danger. An import that assigns a wrong date creates a missed deadline with no warning, and nobody discovers it until the date passes.
Verify every imported deadline against the source, not against the old system. If the old system was wrong, the import faithfully reproduces the error.
Run both for at least one fee cycle. The duplication is tedious and cheaper than the alternative.
Reporting and client visibility
| Report | Used by |
|---|---|
| Upcoming deadlines by matter | Everyone |
| Portfolio status summary | Clients, management |
| Fee forecast by quarter | Budgeting |
| Prosecution history per matter | Attorneys |
| Filing and grant statistics | Management |
| Cost per matter | Finance |
Fee forecasting is the report that gets used most and specified least. Projecting three years of maintenance fees and annuities, per matter, at the correct entity rate, is what turns a docket into a budget.
Ask whether reports export cleanly. They end up in board papers and client updates, which exist outside the platform. See patent portfolio management.
What software does not fix
| Problem | Why software does not solve it |
|---|---|
| Nobody owns the docket | A system needs an operator |
| Deadlines not acted on | Alerts are not decisions |
| Wrong data entered at the start | Garbage in |
| Entity status not rechecked | A judgement, not a calculation |
| Whether to keep a patent | A decision, not a deadline |
Entity status is a judgement made at each payment. Growth past 500 employees or a licence to a large company ends small entity eligibility, and no system detects that automatically. See small entity status.
The renewal decision is not a deadline. Software tells you $8,280 is due; it does not tell you whether the patent is worth keeping, and 58.6% of US utility patents are abandoned because owners decide it is not. See the patent survival curve.
Patent prosecution software: the checklist
- Test date calculation first, on twenty matters the team already knows.
- Include a continuation chain in the test. Term must run from the earliest parent.
- Check terminal disclaimer handling, since it overrides the arithmetic.
- Confirm foreign annuity coverage for every jurisdiction you touch.
- Ask how a deadline reaches a person, not how it is stored.
- Insist on calendar integration, not a dashboard nobody opens.
- Check what updates automatically from the USPTO, and how often.
- Verify every imported deadline against the source during migration.
- Run old and new in parallel for a full cycle before decommissioning.
- Name one owner. Software without an operator is a list nobody reads.