IP portfolio management is mostly a calendar problem with a decision attached.
The calendar part is unglamorous and unforgiving. Rights end permanently when a deadline passes, without notice and without appeal.
The decision part is where the money is. Renewing everything is expensive and easy; releasing what no longer earns its cost is cheaper and requires someone to have asked the question.
Doing it across five different rights types, each with its own cadence, is what makes it a discipline rather than an administrative task.
Why it fails when nobody owns it
The most common structural failure is diffusion, not incompetence.
| Right | Who typically holds it | Who watches the deadline |
|---|---|---|
| Patents | Engineering or legal | Outside counsel, sometimes |
| Trademarks | Marketing or brand | Often nobody |
| Copyright | Whoever created the work | Nobody |
| Designs | Product or industrial design | Often nobody |
| Trade secrets | Operations | Nobody |
Each function knows about its own rights and none has the whole view. A trademark renewal sits with marketing while patent fees sit with legal, and neither list is complete.
Copyright and trade secrets have no deadlines at all, which is precisely why they get no attention — and why trade secret loss, the only irreversible failure in the table, happens quietly.
One named owner with one calendar fixes most of this. The role does not need seniority or legal training; it needs a list, a diary and the authority to ask whether a right is still used before an invoice is paid.
The rights, and how they behave
| Patents (US) | Patents (foreign) | Trademarks | Copyright | Trade secrets | |
|---|---|---|---|---|---|
| Renewal | 3 fees after grant | Annual annuities | Every 10 years | None | None |
| Extra filings | — | — | §8 declaration years 5–6 | — | — |
| Term | 20 yrs from filing | 20 yrs from filing | Indefinite | Life + 70 | While secret |
| Fails when | Fee missed | Annuity missed | Renewal or use lapses | Term ends | Disclosed |
| Recoverable | Sometimes by petition | Sometimes | Refiling possible | No | Never |
| Ongoing duty | Pay | Pay | Use and police the mark | — | Maintain secrecy |
Three different failure modes. Patents fail by non-payment. Trademarks fail by non-use or non-policing as well as non-payment. Trade secrets fail by disclosure, and that failure is permanent and unfixable.
Only trade secret loss is irreversible. Which means the measures protecting it deserve attention out of proportion to their cost.
Unregistered rights are still rights
Portfolios routinely omit everything that was never filed anywhere, which is often the majority of what a company actually relies on.
| Unregistered right | How it arises | How it is lost |
|---|---|---|
| Trade secrets | Reasonable secrecy measures | Disclosure — permanently |
| Unregistered copyright | Automatically on creation | Term expiry only |
| Common-law trademark rights | Use in commerce | Non-use, or a registered mark elsewhere |
| Trade dress | Distinctiveness through use | Loss of distinctiveness |
| Know-how and process data | Accumulated practice | Departure of the people holding it |
Know-how held only in people's heads is the most fragile asset most companies own. It does not appear on any register, generates no invoice, and leaves when they do.
Document it or accept losing it. Process parameters, failure modes and supplier tolerances are all licensable assets when written down and worth nothing when they are not.
Include unregistered rights in the inventory with a status field rather than a deadline. They need review, not payment.
The inventory
Everything starts here, and most inventories are incomplete.
| Field | Why it is needed |
|---|---|
| Right type and jurisdiction | Determines the cadence |
| Registration / application number | Identity |
| Owner entity, exact legal name | Chain of title |
| Filing and grant dates | Term calculation |
| Next deadline and cost | The operative field |
| Status | Live, pending, lapsed |
| Encumbrances | Security interests, existing licences |
| Related family members | Continuations, foreign counterparts |
| Is it actually used? | The field decisions depend on |
The last field is the one that is usually missing. An inventory listing two hundred patents with no indication of which cover shipping products cannot support a renewal decision, so the decision defaults to paying.
Owner entity must be the exact legal name. Portfolios acquired through mergers routinely list rights under entities that no longer exist, and that is a chain-of-title problem waiting for a transaction to expose it.
Chain of title as maintenance
Ownership records decay quietly, and the decay is invisible until a transaction exposes it.
| Decay source | Effect |
|---|---|
| Corporate name changes not recorded | Chain appears broken |
| Acquisitions with unrecorded schedules | Gap between recorded owners |
| Contractors and departed founders never assigned | Possible co-ownership |
| Dissolved subsidiaries | No authorised signatory remains |
Fix gaps while the people involved are still reachable. An assignment from a current employee takes a signature; the same document from someone who left six years ago may take months or prove impossible.
Audit title annually alongside the renewal review. It costs an hour per tranche and it is the difference between a portfolio that can be sold and one that merely exists.
One calendar
Deadlines across rights types and jurisdictions have to sit in one place.
| Deadline type | Frequency |
|---|---|
| US patent maintenance fees | 3 per patent |
| Foreign annuities | Annual, per country, per patent |
| Trademark §8 declaration | Years 5–6 |
| Trademark renewal | Every 10 years |
| Office action responses | As they arrive |
| Continuation filing windows | Before parent grants |
| PCT national phase entry | 30 months from priority |
| PTE application | 60 days from FDA approval |
Foreign annuities dominate by volume. Four patents in five countries is twenty deadlines a year, each in a different jurisdiction with a different agent.
The continuation window is the one with no reminder. Nothing prompts you; the parent simply grants and the option closes permanently.
Diarise decisions, not just payments. A deadline entry saying "pay $8,280" produces a payment. One saying "decide: keep, sell or lapse — $8,280 due" produces a decision.
The annual review
| Question | Applies to |
|---|---|
| Does this cover a product we sell? | Patents, designs |
| Is the mark in use in the classes registered? | Trademarks |
| Would a competitor want this? | All |
| Is anyone infringing? | Patents, trademarks |
| Does remaining term justify escalating cost? | Patents |
| Are secrecy measures actually operating? | Trade secrets |
| Is this licensed, and does the royalty cover the fees? | All |
| Could it be sold rather than released? | Patents, trademarks, designs |
Trademark non-use is a distinct risk. A mark registered in classes where it is not used is vulnerable to cancellation, and the §8 declaration requires a statement of use. Renewing a mark you no longer use is spending money to keep something that may not survive a challenge.
Trade secret review is behavioural, not administrative. Confirming that confidentiality agreements are signed, access is limited, and marking protocols are followed is the whole of the protection.
Where the budget goes
| Portfolio age | Typical split |
|---|---|
| Under 5 years | Mostly new filings and prosecution |
| 5–10 years | Roughly balanced |
| Over 10 years | Mostly renewals |
Renewals accumulate quietly. New filings get discussed and approved individually; renewals arrive as an invoice, and the total grows every year until someone examines it.
Escalation compounds the effect. US maintenance fees rise from $2,150 to $8,280 for a large entity; foreign annuities rise every year. A portfolio held flat in size gets more expensive annually with no decision having been made.
Which is the argument for the annual review. It is the only mechanism that converts an escalating default into a set of choices.
What the data says about normal attrition
| Ipiry Patent Survival Curve v1.0 | Rate |
|---|---|
| Survive the 3.5-year fee (2022 cohort) | 85.8% |
| Survive the 7.5-year fee (2018 cohort) | 64.6% |
| Reach full term (2014 cohort) | 41.4% |
| Abandoned before full term | 58.6% |
| Lost at first fee | 14.2% |
| Lost at second fee | 21.2 points |
| Lost at third fee | 23.2 points |
Computed from 27,273,654 USPTO maintenance fee records covering 8,262,336 US utility patents — see the patent survival curve.
Releasing most of a patent portfolio is what the market does. A company renewing everything is an outlier, and usually an unexamined one.
Attrition accelerating at each stage is the pattern to expect. Rising fees against a shrinking window make later renewals progressively harder to justify.
Selling rather than lapsing
Both end the cost. Only one pays.
| Route | Cost ends | Revenue |
|---|---|---|
| Renew | No | — |
| Sell | Yes | Yes |
| License | No — fees continue | Yes, ongoing |
| Lapse | Yes | None |
Timing is the constraint. A patent sold before its fee deadline is worth something; the same patent after the grace period closes is worth nothing.
Which means the annual review must run well ahead of deadlines, not alongside them. A sale takes months. See patent portfolio pruning for the review process and patents for sale for who buys.
Tooling
| Portfolio size | Approach |
|---|---|
| Under ~20 rights | Spreadsheet plus shared calendar |
| 20–100 | Spreadsheet plus annuity service for foreign renewals |
| Over 100 | Docketing system, annuity service, named owner |
Below a hundred rights the discipline matters more than the software. A maintained spreadsheet with every deadline beats an unused system.
Keep your own calendar even when a service pays the fees. Outsourcing payment does not outsource the consequence, and a duplicate diary is cheap.
Name one person. Diffusion across engineering, marketing and legal is the most common structural failure — nobody has the whole view, and deadlines fall between functions.
IP portfolio management: the checklist
- Build one inventory covering every right type, including trade secrets and unregistered rights.
- Add a "commercially used" field. Without it, renewal decisions default to paying.
- Verify owner entity names against the registers and fix chain-of-title gaps while records still exist.
- Put every deadline in one calendar — patents, annuities, trademark declarations and renewals, continuation windows, national phase entry.
- Phrase deadline entries as decisions, not payments.
- Run one annual review of the whole portfolio, well ahead of individual deadlines so selling remains possible.
- Check trademark use in every registered class, since non-use is a cancellation risk regardless of renewal.
- Audit trade secret measures behaviourally — agreements signed, access limited, marking followed. Loss here is permanent.
- Track renewal spend as a share of total IP budget. Above ten years old it should be the majority, and rising without decisions is a warning.
- Sell rather than lapse wherever value remains, and start early enough that a sale can actually close before the deadline.