An inventor with a workable idea and no experience of the system is, for a short period, an unusually easy customer. The vocabulary is unfamiliar, the deadlines are opaque, and the cost of getting it wrong is invisible until years later. That combination is why the first money spent on an invention is so often the worst-spent money in the whole project.
None of this argues for going it alone. Drafting a patent application without training produces documents that grant narrowly or not at all, and the search work that should precede drafting is a research discipline in its own right. The argument is narrower: that the person or firm taking your money should be able to answer five specific questions clearly, quickly and in writing, and that an evasive answer to any of them is worth more attention than a persuasive answer to all the others.
What follows is not a warning about the field. It is a checklist. Every question below has a right answer, a wrong answer, and a characteristic non-answer, and the non-answers are the ones worth learning to recognise.
Question One · Standing
Who Is Actually Permitted to File on Your Behalf?
Representation before a granting office is a regulated activity almost everywhere. The office maintains a register, entry to it requires passing an examination, and in most territories a technical or scientific qualification is a precondition of even sitting that examination. Anyone on that register can be looked up by name, and the lookup takes about ninety seconds.
This is the first question because it is the only one with a binary answer. Either the individual who will sign your application is on the register, or they are not. Consultants, marketers, prototype engineers and business advisers may all do useful work, and none of them can prosecute an application. That is not a criticism of them; it is a boundary in the rules. The problem arises only when the boundary goes unmentioned and an inventor assumes that everyone in the room is qualified to give legal advice about scope.
Ask for the registration number and the name of the register. Ask who will personally draft the claims, as distinct from who will manage the account. The distinction between the two roles, and the different routes into practice behind them, is worth understanding before the first invoice, and it is set out in more detail on our page about how representation before the granting office actually works.
What a good answer sounds like
- A name, a registration number, and an invitation to verify it yourself.
- A clear statement of who drafts, who reviews, and who signs.
- An acknowledgement of what the firm does not do in-house.
Most first approaches happen exactly like this: alone, on a phone, in a gap between other obligations. The asymmetry is the point. The inventor is making a decision once; the firm has made it thousands of times.
Which is why the questions should be written down before the conversation rather than improvised during it.
Question Two · Scope
What Exactly Is the Deliverable, in Writing?
"We will help you protect your idea" is not a deliverable. A deliverable is a named document, produced by a named date, in a named format. A prior-art search report listing the references found and the databases queried is a deliverable. A drafted specification with claims, sent for your review before filing, is a deliverable. A market assessment is a deliverable, provided the engagement letter says what data it draws on.
The reason to insist on this level of specificity is not pedantry. Invention support covers several distinct trades — searching, drafting, prosecuting, prototyping, engineering, and commercial presentation — and a single fee that covers "the process" makes it impossible to tell which of those you have bought. It also makes it impossible to tell when you have received them. Ask for the deliverables as a numbered list with dates attached, and ask what happens to the schedule if the granting office responds slowly, because it usually does.
Pay attention, too, to what the deliverable is not. A submission package sent to companies is not a licence. A design report is not a patent. A filing receipt is not a grant. Each of these confusions is easy to fall into when the paperwork arrives looking official, and each one has cost inventors years of misplaced confidence. Understanding the different types of patent application and the proceedings attached to them makes the distinctions much harder to blur.
A filing receipt is not a grant, a submission package is not a licence, and a design report is not protection. Most disappointment in this field starts as a category error.
The three confusions worth pre-empting
Question Three · Money
How Is the Fee Structured, and What Triggers the Next One?
Costs in this field arrive in stages rather than as a single figure, and the stages are genuinely hard to predict, because the largest variable — how the examiner responds — is outside everyone's control. An honest quote acknowledges that and separates the components: professional time, official fees paid to the granting office, drawing preparation, and the open-ended cost of prosecution once examination begins.
Three specific numbers are worth extracting before signing anything. First, the total payable to reach a filed application, itemised. Second, the expected cost of responding to a single objection from the examiner, since most applications receive at least one and many receive two or three. Third, the maintenance fees that fall due at intervals across a granted patent's twenty-year term, and who is responsible for diarising them. A patent lapses quietly when a renewal is missed, and the lapse is often discovered only when the holder tries to enforce it.
Be equally direct about how the firm itself is paid. A business that earns its revenue from up-front fees has a different set of incentives from one that earns it on the commercial success of the invention, and neither model is disqualifying — but you should know which one you are dealing with, and the answer should not require excavation. Anyone offering help with patenting invention ideas should be able to describe their own commercial model in two sentences without hesitating.
Three numbers to get in writing
- Total cost to reach a filed application, itemised by component.
- Expected cost of answering one examiner objection.
- Renewal fee schedule across the term, and who tracks the dates.
Question Four · Disclosure
What Happens to Your Description the Moment You Send It?
Novelty is assessed against everything already available to the public at the moment of filing. That principle is close to universal, and its practical consequence is that describing an invention to the wrong audience at the wrong time can destroy rights that did not need to be lost. Some jurisdictions offer a grace period of a year after a disclosure by the inventor; many offer none at all, and a single public description can foreclose protection in those territories on the day it appears.
So the question is procedural, not sentimental. Who inside the organisation will read your submission? Is the file held on a system with access controls? What written undertaking covers it, and how long does that undertaking last? Does the firm ask you to sign anything that limits its own liability for onward disclosure? Reputable operators answer these questions readily because they have already thought about them; the paperwork exists, and they will send it before you ask twice.
It is also worth knowing what a signed undertaking does and does not achieve. It creates a contractual obligation between two parties. It does not make a disclosure private in the eyes of the granting office if the terms are loose, and it is close to worthless against a party with nothing to lose. The circumstances in which such paperwork is genuinely load-bearing, and the circumstances in which filing first is the only real protection, are covered in our note on what a signed non-disclosure undertaking actually does.
Timing is the whole discipline. Rights are measured against a single instant — the filing date — and everything said publicly before that instant counts against you.
File first. Talk second. Almost every irrecoverable mistake in independent invention is those two steps in the wrong order.
Question Five · Evidence
Ask for Numbers, Not Anecdotes
Every organisation in this field can produce a success story, and the good ones are genuinely instructive: a specific problem, an ordinary person, and a long unglamorous stretch of documentation and negotiation between the idea and the shelf. Accounts such as the route a vibration-based pain relief product took to market are useful precisely because they show the intervening steps rather than the leap.
But an anecdote is a sample of one, and the number that matters is the denominator. How many clients were taken on in a given year? How many reached a filed application? How many secured a licence agreement or a product on sale? Rates in independent invention are low — that is a fact about the field rather than about any one firm, and a candid operator will say so plainly. What should worry you is not a modest conversion rate but an unwillingness to state one, or a set of testimonials with no numbers behind them at all.
Ask also for the failure case. What happens if the search comes back badly? Does the engagement stop, or does it roll on into services that no longer have a defensible purpose? The willingness to tell a client that an idea is anticipated, early and at a cost of a few hundred rather than several thousand, is the single most reliable indicator of a serious adviser. A broader survey of how ideas move from concept to product — including the stages where most stall — is set out in this guide to turning invention ideas into reality, and the pattern it describes matches what the numbers show.
Putting It Together
The Order of the Questions Matters as Much as the Questions
Ask about standing first, because it is verifiable and takes a minute. Ask about deliverables second, because the answer determines whether the fee discussion is even meaningful. Ask about money third, once you know what is being bought. Ask about disclosure fourth, before you have sent anything of substance. Ask for evidence last, when you have enough context to judge whether the figures you are given are plausible.
Run in that order, the sequence takes one conversation and a follow-up email. It will not tell you whether your invention is any good — nothing at this stage can — but it will reliably separate operators who work to a defined scope from those who work to a fee. Firms that put their process, pricing and outcomes in writing tend to answer all five without prompting, which is itself the shortest version of the test.
The five questions are cheap. The situations they prevent are not. An inventor who insists on written answers before the first payment gives up nothing but a few days, and buys the one thing the process never gives back later: the ability to change course while changing course is still free.
Ask early · Ask in writing · Then decide