Think about the mistakes available to you in a project like this. Pay too much and money comes back. Miss a date and the date moves. Draft something badly and it gets redrafted. File the wrong thing and it gets refiled. Nearly every error in business is recoverable, which is why people treat decisions here with a certain ordinary confidence. There is exactly one exception, and nobody points it out until afterwards. Telling somebody cannot be undone. It takes two seconds. It costs nothing. Its effect is permanent. No other act available to you has that ratio — and because it is the cheapest thing in the project, it gets the least thought. Then the first correction, because almost everybody who says it is wrong: we have not told anybody. Did you show it to a supplier to have it made? Photograph it? Demonstrate it? Put it in a deck? Describe it at an event? Post the prototype? People hear told as meaning a careful explanation given to a stranger, and most disclosure looks nothing like that. And the part that catches even careful people: showing is heavier than telling. A sentence conveys what you chose to say. A photograph, a demonstration or a working sample conveys everything visible in it — including the parts you never thought about and would never have described. So the first deliverable in this whole subject is not a filing. It is a disclosure log: one row for every person outside your core group who has seen or heard anything, with who, when, exactly what they saw, in what form, and on what terms. It takes an afternoon, almost nobody has one, and — this is the point — it cannot be reconstructed later, because disclosure happens in ordinary moments nobody records. A conversation at a trade fair. A photograph sent to a machinist. A slide shown to four investors, two of whom you can no longer name. A post that is still online. None of them felt like an event, and things that did not feel like events are not stored in anybody’s memory as events. From there the rule writes itself: file first, talk later, wherever the sequence allows — and where it genuinely does not, make the disclosure a decision rather than a reflex, with terms in place and a line in the log. Two more. An agreement does not un-show the thing; it creates obligations, which is worth having and is not the protection people assume it is. And the inventor list is not a courtesy list — adding somebody senior is not generosity, leaving out a junior is not economy, and both are inaccuracies in a document meant to be accurate. Finally, the boundary, stated before you order: whether this can be patented, what the prior art is, and how anything should be drafted is the work of a registered patent agent, not ours. We build the half that must exist before that person can start — and that is the half almost everybody arrives without.
What this guide covers
Start by noticing something about every other decision in a project like this, because it explains the confidence people bring to this one.
Every one of those is a setback rather than an ending, which is why people treat decisions in a project like this with an ordinary, reasonable confidence. That confidence is correct almost everywhere.
There is exactly one exception, and nobody points it out until afterwards. Telling somebody cannot be undone.
There is no version of this in which you un-tell a person something. No correction, no refiling, no second attempt. Whatever has left your hands has left them permanently, and no amount of money, effort or regret retrieves it.
That is not a dramatic way of putting it. It is simply the one place in this project where the usual rules about recoverable mistakes do not apply.
Look at the proportions, because they are what make this so dangerous in practice.
| Act | Effort | Cost to reverse |
|---|---|---|
| Signing a bad contract | Weeks of work | Difficult, possible |
| Ordering wrong tooling | Days, and money | Expensive, possible |
| Missing a deadline | None, it just happens | Usually recoverable |
| Showing somebody the thing | Two seconds | Impossible |
No other act available to you sits in that bottom row.
Place Order — Free · pay after work
And here is the cruel mechanism, which is about attention rather than carelessness.
We allocate deliberation in proportion to effort and visible cost. A contract gets a meeting. Tooling gets three quotes. Showing somebody a prototype gets nothing at all, because it costs nothing and takes no time, and nothing in the moment signals that a decision is being made.
So the single irreversible act in the project is also the only one that is never deliberated. That is not a failure of judgement — it is what happens when consequence and effort point in opposite directions.
This is the sentence we hear most often at the start, and it is almost always sincerely meant and almost always wrong.
Not because people are concealing anything. Because the word told is doing work that nobody notices it doing.
Ask the same person a different set of questions and the picture changes completely.
In our experience several of those turn out to be yes, and the person answering is genuinely surprised by their own answers.
People hear told as meaning something formal: a careful explanation, given deliberately, to a stranger, with the intention of conveying the thing.
Almost no disclosure looks like that. It looks like ordinary commercial life — getting a quote, showing a customer, explaining why your thing is better, answering a question at a stall. Nobody experiences those as acts of disclosure, which is precisely why they happen without any thought at all.
Place Order — Free · pay after work
Now the part that catches even careful people, including people who have deliberately decided to be cautious about what they say.
A sentence conveys what you chose to put in it. A photograph, a demonstration or a sample conveys everything visible in it.
When you describe something you are selecting, naturally and continuously, and you hold things back without even noticing you are doing it. When you show something you have stopped selecting entirely.
Specifically, an image gives away the things you never consciously thought about — which are frequently the things that matter.
The arrangement you arrived at after four attempts and now take for granted. The proportion that turned out to be the one that worked. The small detail you stopped seeing a year ago because it had become obvious to you. Obvious-to-you is exactly the category of information you would never have described, and it is fully present in any photograph.
So when you are listing what you have given away, pictures and samples belong at the top of the list rather than the bottom. People instinctively rank them lower because sending a photograph feels like saying less than talking for ten minutes.
Heavier again, and worth separating because they are so routine that nobody counts them.
A physical sample conveys everything a photograph does plus everything that can be learned by handling it. A demonstration adds the behaviour — how it performs, what it does under use, what problem it solves and how well.
Both are completely normal commercial acts. Neither feels like a disclosure. Both belong on the log with a clear note of exactly what was handed over or shown, because “they saw it work” and “I described it” are very different rows.
So the first deliverable in this entire subject is not a filing and it is not a conversation with a patent agent. It is a document, and it is one almost nobody has.
A disclosure log: one row for every person or organisation outside your core group who has seen or heard anything at all.
It takes an afternoon. It requires no expertise. And it is the single thing that makes every later conversation — with an agent, with an investor, with anybody — shorter and more useful.
Place Order — Free · pay after work
| Column | What goes in it | Why |
|---|---|---|
| Who | Person and organisation, by name | “A supplier” is not a row |
| When | A date, or the closest you can get | Order matters more than exactness |
| What they saw | Described, not summarised | “The idea” tells nobody anything |
| In what form | Spoken, drawing, photo, sample, demo | The form decides the weight |
| On what terms | Agreement, verbal, or nothing | “Nothing” is a valid and useful entry |
| Still public? | For anything posted or published | The one column you can go and check |
This is the reason the afternoon is worth spending now rather than when somebody asks for it.
Disclosure happens in ordinary moments that nobody records, and moments that did not feel like events are not stored in memory as events. You can recall a meeting because it was in a calendar. You cannot recall a two-minute conversation beside a stall at an exhibition eighteen months ago, and nothing exists anywhere that would remind you.
Six months from now this list is irretrievable. Today it is an afternoon.
Place Order — Free · pay after work
For anything that might exist in a durable form, check rather than recall — and this is the one part of the exercise where you can actually be certain.
Scroll back through your own posts and your business accounts. Search your sent messages for the product name and for the words you use about it. Look in your sent email for attachments. Open old decks and see what the slides actually contained rather than what you remember putting in them.
People frequently remember a post and forget that it carried a photograph, or remember the photograph and forget the comment underneath explaining how the thing worked.
Write the uncomfortable rows. A log that quietly omits the two disclosures you wish had not happened is not a cautious log — it is a misleading one, and its only effect is to produce confident advice built on a false picture.
Nobody here is going to be disappointed in you, and the agent will not be either. Both of us have seen this many times, and the disclosures almost always happened for sound commercial reasons.
Where you are unsure whether something counts, write it in and mark it uncertain. An uncertain row is a question somebody can answer. A missing row is invisible.
The commonest row on every log, and the least avoidable, because you cannot have a thing made without somebody seeing how it works.
We are not going to tell you that was a mistake. It usually is not one. The point is that it happened, that it belongs on the log, and that what terms were in place at the time matters. Where nothing was in place — which is the usual answer — record that plainly.
What you can change is how you handle the next one, and the next one is usually only a few weeks away.
Before any supplier meeting, ask the question nobody asks: what does this person actually need in order to do the thing I want?
The answer is almost always less than the instinct. A quote for one component does not require the assembly. A moulding price does not require the mechanism. A finishing quote does not require knowing what the object is for. A properly written supply arrangement helps here too, because it forces somebody to state what is actually being supplied.
Sending the whole thing is the default because it is easier, and easier is how most disclosure happens.
Where you genuinely must show something, put terms in place first. Signed before, not circulated afterwards — an agreement produced after the meeting covers a conversation that already happened.
This is ordinary and people accept it without offence; a supplier who has quoted for other people’s products has signed these before. The awkwardness is almost entirely in your own head, and it lasts about ten seconds.
Worth being precise about, because this is where comfortable misunderstanding does the most damage.
A confidentiality agreement creates obligations between you and that person. It is worth having, it changes what they may properly do, and it gives you something to point at. What it does not do is un-show the thing.
Those are two different kinds of protection and people treat them as one. The agreement is a promise about behaviour. It is not a reversal of the fact that somebody now knows.
Which produces a specific pattern we see often: somebody signs an agreement, feels protected, and from that point shares far more freely than they would have otherwise.
The agreement made the disclosure safer in one sense and did nothing about it in another — and the relaxation it caused typically increases the total disclosure by more than the agreement protects.
So treat the agreement as a floor rather than a licence. The question what does this person actually need to see does not stop applying because a document has been signed.
The rule that falls out of all of this, stated plainly because it is easy to say and hard to follow.
File first, talk later — wherever the sequence allows it.
Note the qualification, which is doing real work. We are not telling you to go silent for a year, and anybody who tells you that has not run a business.
Place Order — Free · pay after work
Most people do the reverse and not out of carelessness. Talking is how a project moves.
You need a manufacturer before you can have a sample. You need a sample before a customer takes you seriously. You need a customer before anybody funds the tooling. You need all three before you can afford a patent agent. The sequence that protects you and the sequence that gets anything built point in opposite directions, and pretending otherwise is useless advice.
So the realistic version is not silence. It is deliberateness.
Four questions before any disclosure, which together take about a minute.
Then write the row. The whole discipline is one minute of thought and one line of record, repeated.
A category of its own, because it is unavoidable, it is high-volume, and the people on the other side see a great many things.
Investors need to understand what they are backing, so some disclosure is genuinely necessary. What is not necessary is doing it without a record. Every conversation is a row: who, when, what was shown, in what form.
The forgotten ones are always the people who passed. A rejection is filed in memory as a rejection rather than as a disclosure, which is why the deck shown to four investors is usually remembered as two.
The practical approach is to decide in advance how much detail each stage of a conversation actually requires, rather than deciding in the room while somebody is interested.
An early conversation generally needs the problem, the market and the fact that you have solved it. It rarely needs how. The how belongs to a later stage, by which time terms are normally in place as a matter of course.
Deciding this beforehand is the whole technique, because in the room, with somebody engaged and asking good questions, nobody has ever successfully decided to say less.
Place Order — Free · pay after work
Academic and professional sharing is a large category and it almost never feels like disclosure to the person doing it, because it is the normal and admirable behaviour of that world.
A paper. A conference presentation. A thesis sitting in a department. A demonstration at a departmental event. A poster. A talk to a student group. All of them are rows.
Whether any of it matters is entirely the agent’s question and we will not guess at it. Whether it is on the log is ours, and this is a category where people under-report heavily, because the sharing was professionally correct and therefore does not feel like something to confess.
The one part of the whole exercise that can be checked rather than remembered, so check it properly.
Your own accounts and the business accounts. Old posts, including ones you have forgotten the existence of. Anything a partner, a supplier or a customer posted about you. Competition entries and incubator pages. An old website version. A directory listing with a product description on it.
Note what is still visible and what is no longer visible, and keep a copy of each for the file — because a page coming down later does not change the fact that it was up, and you will want the record of what it actually said.
Alongside the log, the other thing that is cheap now and unobtainable later: a dated, contemporaneous record of what you made and when.
Not a polished document. The actual working trail, kept rather than replaced. The entire value of this material is that it is dated and that it was made at the time, which is precisely the property that cannot be added to it afterwards.
Place Order — Free · pay after work
The instinct, when somebody asks for your records, is to produce something presentable. Resist it completely.
A cleaned, retyped, nicely organised version of a contemporaneous record has stopped being a contemporaneous record.
The scruffiness is doing the work. Crossings-out, dead ends, a page with the date written in the corner and coffee on it — all of that is evidence of when and how. A neat summary is evidence of somebody having written a neat summary.
If you want something presentable, make it alongside rather than instead: keep the originals exactly as they are, in their original files, and build a clean index that points to them.
That gives you both — something a person can navigate, and underneath it material that has not been touched. It is also exactly how we keep these files, and keeping the set properly afterwards matters more here than in most subjects, because the gap between making something and needing to prove when you made it can be many years.
A hard line rather than a preference, and it applies to you as much as to us.
We will not produce any document carrying a date it did not earn, for any reason, at any price, and a request to do so ends the engagement. We will not reconstruct a “record” of something that was not recorded at the time and present it as though it had been.
Where a record does not exist, the honest position is that it does not exist, and that is a workable position. A fabricated one is not merely weaker — it contaminates everything genuine sitting beside it.
The last of the three things that must exist before anybody files, and the one where otherwise sensible people make a mess for entirely human reasons.
The inventor list is a factual statement, not a political one.
It records who actually did the inventing. Not who runs the company, not who paid for it, not who has been loyal, and not who would be pleased. Those are all real considerations in a business and none of them belongs in this particular document.
Extremely common, almost always well intentioned, and an inaccuracy.
A founder who funded the work. A head of department. Somebody whose support made the project possible. All of that is real and none of it is inventing. Adding them is not generosity — it is a false statement in a document whose entire function is to be true.
Where somebody deserves recognition, there are other ways to recognise them, and the arrangements between founders or the structure of the company itself are the places where contribution and reward actually get settled.
Place Order — Free · pay after work
The mirror image, equally common, and worse because it tends to be the one that comes back.
A junior engineer who solved the problem. A freelancer who was paid once. A student on a placement. An employee who has since left on poor terms. Leaving them off is not economy and it is not tidiness; it is the same inaccuracy pointing the other way, and unlike the first one it creates somebody with a reason to object.
We will ask who did what, write down what you tell us, and say so plainly where the answer appears uncomfortable rather than smoothing it over. That is the only useful thing we can do with this.
Deal with it now, while they are reachable and still well disposed towards you. In two years they may be neither, and the request will be much larger.
Record what they contributed and roughly when. Find whatever written terms existed. Where nothing was ever written, a short confirmation obtained today costs nothing and is often simply impossible later — not because anybody refuses, but because people move, change numbers and stop answering strangers.
For every person on the inventor list and every organisation on the disclosure log, the same question: what was agreed, in writing, at the time?
Where genuinely nothing exists, that is an answer rather than a failure — and it is a far more useful answer than a vague one.
Two recurring cases worth naming because they are handled casually and surface awkwardly.
The freelancer who did a defined piece of work, was paid, and left — frequently with nothing in writing beyond an invoice. The student or intern who produced something substantial in a few weeks, often as part of a course, and whose institution may have had its own arrangements.
Both are ordinary. Both need a row on the inventor list if they contributed and a row on the log if they saw things. And in both cases a short written confirmation obtained now is a two-minute favour that becomes an impossibility surprisingly quickly.
Place Order — Free · pay after work
Put together, the whole intake is four things, and none of them requires any technical knowledge.
Plus anything already public, which we would much rather hear from you than find ourselves in week three.
It is worth being concrete about why this is worth paying for separately rather than hoping it emerges.
A patent agent works from what you bring. They are not going to telephone your old machinist, go through a phone for dated photographs, or work out whether the person who left in 2024 contributed. If you arrive without that, the conversation becomes a series of meetings in which they ask and you go away and find out.
Arriving with a complete picture turns three conversations into one, and it turns the agent’s time into technical work rather than fact-gathering — which is what you are actually paying them for.
Which is how most people arrive, and it is worth saying clearly: that is normal.
The worst response is to avoid writing it down, or to write a hopeful version. The second worst is to assume it is hopeless and do nothing — that is a conclusion for somebody qualified to reach, not for you and certainly not for us.
This matters enough to say directly, because the embarrassment is real and it stops people from being complete.
Almost everybody has disclosed more than they should have, and almost always for sound commercial reasons — they needed a sample made, they needed a customer, they needed somebody to believe in it. You behaved like somebody trying to build a business, which is what you were doing.
We have seen this many times and so has every agent. Nobody is keeping score, and the only thing that actually harms you now is an incomplete account.
Place Order — Free · pay after work
The practical argument, for anybody not persuaded by the reassurance.
An agent advising on a partial picture gives advice that is confidently wrong, which is worse than no advice because you act on it. A strategy built on the belief that nothing has been disclosed, where several things have, is not a cautious strategy — it is a wrong one, and the error is invisible until it is expensive.
The uncomfortable rows are precisely the rows that change the advice. Those are the valuable ones.
Worth knowing they exist, because people arrive having decided on a patent when something else is the actual fit — and because the alternatives have their own clocks.
| If what matters is… | Then the exercise is… |
|---|---|
| How it works, technically | A patent — the agent’s question |
| How the article looks | A design registration |
| The name you sell it under | A mark, starting with a search |
| Something you made and fixed | A copyright registration |
| Something not yet made, to discuss | An agreement, which is a contract rather than a register |
Sometimes a competent agent will tell you that nothing here fits, or that what fits is not worth what it costs for a business your size, or that too much is already public for the route you wanted.
That is them doing their job properly, and it is worth more than a filing sold to somebody who should not have had one. If you ever get that answer, the documentary work still has value: the log, the inventor list and the dated trail are the materials for every other conversation you will have about this — with a partner, a buyer, an investor or a licensee.
The documentary half does not change at all. The log matters exactly as much, the inventor list matters exactly as much, and dated records matter more if anything, because software has no physical prototype to date for you.
What form of protection is available or appropriate is a specialised question and squarely the agent’s. Meanwhile the commercial paperwork still has to exist, and in software it is usually the weakest part of the file: development agreements, employment terms, and a clear account of who built which part.
A closing point that is easy to lose among the more dramatic material above. Most of what protects a small business in practice is not a register at all. It is the set of ordinary documents saying who did what, who may use what, and who agreed to what.
Those exist whether or not anything is ever filed, they cost very little, and they are the part we can actually do. Where a recognition or scheme is also in play, the same underlying facts feed it, which is another reason to establish them once and properly.
Place Order — Free · pay after work
None of those is a technical failure, and not one of them would have been prevented by hiring a better agent. They all happen before an agent is involved, which is exactly why this half needs somebody looking after it.
Place Order — Free · pay after work
Worth explaining rather than asserting, because people reasonably assume anybody in this field can advise on all of it.
Patent work is a qualified, regulated role. Drafting an application is a specialised technical and legal skill, and the quality of the drafting determines what the thing is actually worth. It is not administrative work and it is not something to buy on price.
We are not that, we will not pretend to be, and we would be wary of anybody at our end who blurs the line. Their fee is engaged and paid by you directly, and we have no interest in it either way — which is the only honest basis on which to be telling you to go to one.
Our part for this groundwork — building the disclosure log by interview, checking what is actually public and keeping copies, recording the inventor list as fact, auditing what terms existed with every person and supplier, drafting and chasing the confirmations that are still obtainable, collecting the dated evidence in original condition, and assembling one file an agent can read in twenty minutes — is ₹14,999, with a turnaround of 30 – 90 days.
What moves that range is reconstruction, not writing. Somebody who comes to us before showing anybody anything has a log with two rows and current evidence. Somebody three years in, with several suppliers, two former colleagues and an old post still online, is at the long end — and most of that goes on establishing facts rather than recording them. You will know which end you are at in the first conversation, and we will tell you if we think the honest answer is that you should be talking to an agent first.
These stay outside, each for a reason:
And the framing. You are not buying an opinion about your invention — we are not qualified to have one, and we will not let anybody at our end imply otherwise. You are buying an accurate account of what has already happened, assembled while it can still be assembled, so that the person who is qualified starts from facts instead of from your best recollection.
Court work is for your advocate, whose fee is engaged and paid by you directly; we do not quote, collect or share it.
The companion question, and the one this page keeps pointing at. This route protects how a thing works. The way it looks sits in a different register, which covers only the features that did not have to be that way — see design registration for that half, and bring the showing question back here first.
Completing the map on one product. This route covers the mechanism; name and appearance sit in their own registers; and the number on the carton is neither a right nor a protection, however much the word “registration” suggests otherwise.
We build the disclosure log by interview rather than by form, because the rows that matter only come out when somebody asks the follow-up question. We check what is actually public, record the inventor list as fact rather than diplomacy, audit what terms existed with every person and supplier, chase the confirmations still obtainable, and collect your dated evidence in original condition — so a registered patent agent starts from a complete picture. We do not assess patentability and we do not backdate.
Why this page is written the way it is, and what it refuses to contain
No provision numbers. No test for what can or cannot be patented. No grace periods, no fee figures, no timelines, no office procedure, no country comparison, and no guidance at all on drafting. On a page about patent filing those are large omissions, so they need a reason rather than an apology.
Every one of those is the work of a registered patent agent or attorney, and that is a qualified, regulated role. Patentability turns on technical and legal judgement applied to a specific invention; drafting determines what a filing is actually worth; a search means nothing without somebody competent reading it. A general web page cannot do any of that, and a page that gestured at it would produce exactly the harm it pretended to prevent — a reader deciding, on the strength of something written for nobody in particular, that they are fine or that they are finished. The single most dangerous figure we could have published is a period of time, because a reader would act on it.
What is left when all of that is removed is not small, and it has no owner. It is the factual groundwork: who has seen what, when, in what form and on what terms; who actually invented it; and what dated evidence exists. None of it requires technical qualification, all of it has to exist before an agent can usefully begin, and almost nobody arrives holding any of it. An agent will not interview your former supplier or recover dated photographs from a phone, and it is unreasonable to expect them to.
Why the page is organised around one asymmetry. Because that asymmetry explains the behaviour rather than merely describing it. Disclosure is the only irreversible act in the project and also the cheapest, and attention is allocated in proportion to effort rather than consequence — so the one act that cannot be undone is reliably the one nobody thinks about. That is not a character flaw in anybody and it is not fixed by being told to be careful. It is fixed by a log, four questions and one line of record.
What is deliberately absent. No statement about whether anything is protectable. No period, deadline or grace figure of any kind. No fee other than our own. No drafting guidance, no claim language, no prior-art method. No assessment of anybody’s invention. For your own matter the operative sources are a registered patent agent or attorney on your technical facts; your own disclosure log and dated records for what actually happened; and an advocate instructed on your facts if anything has become contentious.
Place Order — Free · pay after work
Two doors, both free. Clients search a factual directory of enrolled advocates. Advocates apply to be listed on it — no fee, no commission, nothing paid in either direction.
Search Bar Council enrolled advocates by what your matter is about, by court, or by city. Searching and sending a request are both free.
Enrolled advocates anywhere in India can apply to be listed. Your entry is published only after we verify your enrolment number with your State Bar Council.
Would rather not pick anybody? Tell us the matter instead — one form goes to every advocate at once, and the first to take it up calls you.
This directory carries no ratings, no reviews, no rankings and no fees — only the factual particulars the Bar Council of India permits, published at each advocate's own request. Browse the network · Terms for Advocates