You can evaluate your own invention in a weekend, for free, across five dimensions: market pull, manufacturing cost, patentability, competitive density, and personal fit. Score each from 0 to 5, apply the weights below, and you have a number out of 100 that tells you whether the next paid step is justified. Under 45, stop or redesign. From 45 to 64, one specific weakness is dragging it, and the work is to fix that one thing. Above 65, the concept has earned a professional prior art search and a prototype budget.
This is the assessment that paid “invention evaluation” reports are supposed to perform. Doing it yourself produces a better answer, because the inputs are things only you can gather and the scoring is only useful if you are honest.
Why Self-Assessment Beats a Paid Evaluation
Reports sold to inventors at $500 to $1,200 arrive in a few weeks and read the same regardless of the invention. The structural reason is that nobody can evaluate a product concept without knowing its manufacturing cost, and manufacturing cost cannot be known without a design. A report produced before the design exists is a template with your product name inserted.
The inputs that matter are cheap to collect and nobody can collect them for you: what the workaround costs the person with the problem, what similar products sell for on the shelf, how many patents already sit in the classification code, and how many parts your mechanism needs. Two days of your own work beats a purchased opinion built on none of that.
Dimension 1: Market Pull (weight 25)
Not market size. Pull. A large category with no unmet need is worse than a small one where people are actively taping things together.
Answer these before scoring:
- What do people currently do instead, and what does that cost them in time or money?
- How often does the problem occur? Weekly, monthly, once a decade?
- Is the person with the problem the person with the budget?
- What is the shelf price of the closest existing product?
- Have you talked to twenty potential buyers, in person, about the actual task?
| Score | Condition |
|---|---|
| 0 | You have not spoken to anyone with the problem |
| 1 | People agree it would be nice. Nobody does anything about it today |
| 2 | An observable workaround exists but it costs almost nothing |
| 3 | Workaround costs real money or real time, occurs monthly or more |
| 4 | Buyers named a price they would pay, unprompted, more than once |
| 5 | Buyers asked when they could have one |
The last twenty conversations are the whole dimension. Reviews, forum posts, and warranty complaint categories add supporting evidence. Guessing does not.
Dimension 2: Manufacturing Cost Against Price (weight 20)
Work backward from the shelf. For a licensed consumer product sold through retail, manufactured cost has to land near 20% to 25% of retail price for the margin structure to survive the distributor and the retailer. A $19.99 product needs to be made for roughly $4 to $5.
Estimate your cost roughly: count the parts, estimate material volume, assume $12,000 to $60,000 for production steel tooling on injection-molded parts, add assembly labor at a few cents per operation, add packaging and freight. The exact number comes later. What you need now is whether you are in the right order of magnitude.
| Score | Condition |
|---|---|
| 0 | Estimated cost exceeds the category’s shelf price |
| 1 | Cost is 60% or more of achievable retail |
| 2 | Cost is 40% to 60% of retail |
| 3 | Cost is 25% to 40% of retail, with a path to reduce |
| 4 | Cost is near 20% to 25% of retail at modest volume |
| 5 | Cost is under 20% of retail, or the product commands a premium the category supports |
Part count is the lever with the most travel. Every part carries piece price, tooling, inventory, and assembly labor, and consolidating two into one commonly saves 30% to 60% of the combined cost at volume. The moves that get you there are in design for manufacturability. A concept scoring 1 here is often a 3 after one consolidation pass, which is why this dimension is worth re-scoring rather than abandoning.
Dimension 3: Patentability (weight 20)
Two questions: is anything on file that discloses your mechanism, and is your difference more than a size, material, or quantity change?
Run the free search first. Work from classification codes rather than keywords, because keyword-only searching misses references written in different vocabulary, which is most of them. The USPTO search tools cover U.S. and international art, and the technique is in how to search Google Patents.
| Score | Condition |
|---|---|
| 0 | A live patent claims your exact mechanism |
| 1 | An expired patent discloses it. Free to build, not patentable |
| 2 | Close art exists and your difference is size, material, or quantity |
| 3 | Close art exists and your difference is structural but predictable |
| 4 | Your difference produces an interaction the art would not predict |
| 5 | The classification code is thin and nothing close appears in it |
Score 1 deserves attention rather than dismissal. An expired patent that discloses everything means no patent for you and no infringement risk either. That can still be a business built on cost, distribution, or brand. It is not a patent business, and knowing which one you are in changes every decision after it. The distinction between novelty and non-obviousness, and why the second one ends more applications than the first, is covered in what prior art means for a patent application.
Dimension 4: Competitive Density (weight 15)
Count what is already there. Search the classification code and count active patents from the last ten years. Then count products on the shelf and on the major marketplaces that address the same task.
| Score | Condition |
|---|---|
| 0 | Dominated by two or three large brands with deep patent portfolios |
| 1 | Dozens of recent patents in the code, most from operating companies |
| 2 | Crowded shelf, commodity pricing, little differentiation |
| 3 | Several competitors, none owning the specific failure mode you attack |
| 4 | Few direct competitors, category growing |
| 5 | No product addresses this task, and you can explain why it is buildable now |
A 5 here needs a hard second look. If nothing exists, either the problem is not worth solving, or it was not buildable until recently. Name which one. A material that dropped in price, a component that became available, or a regulation that changed are all legitimate answers. “Nobody thought of it” almost never is.
Dimension 5: Personal Fit (weight 20)
The dimension most inventors skip and the one that determines whether the project finishes. It has no bearing on whether the invention is good and complete bearing on whether it reaches a market.
- Do you know this industry, or would you be learning it from zero?
- Can you fund $2,000 to $20,000 over 24 months without wrecking anything?
- Do you have 5 to 10 hours a week for two years?
- Are you willing to license, or does the plan require you to build a company?
- Do you have access to buyers, retailers, or manufacturers in the category?
| Score | Condition |
|---|---|
| 0 | No domain knowledge, no budget, no time |
| 1 | Interest only, in an unfamiliar category |
| 2 | Some budget and time, learning the industry from scratch |
| 3 | Working knowledge of the category, budget for the first $2,000 |
| 4 | You work in this industry and can fund through a utility filing |
| 5 | Industry knowledge, contacts, budget, and a licensing plan you can name |
Domain knowledge outperforms novelty instinct. An inventor who works in the trade knows the tolerances, the price points, and who buys, and those three facts save more time than any other advantage available at this stage.
The Scorecard
| Dimension | Weight | Your score (0 to 5) | Weighted |
|---|---|---|---|
| Market pull | 25 | score x 5 | |
| Manufacturing cost against price | 20 | score x 4 | |
| Patentability | 20 | score x 4 | |
| Competitive density | 15 | score x 3 | |
| Personal fit | 20 | score x 4 | |
| Total | 100 |
Under 45. Stop, or change something structural. Re-scoring the same concept with more optimism is not a strategy. Either the mechanism changes, the market changes, or the project ends. Ending a project at zero cost is a good outcome, and it is what the five mistakes independent inventors make are mostly about avoiding.
45 to 64. One dimension is dragging. Find it, fix that one thing, re-score. A cost problem is a design problem. A patentability problem may be a claim-narrowing problem. A market problem is usually fatal and worth accepting quickly.
65 to 79. Worth a professional search and a prototype budget. Spend the $399 to $2,500 on the search before anything larger, because the search result determines whether the rest is worth funding.
80 and above. Move. File a provisional, build, and start conversations. Concepts scoring here are uncommon and they are also perishable, since a category that obvious will not stay open.
Nothing in this scorecard predicts revenue. It measures whether the next expense is justified by what you know today, which is the only question a self-assessment can answer.
A Worked Example
A contractor’s tool that holds a joist hanger in position for one-handed nailing.
- Market pull: 4. Every framer holds the hanger with one hand and the nail with the other, and drops both. Twelve of twenty framers named a price without being asked. Score 4 rather than 5 because none of them asked when they could buy one.
- Manufacturing cost: 3. Two stamped steel parts, one spring, one molded grip. Estimated $3.10 landed against a $16.99 shelf price in the contractor supply channel. Reachable, with the grip as the obvious consolidation target.
- Patentability: 3. Six references in the classification code, one close. The difference is a detent that indexes to joist width, which is structural but arguably predictable from the closest reference.
- Competitive density: 3. Three products address the task, none with the indexing feature. Commodity pricing pressure exists.
- Personal fit: 5. The inventor is a framer with twenty years in the trade, knows two distributors, and can fund $6,000.
Weighted: 20 + 12 + 12 + 9 + 20 = 73. That is a concept worth a professional search and a prototype, with the patentability score as the open question. A $399 patent search resolves the item most likely to end the project, and it costs less than one week of the prototype budget.
Re-Score in 90 Days
Scores move. A cost estimate drops after a part-count pass. A patentability score changes when the professional search comes back. A market score rises after twenty more conversations. Re-score every 90 days and write down what changed and why.
What should not move is the standard. Adjusting the thresholds because you want the answer to be yes is how inventors end up $15,000 into a project that scored 40 the first time. The full spend at stake across the whole path is in the complete inventor cost breakdown, and the first concrete moves once a concept clears are in first steps after an invention idea.
If the score clears and the gap is engineering rather than judgment, that is the point where product development work starts paying for itself. Business-side questions, including pricing and market sizing, can be worked through at no charge with a Small Business Development Center advisor funded through the Small Business Administration.
FAQ
How do I know if my invention idea is good?
Score it against market pull, manufacturing cost relative to shelf price, patentability, competitive density, and your own fit with the category. A concept clearing 65 out of 100 has earned a professional prior art search. One under 45 needs a structural change or an honest ending. No single dimension decides it, and a strong score on novelty with a weak score on cost is a common and expensive failure pattern.
Should I pay for an invention evaluation report?
No, not as a first step. The inputs that determine whether a product works commercially are the manufacturing cost, the shelf price, and what buyers say, and none of those can be assessed before a design exists. Reports produced at the concept stage are templates. Spend the same money on a prior art search, which produces a document you can act on.
What is the most common reason an invention idea fails?
Cost against price. The mechanism works, the patent is available, and the product cannot be made for the 20% to 25% of retail that the distribution channel requires. This shows up late for inventors who evaluate patentability first and cost last, which is the wrong order. Estimate cost in week one.
How many people should I talk to before deciding?
Twenty, minimum, and they have to be people who perform the task rather than friends who find the idea interesting. Watch what they physically do rather than asking whether they like it. The number that matters is how many named a price without being prompted.
Can a good idea fail the scorecard and still be worth pursuing?
Yes, in one specific case: a low patentability score with strong market pull and strong cost economics. That is a product business rather than a patent business, and it competes on execution, brand, and distribution. Know which one you are running before you spend, because the two paths allocate money in different orders.