A patent protects a technical solution, not the wish to achieve a result. That is why the patent always comes after the idea.
Until then, the real work will be finding a solution defined clearly enough to be described, compared with what already exists and assessed against the patentability requirements. Only at that point will it make sense to move on to filing, claims, priority and international extensions.
We have already discussed how to patent an idea or an invention in 2026. The question we want to ask in this article is a different one: when has an insight reached a level of maturity that makes it genuinely sensible to talk about a patent?
Let us find out.
An abstract idea is not enough
The Italian Industrial Property Code allows inventions in every field of technology to be patented when they are new, involve an inventive step and are susceptible of industrial application. It excludes, however, considered as such, categories such as discoveries, scientific theories, mathematical methods, methods for doing business and computer programs.
The EPO adds a decisive element: there must be technical character. A purely commercial solution can be innovative for the market without being patentable.
‘I’d like a battery that lasts twice as long.’
No, as things stand
It describes a result.
“I have designed a new electrochemical architecture that produces that result.”
To be assessed
There is a technical solution.
‘I have defined the solution, checked the prior art and described how it works.’
May enter the patent process
There is a concrete basis to examine.
→The distance between the first and third cards is the space in which an idea becomes an invention.
The distance between the first and third lines is the real subject of this article.
The test that separates an objective from an invention
Many ideas are framed in terms of the desired result: consuming less, delivering faster, forecasting better, keeping cool for longer, reducing waste.
A patent requires a further step: what structure, process, mechanism or technical combination produces that result?
A useful question is this:
if I removed the final benefit from the sentence, could I still describe the concrete way in which the solution works?
If the answer is no, we are probably still in the territory of the idea. If the answer is yes, the project may be mature enough to warrant a patentability assessment.
The requirements: five gates to pass through
There must be a technical solution
A business model, a pricing rule or a different way of organising the business can create value without falling within the scope of a patent. Not ‘is it innovative?’ but ‘does it solve a technical problem through technical features?’
It must be new
The solution must not already form part of the state of the art: publications, earlier patents, presentations, websites, videos and trade fairs may all become relevant. Talking about a project too early can compromise its novelty.
It must involve an inventive step
New does not mean inventive: a modification may never have been made in that form and still be obvious to a person skilled in the art. It is the hardest requirement to judge from the inside.
It must have industrial application
The solution must be capable of being manufactured or used in industry. It does not need to be a product already in production, but a concrete and workable implementation.
It must be capable of sufficient description
The application must enable a person skilled in the art to carry out the invention. “A battery that lasts ten times longer” defines the result; it does not teach how to achieve it.
1. There must be a technical solution
The EPO requires the invention to have technical character. A new business model, a pricing rule or a simple way of organising activities may create economic value without, for that reason alone, falling within the scope of patent protection.
The question is therefore not ‘is it innovative?’ but ‘does it solve a technical problem through technical features?’
2. It must be new
The solution must not already form part of the state of the art. Publications, earlier patents, presentations, websites, videos, trade fairs and other disclosures may become relevant.
This is where a practical risk emerges: talking about a project too early can compromise novelty. WIPO recommends keeping the invention confidential before filing, subject to specific assessment of the applicable rules.
3. It must involve an inventive step
New does not automatically mean inventive. A modification may never have been made in exactly that form and still be obvious to a person skilled in the field.
This is often the hardest point to judge internally: those who developed the project tend to see as original whatever cost them time; a patent, by contrast, looks at how far the solution departs, in a non-obvious way, from the state of the art.
4. It must be capable of industrial application
The solution must be capable of being manufactured or used in industry. It does not necessarily need to be a product already in production, but there must be a concrete and workable implementation.
5. It must be capable of sufficient description
The application must contain information clear and complete enough to enable a person skilled in the art to carry out the invention. UIBM and WIPO explicitly refer to this requirement.
This is one of the reasons why a mere intuition falls too far short: ‘a battery that lasts ten times longer’ defines the result; it does not teach how to achieve it.
The right level of maturity: defined enough, not necessarily finished
There is no need to wait until the product is ready for market. In fact, waiting too long can increase the risk of disclosure and narrow the options for protection.
At the same time, filing too early can be just as dangerous. WIPO points out that if the invention evolves after filing, it is not normally possible to freely add substantial new technical content to the original application.
The best window opens when the technical principle is stable enough to be described and claimed, but before the project is shown publicly without a strategy.
Five questions before talking about patents
What technical problem am I solving?
that a real technical problem exists
What is the concrete solution?
that we are not merely protecting an objective
Which part do I consider new?
that the distinctive core is identifiable
Why would it not be obvious to an expert?
that a possible inventive step exists
Could I describe it well enough for a technician to understand it?
that the solution has reached sufficient maturity
If the answers are vague, the project still needs to mature. If they become precise, it makes sense to invest in a prior art search and a professional assessment.
Seven examples to understand the boundary
A social network with a new interaction model
The platform concept and the social rules belong mainly to the business model. A specific technical solution to a technical problem, however, may call for a different analysis.
A bottle that keeps its temperature for longer
The benefit alone is not enough. A new insulation system or a particular technical structure may enter patentable territory.
An algorithm that forecasts sales
A mathematical formula as such runs into limits. A computer-implemented solution can be assessed when it produces a technical effect and solves a technical problem.
A chair with a new shape
If the novelty is mainly aesthetic, a design right may be more relevant. If it introduces a particular functional benefit, a utility model may also become relevant.
A delivery service with innovative pricing
It can be a good commercial innovation without being patentable in nature.
A machine that reduces production waste
There is typically patentable ground here: a technical problem, a concrete solution and an industrial application. Novelty and inventive step remain to be verified.
A new chemical formulation
It may fall within patentable scope when it meets the required criteria and is sufficiently described.
Registering an idea: the expression conceals different instruments
“Registering an idea” does not refer to a single right. A project may contain different elements and require different forms of protection.
| Project element | A tool that may become relevant |
|---|---|
| Technical solution | Patent for invention |
| Functional improvement of certain objects | Utility model |
| Shape and appearance of the product | Registered design |
| Name and distinctive sign | Trademark |
| Software, texts and original works | Copyright |
| Confidential information and know-how | Trade secret |
The strategic question is therefore not ‘how do I register my idea?’ but ‘what assets does the project contain, and what protection suits each one?’
Software and AI: adding an algorithm is not enough
In the European context, computer programs as such are excluded. Computer-implemented inventions can, however, be patentable when they have technical character.
In 2026 the EPO updated its guidance on digital inventions: software, algorithms and mathematical models can fall within the scope of protection when they contribute to a technical effect by solving a technical problem.
The same applies to AI. The presence of a machine learning model does not make a solution patentable by definition; what matters is the technical contribution that solution makes.
A prior art search also helps improve the project
A prior art search is often seen as a bureaucratic check: finding out whether someone else has already patented something similar.
In reality, it can serve a design function. It forces the team to pinpoint the novel element more precisely, separating what already belongs to the state of the art from what truly represents the invention’s contribution.
WIPO stresses that a preliminary search can avoid investing in an application destined to run into overly strong prior art.
Disclosing before filing: the problem of sequence
Many projects go public precisely when they are most fragile: fundraising, pitches, trade fairs, market tests, partnerships, communications.
The sequence matters.
Develop → assess → protect → disclose
is very different from:
disclose → seek protection later.
Non-disclosure agreements and access controls may form part of the safeguards, but the strategy must be defined according to the project and the jurisdictions concerned.
From idea to asset
A patent gains meaning when it becomes part of a strategy: it can protect a product, widen the gap with competitors, and support licensing, partnerships, fundraising or due diligence.
It can also cost more than the value it protects, if filing is treated as an end in itself.
The final question is therefore more useful than the initial one: what are we building, how much does it matter to the business, and which part of that value deserves protection?
When the answer genuinely leads to a technical invention, the next step is the complete guide to patents: search, drafting, filing, priority and extensions.
Domande frequenti
Can you patent an idea without a prototype?
A prototype is not a universal requirement. The solution must, however, be sufficiently defined and described to enable a person skilled in the art to understand and implement it.
Does registering an idea mean obtaining a patent?
No. A project may call for a patent, design, trade mark, copyright, utility model or know-how protection, depending on the asset.
Can I patent a project that has already been shown publicly?
Prior disclosure can affect novelty. You need to establish what was disclosed, when and under what conditions before assuming that patentability remains intact.
Can software or an AI system be patented?
In Europe it can be, when the computer-implemented solution has technical character and contributes to solving a technical problem, in addition to meeting the other patentability requirements.
Can Bliss replace an industrial property adviser?
No. Legal assessment and drafting the application require specialist expertise. Bliss can work on the strategic value of the asset, its competitive role and the alignment between intellectual property, positioning and business development.
Fonti e riferimenti
- Ufficio Italiano Brevetti e Marchi, Istruzioni per la presentazione delle domande di brevetto per invenzione industriale o modello di utilità
- European Patent Office, Is it patentable?
- European Patent Office, Patentability requirements, Guidelines for Examination 2026
- European Patent Office, Patentability of digital inventions
- WIPO, How to Protect Inventions through Patents
- Bliss, Come brevettare un’idea o un’invenzione nel 2026: la guida definitiva