How to complete a patentability search in IPRally in minutes using natural language text

What is a patentability search?

A patentability search helps determine whether your invention is novel and eligible for patent protection. It supports R&D teams, patent professionals, and startups in assessing the risk of rejection and optimizing the scope of a patent application before filing it. Usually, it also gives valuable information on competitors’ activities in the field of the invention and helps to assess its business value. 

Traditional methods are manual and keyword-heavy, but IPRally’s AI automates the hard part: understanding the invention and comparing it with millions of documents.

IPRally reads and understands the technical meaning of your search input, then finds similar documents based on conceptual similarity, not just matching words, and finally analyses all the results in depth with professional level reasoning. In this example we'll use text as input, but you can use images as well. 

Here’s how to do it:

1. Paste or type your invention concept

You can start with almost any form of input: a claim set, an invention disclosure, or even an informal technical description.

Your input doesn’t need to be polished. It works even if it’s unstructured, not in claim format or partial or conversational.

Text input example:

An AI-driven snowthrower system that includes a motor - probably electric but maybe gas-powered too - which drives an auger that spins fast enough to shoot snow out the front or side, depending on design. The auger sits inside a housing (could be plastic and metal mixed together, or whatever combo works best to make it tough but not too heavy), and the whole thing’s got a frame that holds the parts together, from the handle all the way to the front end.

There’s a handle part, like what the user holds and pushes the snowthrower with, and it’s also used for steering, maybe with sensors or even buttons to help guide it.

The smart bit - the AI control - is built into the system and takes live inputs while the thing is running. It watches how heavy the snow is (like how clogged or strained the auger feels), what the user’s doing (are they walking fast or slow, turning a lot?), and it somehow reads the terrain, like bumps or slopes or icy patches maybe. Based on all that, the AI adjusts how much power the motor gives and how fast the auger spins, trying to find the best performance - so it doesn’t overload, or throw snow badly, or jerk around.

2. Refine your text input if needed

If your text input is rough or casual like above, just click Refine. This feature cleans up the language and restructures your input into a claim-style format, making it easier for the AI to digest and structure.

This is what the refined version of the original text input above looks like: 

1.  An AI-driven snowthrower system comprising:

  -   a motor,

  -   an auger driven by the motor,

  -   a housing enclosing the auger,

  -   a frame supporting the motor, the auger, and the housing,

  -   a handle for steering the snowthrower, and

  -   an AI control system configured to:

         -  receive live inputs including snow load on the auger, user input, and terrain data, and

        -   adjust motor power and auger speed based on the live inputs.

2.  The system wherein the motor is an electric motor.

3.  The system wherein the motor is a gas-powered motor.

4.  The system wherein the housing comprises a combination of plastic and metal.

5.  The system wherein the handle comprises sensors for steering.

6.  The system wherein the handle comprises buttons for steering.

7.  The system wherein the terrain data includes data relating to bumps, slopes, or icy patches.‍

3. Run the search and browse results

Click “Search Patents” and IPRally will convert your free text to a knowledge graph, which is a tree-like structure of the invention's key features, their functions and their relations.

IPRally then compares this knowledge graph against millions of patents to identify the most technically similar prior art, independently of keywords, language, and official classifications.

4. Use IPRally’s AI assistants to filter and understand the results

Once your initial results are in, it’s time to make sense of them. IPRally gives you two powerful tools to explore and refine the list: Ask AI and Smart Filters.

With Ask AI, you can ask natural language questions across multiple documents without having to read them. Just apply it to all of the search results to get a quick start for your review. Here’s an example giving you immediate useful information for your review process:

“What does the document teach about adjusting the motor power of snowthrowers?”

If your question is binary, i.e. can be answered with a Yes or No, you can apply Smart Filters to filter out the irrelevant hits from the search results. In addition, you will get a full reasoning for the answer, grounded to specific paragraphs in each of the documents. 

In our sample case, suitable binary questions could be for example:

"Does the document disclose a snowthrower with an AI based control system to adjust motor power?"

"Does the control system receive live input on snow load on an auger?"

The results are automatically grouped based on AI’s answer: Yes, No or Maybe. Select “Hide No Patents” to instantly narrow down your results to the most relevant documents with no manual reading required. Open the Smart filters panel to filter on either Yes, No or Maybe.

5. Fine-tune your patentability search

After using Ask AI and Smart Filters, you’ll likely have a few strong hits marked as Favorites. Use Zoom to Favorites to trigger a new AI search, this time centered around the technical features of your favorite documents. This allows you to explore similar documents that may have been missed initially.

Best practices for free text search input

Use natural language. Write your input in full sentences with proper articles and punctuation. This aligns with how the AI has been trained and ensures better understanding of your invention.

Provide structured input where possible: Professionally drafted claims are ideal, since the AI is trained on millions of them. If you don’t have claims, a technical description or invention disclosure works well ​​– especially if it includes specific embodiments, technical functions, or contextual details.

Refine informal text: If your input material is e.g. informal bullet points, conversational or otherwise non-patent in style, use the Refine button. It will convert your text into a mock claim set, making it easier for the AI to parse and compare.

English is optimal. Use English whenever possible, but IPRally also supports German, French, Spanish, Chinese, Japanese, Italian, Dutch, Danish, Finnish, and Swedish through secure in-app machine translations (up to 5000 characters).

Organize with Projects: Make all your searches inside a single Patentability search Project to easily keep track of the relevance of documents, avoid double work, and sandbox your Ask AI questions and Comments within the Project.

Trust data security: All information entered into IPRally is encrypted both in storage and in transit, and IPRally employees cannot access your input.

Pro tips for best results

Provide enough context and be specific: For example “mobile communication device” instead of “device”) - AI is powerful but it cannot read your mind.

Details matter: Functional relationships are particularly important. These were out of question with classical Boolean methods, but IPRally’s AI has changed that.

Don’t shy away from patent jargon: The AI is trained on it and understands it!

Aim for a clear graph: Try to achieve a logical knowledge graph, but perfection is not necessary. Refine button helps to optimize the input for graph AI purposes.

Prefer full terms over abbreviations: Common acronyms are recognized, but full terms are preferred (e.g., “scanning electron microscope” instead of “SEM”), in particular if the context is otherwise shallow.

Learn more from IPRally's customers

"Thanks to IPRally we now spend three hours per search compared to a day and a half."

The semiconductor company Melexis reduced their time spent on patentability studies by a staggering 75% by using IPRally, and the quality of search results allows writing claims around the prior art in an early phase.

“The R&D team can quickly assess if something is worth filing – or even patentable at all.”

Metsä group's Patentability assessments that once required outsourcing, can now be completed in-house within minutes or hours, thanks to IPRally.

Get started with your own patentability search‍

Want to see how fast and easy a professional patentability search can be?

• Paste your invention in natural language
• Let IPRally’s AI do the heavy lifting
• Get novelty-relevant prior art in minutes – no Boolean required

‍Start a free trial – All features mentioned in this guide are included

‍Get a live demo – See invalidity workflows tailored to your specific needs