IPwe’s Valuation Methodology

IPwe’s Valuation Methodology

IPwe’s AI-Based Valuation Methodology

In contrast to other economic goods, patents have 3 peculiarities that make their evaluation and valuation challenging, that IPwe’s Valuation Methodology takes into account:

   1. Patents are unique

   2. Patents are created deliberately abstract

   3. Patents are territorially limited

Patent valuations have 3 dogmatic groups:

   1. Market-based approach

   2. Cost-based approach

  3. Income—based approach*

Smart Intangible Asset Management’s Valuation Methodology

IPwe’s Smart Intangible Asset Management solution utilises an income-based valuation approach. The different established valuation approaches vary largely in their angle of assessment, the key metrics that they’re based on, and the amount of data that needs to be provided by the asset owner. This leads to a drastic variation in results.

Our Patent Valuation Model integrates patent and financial data from multiple data sources in a relief from royalty model**. By automatically feeding the model and applying a uniform qualitative standard of assessment to all valued patents, it eliminates the inherent subjectivity associated with income-based valuations.

IPwe’s core innovation comes from pre-aggregating cleaned and normalised data in the model so that the only input parameter required to carry out the valuation is the target patent number. This results in a previously unattainable level of accuracy in valuation for a process that usually takes significant time and triggers high discovery costs.

Our algorithm carries out the following steps automatically to obtain a sophisticated economic valuation of the target patent:

1) Extracting input data (CPC codes etc…)
2) Market mapping
3) Qualitative analysis of the target patent and the whole peer group in the market 4) Calculating the remaining useful life
5) Forecasting of attributable revenue and annual growth
6) Determining risk-adjusted discount rate
7) Determining applicable royalty rate
8) Determining pre- and after-tax royalty savings
9) Output: DCFA of royalty savings over the remaining useful life

IPwe’s Valuation Methodology provides a new perspective on intangible assets that are rarely valued at scale. To gain a more in-depth insight into the mechanics of our valuation, a detailed published review can be downloaded here.

* Income-based approach: Value an asset based on the past, current, or expected cash flows of the asset.
** Relief from Royalty Model: Calculate value based on the hypothetical royalty payments that would be saved by owning the asset rather than licensing it.

IPwe Analytics, IPwe’s Valuation Methodology

For most of the scenarios stakeholders face, IPwe Analytics is an excellent tool to provide initial guidance and to refine and help more precisely define relevant areas for further inquiry.


The Patent Quality (Q) process replicates the methodology an expert would employ to identify the most significant patents related to a specific patent or technology area.

IPwe operationalises this process by using a random-walk-with-restart (RWR) model***, which is a stochastic process on networks****, and in its simplest form computes for each network node the steady state probability that a random walker will end up on that node.

By doing so, IPwe Analytics gains a comprehensive insight into the relative positioning of the asset compared to its peers. It assesses the asset’s stage in the filing cycle, the level of acceptance by peers through citations, and analyzes the composition of the asset’s structure, including the protected elements and potential fallback positions.

This process is more reflective of what a patent expert would use to define the quality of a patent.


Patent Validity (V) is a term that encompasses the different ways in which a court or administrative process can render a patent unenforceable.

IPwe evaluates validity by considering each claim of a patent separately on an element-by-element basis, ensuring all art nodes are linked.

The methodology IPwe Analytics use to identify prior art is superior to other methods because, by the Art Classification Node***** , IPwe Analytics focuses on the subset of patents or non-patent literature that humans directly or in-directly determined to be relevant.

IPwe Analytics relies on a curated dataset of over 130 million patents and patent application records dating as far back as the early 1900s to analyse all art relevant to a patent.

This accumulation of patent data, alongside over 2 million articles and journals and regular downloads from the internet, amass into the ‘IPwe Analytics Prior Art Library’- a leading database for patent analysis.


*** Random-walk-with-restart: An algorithm that provides a good relevance score between nodes in a weighted graph.

**** Stochastic: Presence of a random variable.

***** Node classification models: Aim to predict non-existing node properties based on already classified nodes.

IPwe Ratings

Financial ratings focus is usually on considerations of purely financial ROI and the associated risk hedging. However, rating types have also emerged whose reference object is the quality of companies and financial instruments.

In these cases, the risk assessment is based on what happens if the reference object does not meet the quality benchmark.

There are no such ratings for IP and especially inventions. If the assessment costs more than the financial benefit gained, then the assessment won’t take place.

These times have changed for IP. AI has enabled batch-scoring across whole industries. While blockchain technology completes this technological breakthrough by providing an unbroken and unforgeable chain of evidence through tokenisation.

IPwe completes this through a patent-to-patent-based assessment and comparison.

Importantly, IPwe breaks down its ratings into industries to avoid misleading cross-assessment of assets. In IPwe’s ratings, only stacks of patents belonging to the same technological area are compared with each other.

Once the target portfolio is split into its respective industries, the rating methodology adopts a bottom-up approach. This bottom-up approach assesses individual inventions and aggregates them to the technology level.

Each invention has a total of 11 attributes associated with it that are structurally relevant for the rating, with each attribute having a score ranging between 1-100:

1) Average Validity Score

2) Average Quality Score

3) Transaction Likeliness

4) Drop Likeliness

5) Patent Family Age

6) Key Markets

7) Technology Hotness

8) Family Grant Saturation

9) Key Market Grant Saturation

10) Enforcement Territories

11) Enforcement Territories Grant Saturation

The attributes are absolute values important for strategic decisions.

Ratings scaling from AAA-D summarize all of this, offering a high-level understanding of the asset’s financial quality.

Jonas Block
Jonas Block