Defining intelligibility vs readability
Defining intelligibility vs readability
We’ve matched the development of the Amplifi tool to the regulatory requirements and the legal definition of intelligibility affecting all communications to consumers.
The underlying legal obligation is for consumer communications to be ‘intelligible’ not just readable. Readability is just one component of intelligibility, yet it is what most tools in the market focus on. The Consumer Understanding Outcome has reinforced the need for documents to be not only readable, but intelligible, and that the information is capable of being used by consumers.
What is Intelligibility and how is it different from Readability?
The law concerning what ‘intelligibility’ means has been defined by previous court cases. It means more than simply ‘capable of being understood’. A more accurate definition (and relevant to the FCA Consumer Duty, for example) would be: “able to be understood by the target audience, including the context, the relevance, and the consequences of the information for the reader.”
Conversely, the concept of readability is not clearly defined in law, hence the wide-ranging and often competing definitions used in academic texts and everyday life. However, in a legal sense, the terms ‘plain’ and ‘legible’ are close to providing a readability definition – i.e. a narrow definition relating to grammar, presentation and accessibility, rather than a measure of likely understanding or comprehension.
Readability frameworks have been shown repeatedly to be poor indicators of whether a communication will be understood. The work of Professor Richard Hyde from Nottingham University has helped to demonstrate this. It is therefore somewhat concerning that the Consumer Duty Guidance from the FCA references readability as part of delivering the consumer understanding outcome.
Intelligibility and foreseeability are two concepts that can be seen reflected in the new Consumer Duty rules. Since CJEU court cases around the time the Consumer Rights Act 2015 came into force, the definition of intelligibility has been widened to also cross-over with the concept of foreseeability. Something being ‘foreseeable’ has long been defined in other areas of law (foreseeable harm, for example).
Foreseeability means that, as well as having the correct information made accessible to them, and being able to understand the language and concepts being used, the average consumer must also be able to understand “potentially significant economic consequences” of the term of the contract. This includes being able to understand the potential risks to the consumer, as well as their rights and obligations.
In terms of concept, intelligibility and foreseeability are intrinsically linked, and hard to clearly distinguish from each other – they exist hand in hand, each necessary for the other.
However, these legal definitions are not widely understood, and many firms overly rely on readability as their key consumer duty understanding metric.
The shortcomings of readability frameworks
Traditional readability metrics measure some grammatical and structural aspects, but intelligibility requires that the information is sufficiently understandable – in terms of the words used, and the concepts being communicated.
And crucially, that the consumer can subsequently then use this information, whether to compare products or services, or to understand their rights, obligations, and the key risks they may face or actions they may take. This has been legally defined as the customer communications being ‘foreseeable’.
Foreseeability and intelligibility are intrinsically linked, and collectively require far more than just being readable.
Drivers of complexity
There are various drivers of complexity we frequently see in regulated consumer communications, such as:
- Prescribed or Defined words as a result of legislation or regulation, or specific information which must be included in certain types of documents.
- Complex concepts could be any term or concept that the average consumer is not used to. Often these may be normal for legal teams or experts to use, but for the average consumer they can be challenging. They may be used to convey complex layers of meaning. But if a word is unusual or complex, it may not be understood.
- Complex sentences may include sub-clauses, or complex words. The overall volume of information can also decrease intelligibility.
- The length and structure of a document affect reading ease. The longer a document is, the more difficult the information is for the reader to understand.
- The use of Passive Tense can make information more challenging to understand.
- Long words and sentences can be replaced with simpler or shorter terms. The average word and sentence length go together to form the readability score. This provides insight into how readable a document might be, but not the intelligibility of words or concepts, and whether the information is likely to be understood.
Measuring Intelligibility
The range of metrics we assess in the tool makes us the only true intelligibility assessment tool available at the moment.
Intelligibility Score
The Intelligibility Score is a unique measure of intelligibility. It uses a range of metrics, which include:
- how readable the document is
- how complex the words and concepts are
- how connected or disconnected the concepts are
- other factors that affect whether the document will be understood.
Our Intelligibility Score is out of 100. The higher the score the more intelligible the document is assessed to be. The scores are aligned to 5 Intelligibility Levels. These help to communicate reach (what proportion of a given population is likely to fully understand the information), and the likely level of education or experience that may be needed for an adult to comprehend it.
The levels range from Level 1 (least intelligible) to Level 5 (most intelligible).
Complexity Score
The Complexity Score takes into account what words actually mean, and how difficult those concepts are to understand. This is calculated at an individual word-level, and whether they are used adjacent to other difficult concepts in a sentence.
The overall Complexity Score is scored out of 100, the higher the score the less complex the document is conceptually, and the more likely it is to be understood (fully), and the reader able to use the information, such as to inform a decision..
Document length and reading time
Length of document can have a significant effect on how intelligible a document will be in practice. This is due to the volume of information being communicated, and factors such as the potential for information to be less clearly structured in a longer document.
Engagement may also be reduced the longer the document or reading time is perceived to be by the reader.
Readability Score
Readability Scores such as Flesch Kincaid concern aspects such as how long the sentences are, how many syllables or how long the words are. The structure of a document and grammar can also have an effect.
These factors have the benefit of being objective and are widely used. They form the basis for readability frameworks. Yet they have been shown to be poor indicators of whether something can be understood, and whether the reader can then apply that information.
This wider concept of understanding, and being able to then apply that information is in essence what intelligibility is, and what the Consumer Duty and the Consumer Rights Act require. Readability is one intelligibility factor, but insufficient on its own to assess whether a communication is likely to be understood.
Readability metrics
The two factors that most impact the Readability Score are the average sentence length, and the average number of syllables per word. They are important factors in what makes a document able to be read fluently.
These two factors are the only two inputs to the Flesch-Kincaid model, which is still often used to provide target reading age by firms. We think this is resulting in compliance risk for those firms, when used as a consumer understanding measure. It is a poor predictor of understanding.
Why does this matter?
Firms must move beyond using readability metrics as a means of testing their communications. Readability (in practice usually taking just sentence length and word length as a measure) has been shown to be a poor predictor of understanding outcomes in a number of studies. The FCA has signalled that they won’t trust standard readability scores as a proxy for understanding:
"Borrowers often struggle to understand complex legal and financial documents, leading to poor financial decisions and confusion. Readability scores alone usually overlook complexity and a user's comprehension…." FCA
Relying on readability data alone leaves significant residual risk, as it fails to measure whether a consumer can truly understand and use the information. We tested current account terms and conditions for a large high street bank, using Amplifi as well as a range of readability testing tools. The difference in scores (for the same content), shown below, is striking: Intelligibility assessment surfaces risks that readability assessments do not identify.
There is a legal requirement that consumer comms are able to be understood, and the information acted upon. Too often this simply isn’t achieved.
This issue is now firmly under the spotlight.
The Gambling Commission has focused on the need for more understandable terms. The FCA too via its new Consumer Duty. And the Competition and Markets Authority is gaining new powers to police the Consumer Rights Act. The risk of being fined 10% of global turnover tends to focus the mind!
These factors are forcing the legal teams that draft and sign off consumer communications to reconsider how understandable their work is to an average customer. This is true whether it’s product information, terms and conditions, employment policies - the list is endless.
It is the need for intelligibility that lies at the heart of regulation and law, in the UK and beyond. Grammatical ‘readability’ alone is not sufficient, and the focus on tools like Flesch Kincaid as a way to measure intelligibility leaves legal teams, their clients, and the consumer at significant risk. Yet too often it is the crutch that legal teams rely on.
How can Amplifi help?
Amplifi is able to delve much deeper, assessing a range of factors. Not only classical and grammatical metrics like Flesch Kincaid, but also to assess how understandable the words and concepts being used are.
This provides a much deeper level of analysis of whether something is truly ‘intelligible’ – i.e., whether the meaning is understandable, and will provide information that a consumer can then use effectively.
The Amplifi platform assesses the likely understanding of all communications by moving beyond surface-level readability metrics to evaluate the deeper intelligibility of a document. We define Intelligibility in line with the Consumer Rights Act and the relevant court cases - i.e. the ability of consumers to be able to read, understand and use the information to inform their actions, and to understand and foresee their risks, rights, costs etc. This is further reflected in the wording of the FCA Consumer Duty around informed decision making.
Utilising proprietary AI and machine learning models trained on legal and regulatory communications, the tool analyses a document's architecture, including its structural density, sentence complexity, and the connection between concepts. This diagnostic capability generates an Intelligibility Score (0-100) and categorises the text into five levels that indicate the ‘reach’ (% of the UK adult population that are likely to fully understand it), and likely education or experience an adult needs to comprehend the content.
By identifying specific intelligibility risks, such as conceptual complexity or poor sentence structure, the tool provides an objective, data-driven audit trail that allows firms to prove whether their communications meet the high standard of clarity required by the FCA’s Consumer Duty.
To ensure the purpose of the communication is fully understood, the solution utilises our Multi-Level Comprehension Framework, which is able to distinguish between simple recall and high-order cognitive processing, and underpins our research and validation of the model. Our framework has been used by the FCA in its own published research (e.g. in its consultation paper “CP25/17 - Supporting consumers’ pensions and investment decisions: proposals for targeted support” and the FCA research note "Reading Between the Lines: Understanding of targeted support in retail investments"), and validation research carried out together with the Solicitors Regulation Authority (SRA). The method assesses a user’s ability to comprehend information at various levels. The diagnostic tools specifically measure factors that affect a consumer's ability to grasp the main message of a document vs their ability to take action or understand a scenario, ensuring that critical information is not obscured by technical jargon or information overload.
We have also worked with the SRA to test our intelligibility technology across the legal sector. We assessed and simplified regulatory guidance communications, then tested them with both consumers and legal professionals. The results validated the approach - the results of simplification included increased understanding outcomes across all the groups we tested, particularly at higher levels of comprehension needed for informed decision making. Simplified text also improved engagement, resulted in faster reading time, and improved trust in the provider of the information.
How does it work?
Amplifi scores every document using its proprietary Cognitive Risk Engine. The score measures how intelligible a communication is likely to be for a given reader group, going beyond standard readability metrics to assess conceptual complexity, ambiguity, and genuine comprehension risk. It is expressed as a numeric score mapped to a five-level intelligibility scale, where Level 1 represents the highest risk and Level 5 represents the lowest.
Users can see intelligibility scores across their entire portfolio and prioritise documents with the highest consumer understanding risk. The score then updates automatically each time a document is edited. There is no need to re-upload or manually trigger a new assessment. Amplifi is designed to support iterative testing across multiple versions of the same document. Teams can test, improve, and retest within a single workflow, seeing the impact of each change immediately and adjusting accordingly. This removes the cadence constraints that limit panel-based testing, where cost, operational effort, and participant fatigue mean testing can only happen infrequently and at fixed points in the drafting process. With Amplifi, testing is continuous and embedded into the drafting process itself.
Every draft, every edit, and every iteration is assessed automatically against the same objective standard (intelligibility). Teams can see the impact of a change immediately and adjust accordingly, compressing what would otherwise be a lengthy review and revision cycle into a single workflow. This makes rapid iteration practical at scale, supporting the kind of ongoing optimisation that panel-based approaches cannot sustain across a full document portfolio.
The FCA expects boards to actively challenge compliance data, evidence good outcomes across all four pillars, and go beyond readability metrics. See here for a practical guide to the Consumer Duty board report requirement, with the 31 July 2026 deadline in focus.
Explore more articles
