When does plain English become a real customer understanding issue?
New CUE research shows which language patterns put Customer Understanding most at risk
Nearly one in two basic investors misunderstood some everyday rule statements when familiar plain-English problems appeared in combination, new CUE calibration research suggests.
The Customer Understanding Exchange, CUE, has completed a recent calibration study designed to test a simple but important question: when does difficult wording stop being a theoretical plain-English problem and become a real customer-understanding risk?
The study asked almost 300 people, across three investor sophistication groups, to answer open-text questions about short everyday rule statements. In total, CUE analysed 1,776 free-text responses across Basic, Informed and Sophisticated target-market segments.
Unlike multiple-choice testing, respondents were asked to explain their answer in their own words. CUE used this format deliberately: open-text answers reduce the risk of people guessing from options and give a clearer view of what the wording actually communicated.
The findings were striking.
Basic investors did not struggle equally with every plain-English issue. A single clean negation, for example, was generally well understood. But where several issues combined, particularly nested conditions, long clause chains and reversal negation logic, misunderstanding rose sharply.
Put more simply, these are everyday drafting habits that make readers work harder. The key message gets interrupted by extra detail. Several things the customer needs to check or do are packed into one sentence. Or the point is framed around what will not happen unless something else happens. Each may seem manageable on its own, but together they can make the meaning much harder to follow.
And the customer understanding failure rates were high. For example, among customers with limited knowledge/experience in investing, statements that combined nested information, clause load and negation saw failure rates of between 36% and 60. For repeated “not unless” style conditions, more than half failed to give the correct answer.
This matters because regulated firms are increasingly expected to show that communications are likely to be understood by their intended audience, not merely that they are written in broadly acceptable language.
CUE’s early research suggests that traditional readability checks can miss the point. Looking at the basics like word count and passive voice matter, but the clearest customer-understanding risks appear when wording forces readers to hold several conditions in mind and work out how those conditions interact.
The finding is not simply that shorter is better. The real issue is cognitive load. A paragraph can look reasonably plain on the surface but still fail customers if it makes them track too many conditions, exceptions and timing rules at once.
Investor experience helped, but only up to a point. Informed respondents with greater investment knowledge & experience performed best, scoring around 12 percentage points higher than Basic respondents across the deliberately challenging test statements. Yet Sophisticated investors were only around 3 percentage points ahead of less sophisticated respondents overall. One possible explanation is that experience can sometimes create overconfidence: more sophisticated readers may move quickly through familiar-looking text, but still miss the precise conditions that determine the outcome. What is crystal clear is that cognitive load is not confined to inexperienced or vulnerable customers. Poorly structured information can overload even experienced readers.
CUE is using the findings to calibrate its Plain English Scoring model and refine the Virtual Customer Personas (VCP’s) used to test communications at scale. In discussions around AI-led testing and Consumer Duty, a central question is how such tools are calibrated to real customers. CUE’s approach is to combine behavioural research, real respondent data and paragraph-level analysis, rather than relying only on theoretical readability rules.
Rather than treating all plain-English issues equally, CUE uses the research to identify combinations of language features most likely to create misunderstanding for a given target market. The model distinguishes between warning signs and stronger red flags, using FCA best-practice benchmark alongside sample-size confidence checks.
The early evidence points to a clear pattern: firms cannot assess customer understanding simply by looking only at a document-level score. The real risk often sits inside individual paragraphs, where particular combinations of language, structure and conditions can make the meaning hard to follow.
For firms subject to Consumer Duty expectations, that is the critical distinction. A document can look acceptable overall while still containing paragraphs that customers are likely to misunderstand. The real test is whether firms can identify those points before the communication reaches the customer.