What does a policyholder actually understand from the letter that says no? We ran 22 user tests for an insurer to measure comprehension of a denial letter as written, and then of the same decision rewritten in plain language. Comprehension began at 37 percent, and the rewrite reached 79 percent, a gap wide enough to change what fairness itself felt like.
I Discrimen
A denial can be correct in law and still fail as communication
The decision and the explanation are two different products, and only one was tested here.
Our study drew a line that claims teams do not always draw. The decision to deny a claim is a matter of policy wording, evidence, and judgement, and our tests did not ask whether that decision was right. They asked whether a reader could understand what had been decided, why, and what could happen next.
In the original letter, comprehension stood at 37 percent. That means most test readers, given time with the document, could not reliably say what it was telling them. A letter that cannot be understood cannot be accepted as fair, however sound the reasoning buried inside it may be.
You will know the reader state this creates if you have ever watched someone receive bad news in writing. They read the first paragraph twice, skip to the end looking for the point, and then decide how they feel based on tone rather than content, a pattern familiar to anyone reading how to choose a PR agency for insurance companies. Confusion curdles into suspicion very quickly.
An unexplained no sounds like an unfair no.
II Machina
Twenty two user tests showed us where readers actually stopped
Watching real readers fail beats asking writers whether the letter seems clear.
Our instrument was a set of 22 user tests built around the denial letter of an insurer. Test readers worked through the letter as a policyholder would, and we recorded whether they could explain the decision back accurately. Comprehension is a stricter measure than readability scoring, because it asks what landed, not what the text looks like on a formula.
We then rewrote the letter in plain language and tested that version the same way. The comparison held the decision constant and changed the expression, which is the only honest way to price the writing itself. One element, a next step box, was examined separately for its effect on perceived fairness.
The approach follows the spirit of the Federal Plain Language Guidelines maintained at plainlanguage.gov in 2025, which treat user testing as the proof that a document works for its intended readers. Teams used to evidence based communication, including those assessing how to choose a PR agency for banks, will recognise the discipline: measure the reader outcome, not the author intention.
If readers cannot repeat it back, the letter did not say it.
III Exemplum
The rewrite took denial letter comprehension from 37 to 79 percent
Same decision, same reader task, and a result more than twice as strong.
Here is the centre of our research. Denial letter comprehension in the original version was 37 percent. After the plain language rewrite, comprehension reached 79 percent. The decision had not softened, the policy had not changed, and the answer was still no, yet readers understood what they had been told.
That difference reframes what a rewrite is worth. Moving from 37 percent to 79 percent is not polishing, and it is not cosmetic editing applied to a finished document. It is the difference between a letter that functions as communication and a letter that functions mainly as a record that something was sent.
| Measure | Insurer test | Benchmark median | Benchmark top quartile |
|---|---|---|---|
| Comprehension of the original denial letter (percent) | 37 | 46 | 71 |
| Comprehension of the rewritten denial letter (percent) | 79 | 46 | 71 |
| Fairness gain from the next step box (points) | 16 | Not benchmarked in the record | Not benchmarked in the record |
| Complaint intent change after the rewrite (percent) | Down 21 | Not benchmarked in the record | Not benchmarked in the record |
fons: Aerovector research record for the study figures and for the Benchmark Claims Clarity median of 46 percent and top quartile of 71 percent in this table.
Set the pair against the benchmarks in our record. The Benchmark Claims Clarity median is 46 percent and the top quartile is 71 percent, so the original letter sat below the median while the rewrite cleared the top quartile. One document, tested honestly, moved from the weaker half of the field to beyond its strongest quarter.
Clarity is not the decoration on a decision, it is half the decision.
IV Damnum
An unclear denial manufactures complaints that the decision never caused
Readers who cannot find the reason go looking for a remedy instead.
Before the rewrite, the letter was producing a predictable secondary cost. A policyholder who cannot understand a denial has three options, all expensive for somebody: call and ask, complain formally, or assume bad faith and escalate. Each route consumes claims time, regulator attention, or trust that took years to accumulate.
Perceived fairness is where the damage concentrates. The Plain Writing Act of 2010, published in full by the U.S. Government Publishing Office, treats plain writing as an obligation because people cannot act on decisions they cannot parse. In insurance the same logic is commercial, since a fairness judgement formed in confusion is rarely revised later in your favour.
Notice, too, who pays first. The reader pays in time and anxiety, on what may already be a bad day involving a loss, an illness, or a bill. Organisations studying member owned communication, such as readers of how to choose a PR agency for credit unions, tend to grasp this quickly: the letter is often the whole relationship at that moment.
Confusion is a complaint you wrote yourself.
V Genus
Denial letters fail in kinds, and each kind has a different repair
Diagnosing the failure precisely is what makes a rewrite fast instead of endless.
Our tests showed failures that fall into recognisable kinds. Some readers could not find the decision itself, because it arrived late in the letter after background and caveats. Some found the decision but not the reason, because the reason was expressed in policy terms the reader did not share.
A third kind found both and still failed, because nothing told them what they could do next. That kind matters disproportionately, since a reader with a path can accept a decision they dislike, while a reader without one experiences the letter as a wall. Health Literacy Online, published by the Office of Disease Prevention and Health Promotion in 2025, makes a parallel point: structure and action steps decide whether information becomes usable.
Sorting failures by kind turns an argument about style into a work list. Move the decision up, translate the reason, and add the next step, and you have repaired the three failures our test readers actually had. Similar letter problems appear across regulated sectors, as the discussion in choosing a PR agency for mortgage lenders shows for lending decisions.
Name the failure before you touch the sentence.
VI Regio
Plain language duties differ by jurisdiction, but reader confusion does not
The legal map changes across borders, and the human response to a dense letter does not.
Claims letters are written inside regulatory frames that vary by place. The International Association of Insurance Supervisors, in its 2025 supervisory material, works across jurisdictions whose conduct expectations differ in detail, and the National Association of Insurance Commissioners, through its 2025 consumer resources and its Center for Insurance Policy and Research, reflects a state based system of consumer protection. Your letter must satisfy the frame you operate in, and that frame is local.
What is not local is the reader. A policyholder in any jurisdiction meets a denial at a kitchen table, without a lawyer or a glossary, after a loss. Build to the stricter standard in front of you: compliance decides whether the letter may be sent, and comprehension decides whether it deserved to be.
The table is the same in every time zone.
VII Occasio
A next step box raised fairness by 16 points on its own
One small panel changed how the whole letter was judged, before any outcome changed.
One element of the rewrite earned its own measurement. Adding a next step box, a short panel stating what the reader could do after the denial, raised fairness by 16 points. Readers who could see a route forward judged the same decision, with the same result for their claim, as fairer than readers left at a dead end.
Think about why that works. The box does not promise success, offer compensation, or reopen the decision, and test readers understood that. It simply converts the letter from a verdict into a stage in a process, and a process can be questioned, timed, and followed, while a verdict can only be resented.
The wider outcome points the same way, and we report it in the next section. For now, hold the design lesson: fairness is partly an information property. Digital.gov, in its 2025 plain language guidance, advises writers to organise content around what readers need to do next, and our 16 point fairness gain is that advice measured inside an insurance letter.
A visible next step lowers the temperature of a no.
VIII Structura
The rewritten letter put the decision first and the route out second
Order is a fairness device, because readers judge a letter before they finish it.
The PR action in this study was a plain language letter rewrite, and its structure can be described plainly. The decision came first, in words a policyholder uses, followed by the reason expressed without policy jargon doing the heavy lifting. The next step box then set out what the reader could do, and the supporting detail followed for those who wanted it.
That ordering respects how letters are actually read. Most readers decide within the first lines whether the writer is being straight with them, and a letter that hides its decision teaches suspicion before it teaches anything else. The World Wide Web Consortium, in its 2025 accessibility principles, likewise expects content people can perceive and understand without specialist effort.
| Letter element | Original pattern in the tests | Rewritten pattern in the tests |
|---|---|---|
| The decision | Found late, after background | Stated first, in plain words |
| The reason | Expressed in policy terms | Explained in reader terms |
| What happens next | Absent or buried | A next step box, raising fairness 16 points |
| Measured comprehension | 37 percent | 79 percent |
fons: Aerovector research record for the comprehension figures, the next step box fairness gain, and the rewrite tested in this study.
Put the answer where the anxiety is.
IX Consilium
Test the letter with readers before the next policyholder has to
Twenty two tests are cheaper than one pattern of avoidable complaints.
Start with your highest volume denial letter, not your most elegant one, and test it the way we did: real readers, a comprehension task, and no help from the author. Record the figure honestly even when it embarrasses the drafting team, because a 37 percent starting point is information, not blame. Then benchmark it, using our record as a reference: a Benchmark Claims Clarity median of 46 percent and a top quartile of 71 percent.
Rewrite against the failure kinds and test again with fresh readers. Advisers reading how to choose a PR agency for financial advisory firms will recognise the final discipline: document the improvement so the next template inherits it. Our support for this work is set out in our public relations services.
The outcome in our study was complaint intent down 21 percent after the rewrite. Treat that as the prize for clarity work, measured in this insurer and this letter, and as a reason to test your own rather than a figure to promise your board. Your letter may gain more or less, and the test, not the hope, is what you report.
Test the letter, not the reader patience.
X Quaesita
Questions
What did the claims letter clarity test measure?
We measured whether readers could understand a denial letter, using 22 user tests with an insurer letter. Readers had to explain the decision back accurately, which is a stricter test than a readability score. We then tested a plain language rewrite of the same decision in the same way.
Why was comprehension only 37 percent for the original letter?
Test readers struggled to find the decision, follow a reason expressed in policy terms, and identify what they could do next. The letter functioned as a record of a decision rather than an explanation of it. Those failures left most readers unable to state reliably what they had been told.
How did the rewrite reach 79 percent comprehension?
The plain language letter rewrite stated the decision first, explained the reason in reader terms, and added a clearly marked next step. The policy decision itself did not change between versions. Comprehension rising from 37 percent to 79 percent therefore prices the writing, not a softer decision.
What is a next step box, and why did it raise fairness by 16 points?
It is a short panel in the letter stating what the reader can do after a denial, such as how to question it or supply more evidence. It raised fairness by 16 points in our tests because readers judge a decision with a visible route forward differently from a dead end. The box changed no outcome, only the reader sense that a process existed.
How do these clarity figures sit against the benchmarks?
Our record sets the Benchmark Claims Clarity median at 46 percent and the top quartile at 71 percent. The original letter, at 37 percent, sat below the median. The rewrite, at 79 percent, sat above the top quartile, which is why we treat structure and wording as primary variables rather than finishing touches.
Did complaint intent really fall by 21 percent?
Yes, complaint intent down 21 percent is the outcome recorded after the plain language rewrite in this study. We report it for this insurer and this letter, tested with these readers. It is evidence that clarity changes behaviour, not a rate we promise for other letters.
How many user tests does a letter need before we trust the result?
Our study used 22 user tests, enough to expose consistent failure points and to confirm the rewrite with fresh reading. Your letter may need more tests if its audience is unusually varied. What matters is testing comprehension with real readers rather than approving drafts by committee.
How this guide was built
This post reports Aerovector's own research record for a claims letter clarity test for an insurer: 22 user tests, the comprehension figures for the original and rewritten denial letters, the fairness effect of the next step box, the Benchmark Claims Clarity median and top quartile stated in the text, the plain language letter rewrite as the PR action, and the complaint intent outcome recorded after it. The study figures are Aerovector's own research record and identities are withheld, so the organisation is described only as an insurer. The study tests one letter in depth, its readers cannot represent every policyholder, and its benchmarks come from the same research record. Third party sources were used only to set context on plain language standards, insurance consumer protection, and health literacy, and each is named with its year where it appears.
Sources
- Aerovector, Claims letter clarity test research, 2026: the first-party study reported in this post, figures from the practice research record, identities withheld.
- Plain Language Action and Information Network, Federal Plain Language Guidelines, 2025. https://www.plainlanguage.gov/guidelines/
- National Association of Insurance Commissioners, consumer insurance resources and guidance, 2025. https://www.naic.org/consumer.htm
- Consumer Financial Protection Bureau, compliance resources for clear consumer financial communication, 2025. https://www.consumerfinance.gov/compliance/
- U.S. Government Publishing Office, Plain Writing Act of 2010, full statutory text, 2010. https://www.govinfo.gov/content/pkg/PLAW-111publ274/pdf/PLAW-111publ274.pdf
- Digital.gov, Plain Language guide, writing for online government audiences, 2025. https://digital.gov/guides/plain-language
- Plain English Campaign, plain English guides and clarity campaigning, 2025. https://www.plainenglish.co.uk/
- World Wide Web Consortium, Web Accessibility Initiative, accessibility principles, 2025. https://www.w3.org/WAI/fundamentals/accessibility-principles/
- Office of Disease Prevention and Health Promotion, Health Literacy Online, clear communication guide, 2025. https://health.gov/healthliteracyonline/
- International Association of Insurance Supervisors, insurance supervision and consumer protection framework, 2025. https://www.iais.org/
XI Incipe
When your hardest letter is also your clearest, Aerovector helps the decision travel
Our Message House discipline applies the same test to public statements that this study applied to a denial letter: say the decision first, give the reason in reader language, and show the next step. A claims letter, a press answer, and an AI search summary should not tell three different stories about the same decision, and Coverage Reporting shows you which version the outside world actually repeated. That consistency is what turns a difficult communication into evidence that your organisation deals straightly.
Ranking for AI search results and guaranteed media coverage sit at the centre of the Aerovector offer, and the coverage is built to generate leads. If your denial letters have never been tested with readers, send us the one you send most often through the form on our contact page, with the decision it carries and the complaints it tends to draw, and we will show you how a tested rewrite programme would be built for your claims team.
Aerovector writes the no so clearly that readers can respect it.
Additional Details
Description: Our Research: Claims letter clarity test for an Insurer: what the study measured, what it found, and what the finding changes for teams facing the same.
Published: 2026-10-25