Tool AT-W47REV. 2026-A
Check readability: how easy is your text to understand?
As a guide for texts meant for everyone: Flesch index (Amstad) of 60 or more, Wiener Sachtextformel no higher than level 9, LIX below 50. The formulas count sentence length and syllables for this. The readability check calculates all three, marks long sentences, long words, passive voice, nominal style and filler words, and estimates the reading time. It shows where a text snags. Whether it is understandable is for your readers to decide.
- free
- no sign-up
- Text stays in the browser
01 / CHECK
Paste a text, read off the values and the problem spots
Check · Readability of German texts
Readability formulas count words, syllables and sentence lengths. Whether a text is understandable is in the end for your readers to decide: the values are a pointer to where it snags, not a verdict on the text.
Calculates in your browser. The text is not sent and not stored.
02 / FORMULAS
Flesch, Wiener Sachtextformel and LIX: what they calculate
All three formulas rest on the same observation: long sentences and long words make a text harder. They differ in how they measure word length (syllables or letters) and in what comes out at the end. Rudolf Flesch's formula dates from 1948 and was built for English; Toni Amstad adapted it to German in 1978, mainly the factor for syllables. The Wiener Sachtextformel was designed for German from the outset. All three are in use, for example in assessing German patient information Betschart et al. 2017.
| Formula | Calculation | Result | Guide value for texts meant for everyone |
|---|---|---|---|
| Flesch (Amstad) | 180 − sentence length − 58.5 × syllables per word | usually 0 to 100, higher is easier | 60 or more |
| Wiener Sachtextformel 1 | 0.1935 × MS + 0.1672 × SL + 0.1297 × IW − 0.0327 × ES − 0.875 | school level 4 to 15 | up to 9 |
| LIX | Sentence length + percentage of words over six letters | usually 20 to 70, lower is easier | below 50 |
Sentence length and SL = words per sentence. MS = percentage of words of three or more syllables, IW = percentage of words over six letters, ES = percentage of one-syllable words. House guide values, not a standard. As of 14 November 2026.
| Value | Readability |
|---|---|
| 90 to 100 | very easy, understandable for 11-year-olds |
| 80 to 90 | easy |
| 70 to 80 | fairly easy |
| 60 to 70 | medium, understandable for 13 to 15-year-olds |
| 50 to 60 | fairly difficult |
| 30 to 50 | difficult |
| below 30 | very difficult, more for academics |
| Value | Text type |
|---|---|
| below 40 | Children's and young adult literature |
| 40 to 50 | Fiction |
| 50 to 60 | Non-fiction |
| above 60 | Specialist literature |
Four Wiener Sachtextformeln
In 1984 Richard Bamberger and Erich Vanecek published four versions. The first uses all four features, the second leaves out the one-syllable words, the third and fourth work only with multi-syllable words and sentence length. The result is a school level: 4 means very easy, 15 very difficult, and from 12 it is more a difficulty level than a school year. The check shows all four and uses the first for the traffic light.
| Version | MS | SL | IW | ES | Constant |
|---|---|---|---|---|---|
| WSTF 1 | 0,1935 | 0,1672 | 0,1297 | -0,0327 | -0,8750 |
| WSTF 2 | 0,2007 | 0,1682 | 0,1373 | – | -2,7790 |
| WSTF 3 | 0,2963 | 0,1905 | – | – | -1,1144 |
| WSTF 4 | 0,2744 | 0,2656 | – | – | -1,6930 |
Coefficients according to Bamberger and Vanecek (1984), matching in two independent sources; we have not seen the book itself.
How the check counts
- A sentence ends at a full stop, exclamation mark or question mark, but not after abbreviations (z. B., d. h., Nr., Dr., ca.), after ordinal numbers (“am 3. Oktober”), or in dates and times (14.30 Uhr). Usw. and etc. may end a sentence.
- Abbreviations count as they are read aloud: “z. B.” as two words with three syllables.
- Numbers, web addresses and e-mail addresses do not count as words. Abbreviations in capitals are spelled out (PDF has three syllables), unless they are spoken as a word (WLAN).
- The check counts syllables by spelling rules: diphthongs such as ei, au, eu once, cases such as Fe-ri-en, Mu-se-um and spe-zi-ell separately, plus a list of English words such as Website and Team. With rare foreign words and names it can be one syllable off.
- Single lines without punctuation at the end (headings, bullet points, buttons) are not counted while the option is on. They would pull the sentence length down.
03 / EXAMPLE
Before and after: the same message, two texts
The scores are based on the original German texts.
Before, in officialese:
Zur Gewährleistung einer reibungslosen Inbetriebnahme Ihrer Datensicherung wird seitens unserer Mitarbeiter im Vorfeld eine umfassende Bestandsaufnahme der vorhandenen Infrastruktur durchgeführt. Die Durchführung der Wiederherstellungstests erfolgt grundsätzlich in regelmäßigen Abständen, wobei die Ergebnisse dokumentiert und dem Auftraggeber im Rahmen eines Abschlussberichts zur Verfügung gestellt werden. Bei Feststellung von Abweichungen wird unverzüglich eine Benachrichtigung der zuständigen Person veranlasst.After, with verbs and short sentences:
Bevor wir Ihre Datensicherung einrichten, sehen wir uns Ihre Geräte und Ihr Netzwerk an. Danach prüfen wir regelmäßig, ob sich die Daten zurückholen lassen. Jedes Ergebnis schreiben wir auf. Am Ende bekommen Sie einen Bericht. Fällt uns etwas auf, sagen wir Ihnen sofort Bescheid.| Value | before | after |
|---|---|---|
| Words | 56 | 44 |
| Sentences | 3 | 5 |
| Words per sentence | 18,7 | 8,8 |
| Syllables per word | 2,66 | 1,84 |
| Flesch (Amstad) | 5,7 | 63,5 |
| Wiener Sachtextformel 1 | 17,8 | 5,3 |
| LIX | 74,0 | 31,5 |
| Sentences with passive voice | 3 | 0 |
| Long words | 7 | 1 |
Fig. 1 · Flesch (Amstad), levels according to Flesch (1948). Values from the check, as of 14 November 2026.
Little has changed in the content and a lot in the form. “To ensure the smooth commissioning” became a sentence with “we” and a verb. The three passive sentences do not say who acts; afterwards, who acts is stated in every sentence. And “notification of the person responsible” became “we tell you”. The formulas therefore jump from 6 to 64 points and from level 18 to level 5. More important is that a customer understands the second text on first reading.
04 / MARKINGS
Long sentences, long words, passive voice, nominal style, filler words
| Marking | From when | What you can do |
|---|---|---|
| Long sentence | from 20 words, very long from 30 | Split it, put asides into a sentence of their own, bring the verb forward |
| Long word | a word part of 14 letters or more | Split with a hyphen or rephrase: Daten-Sicherung, “how we back up” |
| Passive voice | “werden” with a past participle in the same sentence | Say who acts: “Wir schicken die Rechnung” instead of “Die Rechnung wird verschickt” |
| Nominal style | 3 or more nouns ending in -ung, -heit, -keit, -ation, -tät in one sentence | Bring the verb back: “wenn wir prüfen” instead of “bei Durchführung der Prüfung” |
| Filler word | Words such as eigentlich, ziemlich, durchaus, sehr | Delete them if the sentence says the same without them |
Thresholds and word lists are house guide values. The check recognises passive voice and nominal style by patterns, not by grammatical analysis.
Why long sentences slow readers down
A sentence costs the more, the longer the reader has to keep something announced in mind and the further apart words that belong together stand Gibson 1998. In German this is the sentence bracket (Satzklammer): “Wir werden Ihnen die Unterlagen nach Abschluss der Prüfung durch unsere Buchhaltung zusenden.” Only at the end does it say what happens. Bring the verb forward or make two sentences of it. In an experiment with 30 cloze tests, comprehension dropped mainly because of sentences of more than about 140 characters Mikk 2008; in German that is roughly 20 words, which is why the check marks from there. The strip above the marked text shows every sentence as a bar. If it looks like a comb with teeth of equal height, the text is getting monotonous; variety is allowed.
Long words and compounds
German glues words together, and that costs reading time. In eye-tracking studies, gaps between the parts of long compounds helped with breaking them up but got in the way of understanding Inhoff et al. 2000. That is why you split with a hyphen, not with a space. People with low reading ability recognised split compounds faster Pappert and Bock 2019, and for dyslexia shorter words improved comprehension Rello et al. 2013. In the comparison group without dyslexia there was no clear effect: short words help above all those who find reading hard.
Passive voice and nominal style
Passive sentences are misunderstood more often, especially when it is a matter of who does what Ferreira 2003. For deadlines, duties and costs that is no longer a question of style. The check looks for a form of “werden” and a past participle in the same sentence; it deliberately does not detect the stative passive with “sein” (“ist geschlossen”), because that is often simply a state. With nominal style the picture is less clear. Native speakers remembered the important points better when verbs stood in place of nominalisations; for non-native speakers nominal style partly worked well Spyridakis and Isakson 1998. The check therefore marks it only when there are 3 such words in a sentence.
Reading time
Adults read non-fiction silently at around 238 words per minute, most of them between 175 and 300 Brysbaert 2019. The often-quoted 300 words are too high for most. German needs about as many words as English for the same message, so the check uses the same values. For a text with many long words, the lower end is more likely.
05 / LIMITS
What readability formulas do not measure
For a first look the formulas are good: quick to calculate and easy to follow. But they have clear limits, and you should know them before you rewrite a text to fit them.
- They measure length, not familiarity. 68 German patient leaflets averaged a Flesch score of 31.6, although 69 per cent of the vocabulary was understandable to laypeople Wiesner et al. 2020. “Rechnung” (invoice) is long and familiar to everyone, “Skonto” (cash discount) is short and not clear to everyone.
- They are not suited as writing guidance. Texts simplified to fit a formula were not easier for it Davison and Kantor 1982, and using a formula as the yardstick when revising did not help Duffy and Kabance 1982. Anyone who deletes connecting words to shorten sentences makes the connections harder to follow.
- They see neither readers nor design. Prior knowledge, subheadings, lists and font size are missing from every formula Redish 2000. A test with two or three readers from the target group tells you more.
- They often disagree. On the same text, common formulas were up to five school grades apart, and a single formula up to six depending on the excerpt and software Wang et al. 2013. That is why the check shows all values with the working and warns for texts under 100 words.
- There are better models. Newer methods from language processing predict comprehension and reading speed clearly better than the classic formulas Crossley et al. 2019, which mainly count surface features Benjamin 2012. They need word lists and computing power, however, which a tool in the browser does not bring with it.
06 / PLAIN LANGUAGE
Plain Language (Einfache Sprache) and Easy Read (Leichte Sprache)
In Germany around 6.2 million adults between 18 and 64 read and write at best at the level of single sentences, which is over 12 per cent Grotlüschen and Buddeberg 2020. Add to that everyone who speaks German as a second language or reads on a phone while out and about. There are two routes for them, which are often confused.
| Plain Language | Easy Read | |
|---|---|---|
| Who it is for | Everyone who wants to understand quickly, including those with little reading practice; expressly not for readers with cognitive impairments | People with learning difficulties and everyone who needs very simple texts |
| Rules | DIN 8581-1:2024-05, building on DIN ISO 24495-1 | DIN SPEC 33429:2025-03, free |
| Type | Standard | DIN SPEC: recommendations of a consortium, not a standard |
| Typical | Short sentences, familiar words, active verbs, clear structure | Very short sentences, one statement per sentence, split-up words, its own layout |
| Required | No general obligation for businesses | Public bodies: explanations on the home page (BITV 2.0, § 4) |
For a business website, Plain Language is almost always the right yardstick. The formulas help as a control. We did not find a target figure for them in the freely available information on the standard, and the check is no substitute for an assessment under DIN 8581-1. Easy Read is a text type of its own; whether a text in Easy Read works is shown only by a test with people from the target group.
And the Barrierefreiheitsstärkungsgesetz? It is based on WCAG 2.1 at levels A and AA. The reading level criterion, under which difficult texts need a simpler version, belongs to level AAA (WCAG 3.1.5). It is still useful as a guide: the limit there is the level reached after about nine school years, hence our guide value of level 9 in the Wiener Sachtextformel. Whether your website falls under the law and what to do then is shown by the BFSG check.
07 / GOOGLE AND AI
Readability, Google and AI answers
A readability score is not a ranking factor that Google names. The guide to helpful content contains no formula; Google even advises against writing to a particular word count and asks for content made first for people (Google Search Central). An understandable text is therefore not a trick for search. It helps visitors understand what you offer.
Whether AI assistants prefer more readable texts, we know of no reliable figures. How machines read a page is explained under AI visibility. Why a page is not found in the first place is explained in the guide Website not found on Google. The check answers only the question about the text; the overall verdict on your website, with technology, content and visibility, comes from SpotLens. More tools for individual questions on findability are in the overview of SEO and visibility.
08 / METHOD
How to make a text easier
- Decide who the readers are. Who reads the page, with what prior knowledge, on a phone or at a desk? For specialists it can be harder than for home users.
- Run the check. Paste in a whole section, at least 100 words, and look at the marked passages first, not at the number.
- Split very long sentences. Deal with sentences with a red background first. Bring the verb forward and put asides into a sentence of their own.
- Say who acts. Rebuild passive voice into active sentences with “wir” and “Sie”, turn nouns ending in -ung back into verbs.
- Split or replace long words. A hyphen in compounds, the familiar word instead of the technical term where possible.
- Test with real readers. Have two or three customers read it and ask what they would do next. That beats any formula.
How we build website texts and pages for businesses in the region is explained under Web design and at SEO in Rodewisch.
09 / RESEARCH
What the research says
17 reviewed papers underpin the thresholds and notes in the check. The full list with review notes is further down; here are the most important findings.
- Formulas count, readers understand. As writing guidance, formulas have failed Davison and Kantor 1982 and are clearly outperformed by newer models Crossley et al. 2019. The check therefore shows passages rather than just numbers.
- Several formulas, several answers. Up to five grades of difference on the same text Wang et al. 2013; German specialist texts ranged from grade 7 to 14 Betschart et al. 2017.
- Length is not everything. Leaflets with weak formula scores often had an understandable vocabulary Wiesner et al. 2020.
- What helps within the sentence. Long dependencies cost working memory Gibson 1998, passive voice is misunderstood more often Ferreira 2003, split compounds are recognised faster Pappert and Bock 2019.
- It matters especially for many readers. Over 12 per cent of adults read at best single sentences Grotlüschen and Buddeberg 2020.
10 / FOR AI ASSISTANTS
Using this tool from an AI assistant
Assistants such as Claude or ChatGPT that speak the Model Context Protocol (MCP) can call this page’s calculation directly instead of rebuilding it. The server is public, needs no sign-in and stores nothing. The answer always includes the link to this page, the date of the data and the limits of the result.
- Endpoint
https://aureum-tech.com/mcp- Tool
werkzeug_lesbarkeits_check- Transport
- Streamable HTTP, read-only
| Name | Required | Meaning |
|---|---|---|
text | yes | The German text, as plain text (no HTML), 100,000 characters at most. Paragraphs separated by line breaks. |
zeilenumbruch_ist_satzende | no | Default true: every line ends a sentence (web page text). false for text from PDFs with line breaks in the middle of sentences; then only blank lines separate. |
ueberschriften_ausschliessen | no | Default true: single lines without punctuation at the end (headings, bullet points, buttons) are left out of the formulas. |
Questions where an assistant should call it
- “How readable is this text for my website? Here it is: …”
- “What Flesch score (Amstad) and what Wiener Sachtextformel does this paragraph have?”
- “Which sentences in my cover letter are too long, and where is the passive voice?”
- “How long does this text take to read, and is it understandable enough for a broad audience?”
How the connection works and what MCP can do in your own business is explained on the page about MCP integrations.
11 / COMMON QUESTIONS
Common questions about readability
What is a good Flesch score for a German text?
On the Flesch scale (Amstad), 60 to 70 counts as medium, understandable for 13 to 15-year-olds; from 80 a text is easy, below 30 very difficult. For a website that everyone should understand, we recommend at least 60. Our example in officialese scores 6, the same message in short sentences 64. Specialist texts are often between 30 and 50, and that is fine for specialists.
What is the Wiener Sachtextformel?
A readability formula for German non-fiction by Richard Bamberger and Erich Vanecek (1984). From the share of long and multi-syllable words and the sentence length it calculates a school level between 4 (very easy) and 15 (very difficult); from 12 it is more a difficulty level than a school year. There are four versions; the first uses all four features. For texts meant for everyone we advise level 9 at most.
What does the LIX value tell me?
The readability index LIX by Carl-Hugo Björnsson adds the average sentence length and the percentage of words with more than six letters. For German texts: below 40 like children's and young adult books, 40 to 50 like fiction, 50 to 60 like non-fiction, above 60 like specialist literature. For website texts meant for everyone we recommend below 50.
What is the maximum number of words a sentence should have?
There is no fixed limit. The check marks sentences of 20 words or more as long and of 30 or more as very long. In an experiment with pupils, comprehension dropped mainly for sentences of more than about 140 characters, which in German is roughly 20 words. A text made up entirely of equally short sentences does not read well either; the mix is what counts.
Does a good readability score improve my Google ranking?
That has not been shown. In its guide to helpful content, Google names no readability formula and advises against writing to a word count. It asks for content written for people. An understandable text therefore helps readers first; whether a page is found depends on many other things, which SpotLens checks as a whole.
What is the difference between Plain Language and Easy Read?
Plain Language (Einfache Sprache) is for everyone who wants to understand a text quickly, including people with little reading practice. For German it is governed by DIN 8581-1 (May 2024 edition), which is expressly not intended for readers with cognitive impairments. Easy Read (Leichte Sprache) is for exactly those readers and follows its own rules, such as very short sentences and split-up words; recommendations are given in the free DIN SPEC 33429 (March 2025).
Does my website have to be in Plain Language because of the BFSG?
Not as of our review on 14 November 2026. The Barrierefreiheitsstärkungsgesetz (BFSG, German Accessibility Act) is based on WCAG 2.1 at levels A and AA; the reading level criterion (3.1.5) belongs to level AAA. Explanations in Easy Read are required of public bodies by the BITV 2.0. Whether your website falls under the BFSG at all is answered by our BFSG check.
Where does the readability check stop?
It counts sentences, words and syllables and looks for patterns of passive voice, nominal style and filler words. It cannot tell whether the words are familiar to your customers, whether the structure holds up or whether the page says the right thing. We clarify that with you: we write website texts for businesses in Rodewisch and the Vogtland and test them with readers from your target group.
12 / STUDIES AND SOURCES
What the tool is based on
The reference values and notes above rest on these 17 scientific papers. Each has been checked via its DOI or the publisher’s page (last on 14 November 2026): it exists, it has not been retracted, and its finding still holds for the question today.
Davison, A.; Kantor, R. N. (1982): On the Failure of Readability Formulas to Define Readable Texts: A Case Study from Adaptations. Reading Research Quarterly 17 (2), 187–209
Finding Using texts simplified for weaker readers, the authors show that formulas as writing guidance lead people astray. What actually makes a text readable, they say, has first to be established through experiments with readers.
In the tool The tool shows where a text snags and expressly advises against rewriting a text to hit a target number.
doi.org/10.2307/747483 · Journal article
Duffy, T. M.; Kabance, P. (1982): Testing a readable writing approach to text revision. Journal of Educational Psychology 74 (5), 733–748
Finding In experiments with revised texts, a readability formula was not suited as a yardstick for writing, even when the writers were not deliberately writing to the formula.
In the tool That is why the check marks passages in the text instead of just giving a number; changes are made to the sentence, not to the value.
doi.org/10.1037/0022-0663.74.5.733 · Journal article
Redish, J. (2000): Readability formulas have even more limitations than Klare discusses. ACM Journal of Computer Documentation 24 (3), 132–137
Finding Literature review: the formulas come from school books for children, overlook differences between readers as well as content, design and signposting, and count only what is easy to count. A test with real readers tells you more.
In the tool This underpins the section on the limits and the advice to have two or three customers proofread important pages.
doi.org/10.1145/344599.344637 · Journal article
Benjamin, R. G. (2012): Reconstructing Readability: Recent Developments and Recommendations in the Analysis of Text Difficulty. Educational Psychology Review 24 (1), 63–88
Finding Review of two decades of research: the classic formulas have long been criticised, above all because they take surface features as a substitute for comprehension processes and are used to generate texts.
In the tool It explains why the check measures word and sentence length but says nothing about the coherence, prior knowledge and logic of a text.
doi.org/10.1007/s10648-011-9181-8 · Review
Crossley, S. A.; Skalicky, S.; Dascalu, M. (2019): Moving beyond classic readability formulas: new methods and new models. Journal of Research in Reading 42 (3–4), 541–561
Finding Using many readers' judgements of comprehension and reading speed, the authors developed models from linguistic features. These predicted comprehension and reading speed clearly better than classic formulas such as Flesch-Kincaid.
In the tool The tool gives an honest assessment: the classic formulas are easy to follow and quick, but not the most accurate measure.
doi.org/10.1111/1467-9817.12283 · Journal article
Wang, L.-W.; Miller, M. J.; Schmitt, M. R.; et al. (2013): Assessing readability formula differences with written health information materials: Application, results, and recommendations. Research in Social and Administrative Pharmacy 9 (5), 503–516
Finding Applied to the same text, common formulas were up to five school grades apart. Depending on the text excerpt, formatting and software, a single formula varied by up to six grades.
In the tool That is why the check shows all formulas side by side with the working, discloses the counting rules and warns for short texts.
doi.org/10.1016/j.sapharm.2012.05.009 · Journal article
Betschart, P.; Zumstein, V.; Hasan Ali, O.; et al. (2017): Readability Assessment of Patient Education Material Published by German-Speaking Associations of Urology. Urologia Internationalis 100 (1), 79–84
Finding German-language patient information from four professional societies was assessed with the Amstad test, G-SMOG, Wiener Sachtextformel and LIX. The grade-based formulas gave grades 7 to 14, with clear differences between the publishers.
In the tool It shows that exactly these German formulas are used in research, and that specialist texts are often far above the target.
doi.org/10.1159/000480095 · Journal article
Wiesner, M.; Zowalla, R.; Pobiruchin, M. (2020): The Difficulty of German Information Booklets on Psoriasis and Psoriatic Arthritis: Automated Readability and Vocabulary Analysis. JMIR Dermatology 3 (1), e16095
Finding 68 German leaflets averaged a Flesch score of 31.6 and a fourth Wiener Sachtextformel of 11.8. At the same time, 69 per cent of the vocabulary was understandable to laypeople.
In the tool It supports the point made in the text that the formulas measure only length: a text can have familiar words and still score badly.
doi.org/10.2196/16095 · Journal article
Brysbaert, M. (2019): How many words do we read per minute? A review and meta-analysis of reading rate. Journal of Memory and Language 109, 104047
Finding Meta-analysis of 190 studies: adults read non-fiction silently at around 238 words per minute, most of them between 175 and 300. Other languages can be estimated from the number of words they need for the same message; German is close to English in this.
In the tool The reading time is calculated at 238 words per minute and gives the range of 175 to 300.
doi.org/10.1016/j.jml.2019.104047 · Review
Mikk, J. (2008): Sentence length for revealing the cognitive load reversal effect in text comprehension. Educational Studies 34 (2), 119–127
Finding In 30 cloze tests with 17 to 18-year-olds, comprehension fell with the share of sentences above a limit, which was clearest at around 140 characters. Sentences of 130 to 150 characters fitted best.
In the tool Supports marking long sentences from 20 words; 140 characters is roughly 20 words in German.
doi.org/10.1080/03055690701811164 · Journal article
Gibson, E. (1998): Linguistic complexity: locality of syntactic dependencies. Cognition 68 (1), 1–76
Finding The longer a reader has to hold an announced sentence element in memory and the further apart words that belong together stand, the more the sentence costs. Sentences are easier when the heavy parts come at the end.
In the tool It informs the advice on the German sentence bracket: verb early, long asides at the end or in a sentence of their own.
doi.org/10.1016/S0010-0277(98)00034-1 · Journal article
Ferreira, F. (2003): The misinterpretation of noncanonical sentences. Cognitive Psychology 47 (2), 164–203
Finding Passive sentences were frequently and systematically misunderstood, that is, who does what, especially when the statement was implausible. Readers often process sentences only as precisely as seems necessary at the time.
In the tool Justifies the passive voice marking: who acts should be in the sentence, especially for duties and deadlines.
doi.org/10.1016/S0010-0285(03)00005-7 · Journal article
Spyridakis, J. H.; Isakson, C. S. (1998): Nominalizations vs. Denominalizations: Do They Influence What Readers Recall?. Journal of Technical Writing and Communication 28 (2), 163–188
Finding Technical texts with verbs instead of nominalisations helped native speakers remember the important points. For non-native speakers, by contrast, nominal style partly worked well.
In the tool That is why the check marks nominal style only from three nominalisations in one sentence and calls it a question of readership.
doi.org/10.2190/01HD-MHU1-QNX9-R3YE · Journal article
Inhoff, A. W.; Radach, R.; Heller, D. (2000): Complex Compounds in German: Interword Spaces Facilitate Segmentation but Hinder Assignment of Meaning. Journal of Memory and Language 42 (1), 23–50
Finding Spaces between the parts of three-part German compounds shortened naming times and gaze duration early in reading, but later got in the way of grasping the meaning.
In the tool Explains why long compounds slow readers down and why you split them with a hyphen rather than tearing them apart with spaces.
doi.org/10.1006/jmla.1999.2666 · Journal article
Pappert, S.; Bock, B. M. (2019): Easy-to-read German put to the test: Do adults with intellectual disability or functional illiteracy benefit from compound segmentation?. Reading and Writing 33 (5), 1105–1131
Finding Adults with intellectual disability or low reading ability recognised split compounds faster than unsplit ones, whether or not the meaning was transparent. The Easy Read rule of splitting compounds stood up to the experiment.
In the tool It underpins the note on long words and the section on Easy Read (middle dot, hyphen).
doi.org/10.1007/s11145-019-09995-y · Journal article
Rello, L.; Baeza-Yates, R.; Dempere-Marco, L.; et al. (2013): Frequent Words Improve Readability and Short Words Improve Understandability for People with Dyslexia. Human-Computer Interaction – INTERACT 2013, Lecture Notes in Computer Science, pp. 203–219
Finding Eye-tracking study with 46 people: more frequent words let people with dyslexia read faster, shorter words helped them understand better. In the comparison group there was no clear effect.
In the tool Behind the marking of long words and the advice to choose familiar words over rare ones; the benefit is mainly for weaker readers.
doi.org/10.1007/978-3-642-40498-6_15 · Conference paper
Grotlüschen, A.; Buddeberg, K. (eds.) (2020): LEO 2018. Leben mit geringer Literalität. wbv Publikation, Bielefeld (Open Access)
Finding Representative survey of 7,192 adults aged 18 to 64: over 12 per cent, around 6.2 million people, read and write at best at the level of single sentences and mostly very slowly.
In the tool Shows for how many readers short sentences and familiar words make the difference, and supports the section on Plain Language.
doi.org/10.3278/6004740w · Research report
Standards, authorities and manufacturer data
- Flesch, R. (1948): A new readability yardstick. Journal of Applied Psychology 32 (3), 221–233 · American Psychological Association (basis of the formula, no abstract)
- Lesbarkeitsindex: Flesch nach Amstad und Wiener Sachtextformeln · Wikipedia (German), formulas by Amstad 1978 and Bamberger and Vanecek 1984
- Readability index LIX by Björnsson, with levels for German texts · Lenhard and Lenhard, Psychometrica
- DIN 8581-1:2024-05 Einfache Sprache – Anwendung für das Deutsche – Teil 1: Sprachspezifische Festlegungen · DIN Media (2024-05 edition, current)
- DIN SPEC 33429:2025-03 Empfehlungen für Deutsche Leichte Sprache · DIN Media (2025-03 edition, current, free)
- BITV 2.0 § 4: explanations in German Sign Language and in Easy Read · Federal Ministry of Justice (Bundesministerium der Justiz), gesetze-im-internet.de
- WCAG 2.2, Success Criterion 3.1.5 Reading Level (AAA) · W3C
- Creating helpful, reliable, people-first content · Google Search Central
13 / MORE TOOLS
Still have a question?
- AT-W42 · Businesses
Snippet preview
How does your page appear on Google? Title and description with pixel width, for desktop and mobile. - AT-W43 · Businesses
Schema generator
Structured data for your business: LocalBusiness, opening hours and FAQ as JSON-LD, ready to paste in. - AT-W44 · Businesses
Business Profile check
Is your Google Business Profile complete? What is missing, sorted by impact. - AT-W45 · Businesses
NAP check
Are your name, address and phone number the same everywhere? Compare entries from directories. - AT-W46 · Businesses
Keyword sorter
Which search terms belong together, and what is the searcher after? Lists sorted by topic and search intent. - AT-W48 · Businesses
robots.txt tester
Is Googlebot allowed to read this page? Check robots.txt against RFC 9309, rule by rule.
The check shows where a text snags. You still have to rewrite it yourself, or leave it to us: we write website texts that your customers understand and search engines classify well, for businesses in Rodewisch and across the Vogtland.
+49 3744 4399760[email protected]Mon–Fri 8:00–17:00 · Straße der Jugend 5, 08228 Rodewisch