Reading time calculator for tooltips and tours

Reading time is the number of words divided by reading speed, measured in words per minute (wpm):

Default speeds come from the IReST study (International Reading Speed Texts, Trauzettel-Klosinski, Dietz et al., 2012), which measured adults reading standardized, equivalent texts in 17 languages: English 228 wpm, Spanish 218, Russian 184, Portuguese 181. IReST texts were read aloud; silent reading is somewhat faster (a 2019 meta-analysis by Brysbaert puts English non-fiction at about 238 wpm), but interface text is read in fragments and under distraction, so the IReST figures are a sensible conservative default. Words per minute differ between languages largely because word length differs — the same message takes fewer, longer words in Russian than in English. Indonesian, Filipino and Kazakh are not covered by IReST; the calculator uses 180, 180 and 165 wpm as estimates and says so.

Two caveats matter for UI text:

• People skim interfaces. A tooltip is read in fragments between actions, often on a small screen and while the user is thinking about their own task. Real reading is slower per word than in the lab, and much of the text is skipped — both reasons to write less, not to count on the average.
• Reading is not the whole interaction. Before reading, the user notices the hint and orients to the element it points at; after reading, they find and press a button. The "notice and click" field adds this per step to the total tour time; two seconds is a conservative default.

There is no official standard for tooltip length. The limits in this calculator are Clew's editorial guideline, based on how onboarding hints are used in product interfaces and aligned with our tooltip copy checker:

A step is marked Long when it exceeds a limit by up to 50 % and Too long beyond that. Words, characters and seconds are checked separately: long words in German or Russian can break the character limit while the word count looks fine, and a small bubble on mobile is limited by characters, not words.

Why short matters in onboarding

• Hints interrupt a task. The user came to do something; every second spent reading delays it. One idea per step keeps the interruption short.
• Long bubbles cover the interface. A 60-word tooltip hides the very element it explains, especially on laptops and phones.
• Long tours lose people. Each extra step is another chance to click "Skip". If the flow has more than five steps, move later steps into a checklist the user opens when ready, or show a second tour when they reach that feature.

1. Start with the action. "Click New project to start" beats "In order to get started with your work, you can create a new project by clicking the button".
2. One step, one idea. If a step has two sentences about different things, it is two steps — the split suggestion shows where the sentence boundaries are.
3. Cut what the interface already says. Don't repeat the button label or describe what the user can see.
4. Move details out. Explanations, edge cases and "you can change this later" belong in help docs linked from the step, not in the bubble.
5. Use the button as copy. "Invite teammates" as a button label replaces a sentence telling the user to invite teammates.
6. Check readability and tone. Plain words and active voice read faster; the readability checker shows long sentences and hard words.

1. Trauzettel-Klosinski S., Dietz K., IReST Study Group. Standardized Assessment of Reading Performance: The New International Reading Speed Texts IReST. Investigative Ophthalmology & Visual Science, 53(9), 2012 — reading speeds for 17 languages.
2. Brysbaert M. How many words do we read per minute? A review and meta-analysis of reading rate. Journal of Memory and Language, 109, 2019 — silent reading rates for non-fiction and fiction.
3. Nielsen Norman Group. How Users Read on the Web; Tooltip Guidelines — scanning behavior and tooltip content.
4. Clew. Length guideline for in-app onboarding copy (tooltips, tour steps, modals, checklists) — editorial guideline used by this calculator.