Designing for Multilingual Users: A Simple Solution

The Center for Plain Language has created a five-step checklist will guide you through the plain language process. It will help you develop content that’s right for your organization. Step 1: Identify and describe the target audience Step 2: Structure the content to guide the reader through it Step 3: Write the content in plain language Step 4: Use information design to help readers see and understand Step 5: Work with the target user groups to test the design and content Each step has explicit advice on the logistics of making it work. The document or site works when target users can find what they need, understand what they find, and act on it confidently. https://lnkd.in/dZM2U2DZ Image Description: A cartoon-type background. In the center, a speech balloon. In the speech balloon are the words, "Five Steps to plain language." Under the speech balloon are two colored rectangles. The top one says, "Learn what they are." The bottom one is the Center for Plain Language logo, which has two speech balloons on the left. The name Center for Plain Language. And their tagline, "Make it clear."

Two people can look at the same thing - and see completely different worlds.   All because of the language they speak. Recently came across the Sapir-Whorf hypothesis - the idea that the language a person speaks shapes the way they perceive the world. If language A has a word for a concept that language B doesn’t, speakers of A are not only more likely to name it - they’re more likely to notice it. That distinction blew my mind. For example, in English we say “I’m sad” or “I’m anxious” But in some other languages, people say, “I have sadness on me”  or “I have anxiety on me.” That small linguistic shift changes how we view emotions - as something that visits you. Not something that defines you, but your current state. I believe this insight isn't even limited to spoken languages. Every craft has its own language - its own ways of seeing. Designers see symmetry, balance, hierarchy. Actors notice pauses, quirks, speech intonations. Filmmakers think in stories, frames and character arcs. In tech, the language is often built on logic and engineering - powerful, but sometimes limiting when it comes to communicating what you’ve built. Because your customers think in stories - problems, outcomes, and emotions. When we collaborate with marketing and product teams, we help expand their lens.   They bring the precision of product thinking.   We bring our sensitivity to color, movement, sound, character, emotion. 🔐 You don’t just get a video or a campaign - but a new language to tell your story with.

The Language Gap: What We Don’t See, We Don’t Understand. When we talk about language, most people think of talking. We measure how well a child speaks, how many words they use, how clear their speech sounds, and how confidently they can hold a conversation. But spoken language is only one part of the picture. Language develops in several interconnected ways: - Receptive language – understanding what we hear and read - Expressive language – how we organise and share our thoughts in words - Written language – turning ideas into structured sentences and paragraphs - Reading – decoding symbols and constructing meaning We often assume that if a child can talk, they can understand, and if they can speak, they can write. Yet these are different systems that rely on distinct skills and brain pathways. A child might have strong verbal skills but still find it difficult to follow instructions, express complex ideas, or organise their thoughts in writing. When that happens, we can mistake the struggle for disinterest or lack of effort, when in fact it is a language load problem, not a behaviour problem. Understanding the differences between verbal, receptive, expressive, and written language is more than academic. It is the key to effective teaching, therapy, and support. When we start to see language as layered and connected, we can begin to close The Language Gap.

🩺 The Urgent Need for Plain Language in PROs Over half of PRO instruments fail basic Plain Language standards, yet we expect patients to report outcomes accurately. Let’s talk about bridging the gap between accessibility, compliance, and patient trust. Plain Language is not a “nice to have” in PROs. It is an ethical and regulatory imperative for equitable, patient-centred research. While digital innovation has improved access to publications, complex and overly technical language continues to exclude the very people research aims to serve: patients and their caregivers. 💡 Why Plain Language Drives Engagement and Equity We ask patients to engage actively in their own care, yet the language of research often builds barriers instead of bridges. 🔎 Low readability: The average Flesch–Kincaid grade for so-called “Plain Language Summaries” in PRO research is 17.8, far above the recommended eighth-grade level. 🔎 Systematic exclusion: Complex wording disproportionately affects older adults and those with limited health literacy. More than 50% of cancer PROMs fail to meet plain-language criteria. 🔎 Building trust: Translating numerical PRO scores into accessible narratives strengthens comprehension and the perceived value of these measures for patients. Plain Language does not simplify science, but it humanises it. ✅ Let’s Talk Are your current PROMs truly accessible to all patient populations, or are technical barriers compromising your data quality and equity goals? Connect with me today to discuss how we can implement world-class Plain Language strategies, optimize your eCOA migration processes, and ensure your clinical trial communication is both accurate and inclusive. #PlainLanguage #PROM #PRO #eCOA #COA #ClinicalTrials #PatientEngagement #HealthLiteracy #Localization #RegulatoryCompliance #LifeSciencesTranslation #Accessibility #ClinicalResearch #HumanInTheLoop #PlainLanguageMatters

When “good English” starts to sound suspicious… A few days ago, my sister, who’s an English writer, said something that stopped me in my tracks. She said, “I avoid using em dashes (-) now, because they make my writing look like ChatGPT wrote it.” That line hit hard. We’ve officially entered a time where writing too perfectly can make you sound… not human. Where polish, grammar, and structure - the very things we worked so hard to learn - now risk making us sound artificial. For years, people have been advised to speak more effectively, write more clearly, and sound more professional. But now, many are deliberately toning down how they express - just to sound real. It’s strange, isn’t it? If you write too well, people assume it’s AI. If you sound too raw, people think you’re careless. So where does that leave us? Small shifts are happening everywhere: – Simpler words. – Fewer edits. – Sentences that breathe, even if they’re not perfect. Not to sound casual - but to sound human. Maybe perfect English isn’t the goal anymore. The goal may be to sound like someone who feels, not something that performs. Because at the end of the day, authenticity isn’t flawless. It’s imperfect. Thoughtful. And unmistakably human.

🧠 𝗧𝗿𝗮𝗻𝘀𝗹𝗮𝘁𝗲 𝗮 𝘀𝗲𝗻𝘁𝗲𝗻𝗰𝗲, 𝗺𝗶𝘀𝘀 𝘁𝗵𝗲 𝘀𝘁𝗼𝗿𝘆. Sentence-level machine translation translates individual sentences or phrases in isolation - without considering the context around them. It’s fast and practical, which makes it perfect for short texts like emails, chat messages or social media posts, where structure matters less than speed. This approach comes from phrase-based systems, which break text into small chunks, translate each piece using a database of pre-translated phrases, and then reassemble them. The process is efficient, but it sees each sentence as a self-contained island. That’s where the cracks start to show. Without context, pronouns lose their reference, tone shifts, and meaning can drift. Sentence-level MT often misses links between sentences - the “flow” that makes writing sound natural and coherent. In contrast, document-level machine translation reads the bigger picture. It uses surrounding context to resolve ambiguity, keep terminology consistent, and maintain a single voice across paragraphs. It’s slower, but closer to how humans actually read and think. Real-world example: imagine translating a product manual. One line says, “𝘗𝘳𝘦𝘴𝘴 𝘵𝘩𝘦 𝘣𝘶𝘵𝘵𝘰𝘯 𝘢𝘨𝘢𝘪𝘯 𝘪𝘧 𝘪𝘵 𝘧𝘢𝘪𝘭𝘴.” The next says, “𝘐𝘵 𝘴𝘩𝘰𝘶𝘭𝘥 𝘳𝘦𝘴𝘦𝘵.” Without context, “𝘪𝘵” may refer to the wrong device - or be ambiguous. Sentence-level MT might mistranslate “𝘪𝘵” in one of them. But a system aware of the full manual would know which device “𝘪𝘵” refers to, keeping the translation consistent. Sentence-level MT is still essential for quick, functional communication. But when the goal is clarity, tone, and trust - context wins every time. #RespectTheLocals #glocco #Gloccary #localisation #MT #MachineTranslation #languages

IF English is your second language, this prompt can help you LEARN how to improve your English. "I am a native ______ speaker, can you help me improve this email?" When you tell it what you speak natively, it can compare your expectations of sentence structure with English and teach you the principles of how the two languages are different. Don't just have AI correct your emails. Have AI teach you.

🧐 Is the source always right? Not really. And it happens more often than you’d think. Sometimes the source text isn’t even the true original but already a translation itself, and the meaning gets blurrier with every step of the process. Sometimes there are simple human or MT errors, typos, or odd phrasing that slipped through. And sometimes, technical issues like segmentation errors or problems that arise when files are imported into tools such as XTM or Smartling break the flow and make the text harder to understand. Last but not least, there are idioms, expressions, or cultural references that simply don’t work in Italian. You can’t just carry them over, or you’ll end up with something that sounds… odd. Take “I’m feeling blue.” In English, it means “I’m sad.” In Italian, if you translate it literally as “Mi sento blu,” it doesn’t just lose its meaning, it loses all logic. That’s why I always tell my clients when something feels wrong or can be improved. Because localization is not about obeying the source at all costs, it’s about making sure the final, and sometimes even the source, message truly serves its purpose. And honestly, it feels great when a client notices. When they appreciate that extra step, that little “extra mile,” to make their content stronger and more authentic. It turns a regular day into one of those I-really-love-my-job days. ____________________________________ 🧠 𝐀𝐈 𝐦𝐚𝐤𝐞𝐬 𝐢𝐭 𝐟𝐚𝐬𝐭. 𝐇𝐮𝐦𝐚𝐧𝐬 𝐦𝐚𝐤𝐞 𝐢𝐭 𝐟𝐢𝐭. AI and MT are tremendous tools. Useful, powerful, fast. Use them if they help, they give you a nice fabric. But if you want something that truly fits your story, your audience, your brand, and brings real results, you need a tailor. That is when you call me. 📩 𝐃𝐌 𝐦𝐞 𝐚𝐧𝐝 𝐥𝐞𝐭’𝐬 𝐬𝐡𝐚𝐩𝐞 𝐲𝐨𝐮𝐫 𝐯𝐨𝐢𝐜𝐞 𝐢𝐧 𝐈𝐭𝐚𝐥𝐢𝐚𝐧. 🌐 Want to learn more? Check out www.serraluisa.com

Cultural Nuance Could Be Your Missing Data Point Every dataset contains a story, a snapshot of the culture it comes from. Language is often shaped by region and tradition, and its meanings don’t always translate the same way. If that cultural context is missing from how data is interpreted, you’re missing more than information. You’re missing understanding. That’s why human-in-the-loop validation makes all the difference. At Productive Playhouse, our proven workflows include native speakers at every review stage. They do more than check for accuracy. They ensure meaning is preserved. A literal translation doesn’t always carry the same context or intent. Native speakers can catch those shifts before they become errors at scale. Whether your work spans many languages or just one, our native speaker expertise can help keep meaning intact.