📌 Note: This article summarises and applies key takeaways from the Quantitative and Qualitative Empirical Social Research class taught by Susanne Sackl-Sharif, breaking it down and linking it to the context of design work so that readers can implement it in practice. In designing online surveys, it's not just about collecting statistics; it's a gateway to understanding the "heart of the user." However, the main problem designers face is sending out surveys and not getting the desired responses, or people closing the survey midway. 1. Set clear goals Each stage of design requires different answers. Before starting to ask questions, you must first answer for yourself why you are doing this. ⭐️ Before designing: Focus on finding Pain Points and existing behaviours (e.g., "What problems do people usually encounter when making a payment?") ⭐️ During the design phase: Test ideas or layouts (e.g., "Do people understand this layout?") ⭐️ After design: Measure the ease of use and satisfaction after actual use (e.g., is this version of the UX/UI easier to use?). 💡 Additional explanation for designers: Setting good goals must always come with a "hypothesis." Instead of just saying, "I want to know if people like the app," change it to, "We believe that if we move the payment button to the bottom, users will complete their transactions faster." Having a clear hypothesis will help you design questions to "confirm" or "refute" your thoughts accurately and avoid getting sidetracked by minor details that are not related to the main goal. 2. Design questions to be "short, concise, and non-leading." Design questions to be "short, concise, and non-leading." The heart of a good survey is conciseness. ⭐️ Screening questions: Filter out non-target groups from the very first questions to avoid skewed data. ⭐️ Stop using leading questions: Change from "Do you think this app is easy to use?" "On a scale of 1-5, how would you rate the ease of use of this app?" ⭐️ Limit open-ended questions: Open-ended questions that require typing can indeed provide deep insights, but they should not exceed 2-3 questions. Most people are reluctant to type, and this may significantly reduce the completion rate. 💡 Additional explanation for designers: The key technique is to use the Funnel Approach structure. Start with broad, easy-to-answer questions to warm up and then gradually delve into more specific questions. Additionally, the Logic Jump function should be utilised effectively. For example, if a user answers "I have never used feature A," the system should skip to the next question to avoid frustrating the user with irrelevant questions.