User Satisfaction Surveys

Drawing from years of my experience designing surveys for my academic projects, clients, along with teaching research methods and Human-Computer Interaction, I've consolidated these insights into this comprehensive guideline. Introducing the Layered Survey Framework, designed to unlock richer, more actionable insights by respecting the nuances of human cognition. This framework (https://lnkd.in/enQCXXnb) re-imagines survey design as a therapeutic session: you don't start with profound truths, but gently guide the respondent through layers of their experience. This isn't just an analogy; it's a functional design model where each phase maps to a known stage of emotional readiness, mirroring how people naturally recall and articulate complex experiences. The journey begins by establishing context, grounding users in their specific experience with simple, memory-activating questions, recognizing that asking "why were you frustrated?" prematurely, without cognitive preparation, yields only vague or speculative responses. Next, the framework moves to surfacing emotions, gently probing feelings tied to those activated memories, tapping into emotional salience. Following that, it focuses on uncovering mental models, guiding users to interpret "what happened and why" and revealing their underlying assumptions. Only after this structured progression does it proceed to capturing actionable insights, where satisfaction ratings and prioritization tasks, asked at the right cognitive moment, yield data that's far more specific, grounded, and truly valuable. This holistic approach ensures you ask the right questions at the right cognitive moment, fundamentally transforming your ability to understand customer minds. Remember, even the most advanced analytics tools can't compensate for fundamentally misaligned questions. Ready to transform your survey design and unlock deeper customer understanding? Read the full guide here: https://lnkd.in/enQCXXnb #UXResearch #SurveyDesign #CognitivePsychology #CustomerInsights #UserExperience #DataQuality

The MOST critical metric you can use to measure customer satisfaction: (This changed everything for my company) We had a daily deal site with 2 million users. Sounds great, right? But about 18 months in we had a massive problem: → Customer satisfaction was TANKING (we were in the daily-deals business, largest Groupon competitor) Why? Our customers weren't getting the same experience as full-paying customers. They were treated as “coupon buyers”, so they: - Had long wait-times - Didn't get the same food - Got given the cr*ppy tables at the back They went for the full service and they got very low-quality service. And it was KILLING our business model. We tried everything - customer service calls, merchant meetings, forums. Nothing worked. Then I learned about NPS (Net Promoter Score) at EO and MIT Masters. It was an ABSOLUTE revelation. NPS isn't a boring survey asking "How happy are you with our service?" It's way more powerful. It asks, on a simple scale of 0-10: → "How likely are you to recommend this service to a friend or colleague?" 10-9 → Promoters (Nice!) 8-7 → Passive (no need to do anything) 6-0 → Detractors (fix this NOW) It’s such a simple shift on our end and so easy to respond on the customer end: “Hey, would you recommend me or not, out of 10?” “Hm, 7.” “Ok, thank you” — that’s it. Simple reframe, massive impact. We implemented it immediately. But here's the real gold: → We contacted everyone (one-on-one customer service) who used our service and provided a NPS score. They scored us less than 6? - Give them gift cards - Interview them to make them feel heard - Do ANYTHING to flip detractors into promoters Because if they’re scoring you less than 6, they’re actually HARMING your business. These are going to be like e-brakes in your company. NPS became our most important metric, integrated into everything we did. The results? - Improved customer satisfaction - Increased repeat business and customer LTV - Lower CAC (because happy customers = free marketing) - Higher AOV (people were willing to spend more) But it's not just about the numbers. It's about understanding WHY people aren't recommending you and fixing it fast. (Another great feature is that people can also add comments to get some real feedback, but just using the number is POWERFUL). If you're not using NPS, stop what you're doing and implement it tonight. Seriously. And if you are already using it? Double down on those 0-6 scores. Turning your detractors into promoters is where the real growth potential lies. Remember: in business, what gets measured gets managed. And NPS is the ultimate measure of how satisfied your customers REALLY are. So, what's your score? — Found value in this? Repost ♻️ to share to your network and follow Ignacio Carcavallo for more like this!

As UX researchers, we often rely on survey totals. We sum up Likert scale responses across a few items and call it a metric - satisfaction, usability, engagement, trust. It’s fast, familiar, and widely accepted. But if you’ve ever questioned whether a survey is truly capturing what matters, that’s where Item Response Theory (IRT) steps in. IRT is more than just a statistical model - it’s a smarter way to design, evaluate, and optimize questionnaires. While total scores give you a general snapshot, IRT gives you the diagnostic toolkit. It shifts your focus from just what the total score is to how each question behaves across different user types. Instead of treating every item as equally valuable, IRT assumes that each question has its own characteristics - its own difficulty level, its ability to discriminate between users with different trait levels (like low vs. high satisfaction), and even its tendency to generate noise. It mathematically models the likelihood of a particular response based on the person’s underlying trait (e.g., engagement) and the specific properties of that item. This lets you see which items are doing real work - and which ones are just adding bloat. Let’s say you’re trying to measure perceived product enjoyment. You include five questions. One of them - "I enjoy using this product" - is endorsed by nearly everyone. Another one - "This product makes me feel inspired" - gets more varied responses. Under IRT, the first item would be flagged as too easy; it doesn’t help you separate highly engaged users from moderately engaged ones. The second item, if it cleanly differentiates users with different enjoyment levels, would be seen as high in discrimination power. That’s the kind of insight you won’t get from a simple average. One of the biggest advantages of IRT is that it allows you to assess not just people’s responses, but the quality of the items themselves. You can identify and remove redundant or low-informative questions, focus your surveys to measure what matters most, and retain high precision with fewer items. This is a huge win for both survey respondents and UX researchers, especially when you're working in product environments where every question has to earn its place. IRT also enables more advanced applications. You can build adaptive surveys- ones that tailor themselves in real time to each participant. You can create item banks that offer equivalent measurement across time or populations. And you can track individual-level changes in UX perceptions over time more reliably, which is something traditional scoring methods often miss. I use IRT models to analyze UX questionnaires in my own work, especially when I want to make sure each item is pulling its weight. It also leads to clearer communication with designers, PMs, and engineers, because I can show why a certain item matters or doesn’t, backed by data that makes sense.

✅ Survey Design Cheatsheet (PNG/PDF). With practical techniques to reduce bias, increase completion and get reliable insights ↓ 🚫 Most surveys are biased, misleading and not actionable. 🤔 People often don’t give true answers, or can’t answer truthfully. 🤔 What people answer, think and feel are often very different things. 🤔 Average scores don’t speak to individual differences. ✅ Good questions, scale and sample avoid poor insights at scale. ✅ Industry confidence level: 95%, margin of error 4–5%. ✅ With 10.000 users, you need ≥567 answers to reduce sample bias. ✅ Randomize the order of options to minimize primacy bias. ✅ Allow testers to skip questions, or save and exit to reduce noise. 🚫 Don’t ask multiple questions at once in one single question. 🤔 For long surveys, users regress to neutral or positive answers. 🚫 The more questions, the less time users spend answering them. ✅ Shorter is better: after 7–8 mins completion rates drop by 5–20%. ✅ Pre-test your survey in a pilot run with at least 3 customers. 🚫 Avoid 1–10 scales as there is more variance in larger scales. 🚫 Never ask people about their behavior: observe them. 🚫 Don’t ask what people like/dislike: it rarely matches behavior. 🚫 Asking a question directly is the worst way to get insights. 🚫 Don’t make key decisions based on survey results alone. Surveys aim to uncover what many people think or feel. But often it’s what many people *think* they think or feel. In practice, they aren’t very helpful to learn how users behave, what they actually do, if a product is usable or learn specific user needs. However, they do help to learn where users struggle, what user’s expectations are, if a feature is helpful and to better understand user’s perception or view. But: designing surveys is difficult. The results are often hard to interpret and we always need to verify them by listening to and observing users. Pre-test surveys before sending out. Check if users can answer truthfully. Review the sample size. Define what you want to know first. And, most importantly, what decisions you will and will not make based on the answers you receive. --- ✤ Useful resources: Survey Design Cheatsheet (PNG, PDF), by yours truly https://lnkd.in/ez9XQAk3 A Big Guide To Survey Design, by H Locke https://lnkd.in/eJWRnDRi How to Write (Better) Survey Questions, by Nikki Anderson, MA https://lnkd.in/eHpzr-Q6 Survey Design Guide, by Maze https://lnkd.in/e4cMp5g5 Why Surveys Are Problematic, by Erika Hall https://lnkd.in/eqTd-7xM --- ✤ Books ⦿ Just Enough Research, by Erika Hall ⦿ Designing Surveys That Work, by Caroline Jarrett ⦿ Designing Quality Survey Questions, by Sheila B. Robinson #ux #surveys

CSAT measurement must be more than just a score. Many companies prioritize their Net Promoter Score (NPS) as a measure of Customer Satisfaction (CSAT). But do these methods truly give us a complete understanding? In reality, surveys are not always accurate. Bias can influence the results, ratings may be misinterpreted, and there's a chance that we didn't even ask the right questions. While a basic survey can indicate problems, the true value lies in comprehending the reasons behind those scores and identifying effective solutions to improve them. Here’s a better way to look at CSAT: 1. Start with Actions, Not Just Scores: Observable behaviors like repeat purchases, referrals, and product usage often tell a more accurate story than a survey score alone. 2. Analyze Digital Signals & Employee Feedback: Look for objective measures that consumers are happy with what you offer (website micro-conversions like page depth, time on site, product views and cart adds). And don’t forget your team! Happy employees = Happy customers. 3. Understand the Voice of the Customer (VoC): Utilize AI tools to examine customer feedback, interactions with customer support, and comments on social media platforms in order to stay updated on the current attitudes towards your brand. 4. Make It a Closed Loop: Gathering feedback is only the beginning. Use it to drive change. Your customers need to know you’re listening — and *acting*. Think of your CSAT score as a signal that something happened in your customer relationships. But to truly improve your business, you must pinpoint the reasons behind those scores and use that information to guide improvements. Don’t settle for simply knowing that something happened, find an answer for why it happened. Art+Science Analytics Institute | University of Notre Dame | University of Notre Dame - Mendoza College of Business | University of Illinois Urbana-Champaign | University of Chicago | D'Amore-McKim School of Business at Northeastern University | ELVTR | Grow with Google - Data Analytics #Analytics #DataStorytelling

Your customer satisfaction survey is more than a score. Here's how one client used it to leverage a strength and fix a major pain point: 1. Analyze comments Review the survey comments and identify themes for each rating. I can review about 100 surveys by hand in 30 minutes. AI software does this in seconds. Here's what my client's survey comments revealed: 💪 Strengths: employees were frequently mentioned for caring service ❌ Weaknesses: My client discovered that one particular process was a major pain point. Customers felt it was too difficult and inconvenient. 2. Investigate findings Dig deeper to learn more about the strengths and weaknesses the survey helped reveal. Observing employees and workflows is often the best way. My client's observations deepened two insights: 🙏 Employees frequently mentioned in surveys were great at building genuine rapport. Their techniques were easily shared with the rest of the team. ⏱️The painful process was inefficient. The team made changes that made the process more efficient and easier for customers. 3. Experiment Implement new ideas and track the results to see if they work. My client combined observations, anecdotal feedback from customers, and new survey results to assess how the rapport techniques and new process were working. Both were a hit! The painful process in particular stood out. Many customers mentioned how happy they were with the changes. My client had taken a pain point and turned it into a strength! Bottom line --> Follow this process to get more value from your surveys: 1. Analyze comments 2. Investigate findings 3. Experiment

How to REALLY Measure Customer Satisfaction; Understanding customer satisfaction is crucial for  any business looking to thrive. But measuring it effectively requires more than just  a gut feeling—it involves using specific tools and  techniques to gain real insights into how your  customers feel about your service. Here are some of the most effective ways to  measure customer satisfaction: 1. 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝗦𝗮𝘁𝗶𝘀𝗳𝗮𝗰𝘁𝗶𝗼𝗻 𝗦𝗰𝗼𝗿𝗲 (𝗖𝗦𝗔𝗧): Ask your customers to rate their satisfaction on a scale,  like 1-5 or 1-10. By averaging these scores, you can get  a quick snapshot of overall customer happiness. 2. 𝗡𝗲𝘁 𝗣𝗿𝗼𝗺𝗼𝘁𝗲𝗿 𝗦𝗰𝗼𝗿𝗲 (𝗡𝗣𝗦): NPS helps measure customer loyalty by asking how  likely they are to recommend your business. 3. 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝗘𝗳𝗳𝗼𝗿𝘁 𝗦𝗰𝗼𝗿𝗲 (𝗖𝗘𝗦): This score measures how easy it was for customers to  get their issues resolved. A high CES indicates that your  processes are smooth, while a low score highlights  areas that need improvement. 4. 𝗦𝘂𝗿𝘃𝗲𝘆𝘀 𝗮𝗻𝗱 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸: Short, targeted surveys after customer interactions can  provide valuable insights. Mixing rating scales with  open-ended questions ensures you capture both  quantitative and qualitative data. 5. 𝗦𝗼𝗰𝗶𝗮𝗹 𝗠𝗲𝗱𝗶𝗮 𝗠𝗼𝗻𝗶𝘁𝗼𝗿𝗶𝗻𝗴: Tracking brand mentions and sentiment on social  platforms gives you real-time feedback on customer  satisfaction. 6. 𝗜𝗻-𝗗𝗲𝗽𝘁𝗵 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄𝘀: Going beyond numbers, in-depth interviews can uncover  deeper insights into what drives customer satisfaction. These qualitative insights are invaluable for making  informed decisions. 7. 𝗔𝗻𝗮𝗹𝘆𝘇𝗲 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝗕𝗲𝗵𝗮𝘃𝗶𝗼𝗿: Metrics like repeat purchases, churn rates, and  customer lifetime value provide concrete evidence  of customer satisfaction. 8. 𝗖𝗼𝗺𝗯𝗶𝗻𝗲 𝗠𝘂𝗹𝘁𝗶𝗽𝗹𝗲 𝗠𝗲𝘁𝗿𝗶𝗰𝘀: No single metric gives the full picture. Combining various methods, such as CSAT, NPS,  and customer behavior analysis, will provide a more  comprehensive view of customer satisfaction. Measuring customer satisfaction is not just about  collecting data—it’s about understanding your  customers' experiences and using that  information to improve. 

A UX researcher recently told me their team was struggling to get high-quality user feedback. When I asked about their approach, it all made sense. They were relying on email surveys. I knew this problem firsthand. At Weebly, we ran quarterly Qualtrics email surveys, throwing in a mix of questions and hoping they were relevant. We struggled to gain real insights—until we built a better way. Here’s why email surveys fail (and what actually works): 1. Irrelevant Questions At Weebly, we asked about onboarding, free-to-paid upgrades, and churn. The problem? No segmentation. Users who never upgraded got upgrade questions. Churned users got onboarding questions. Fix: In-product surveys trigger at the right moment, based on user actions. 2. Declining Email Response Rates Most recent data shows email survey response rates are 1-1.5%. That means getting 1,000 responses requires 100,000+ emails. Compare that to in-product surveys: 📈 15-20% response rate 📉 Just 5,000-7,000 surveys needed for 1,000 responses Users engage in the product, not their inbox. 3. Email Surveys Are Too Long No one wants to fill out a long, drawn-out email survey. That’s why we keep Sprig surveys to 3 questions or less, triggered at key moments. The result: ✅ Relevant questions ✅ 15-20x higher response rates ✅ More insights, faster I shared this with the researcher. Hopefully, I convinced them. And hopefully, I just convinced you. 🙂

𝗬𝗼𝘂𝗿 𝗖𝗦𝗔𝗧 𝗶𝘀 𝟴𝟱%. 𝗬𝗼𝘂𝗿 𝗰𝗵𝘂𝗿𝗻 𝗶𝘀 𝗰𝗹𝗶𝗺𝗯𝗶𝗻𝗴. Here's why: You survey right after the interaction. Customer feels heard. Gives you a 4 or 5. Three days later? Problem comes back. No survey sent. You're measuring: → Agent politeness → Response speed → Moment-in-time feelings You're missing: → Did it work → Did they call back → Will they stay Try this: Survey immediately for interaction quality. Survey again in 7 days for resolution quality. The gap between those scores is your real performance. High score today, low score next week? Your agents are great, but your fixes don't stick. What's your follow-up survey strategy? #csat #cx #customerservice