Design Thinking Workshops

Last week, I described four design patterns for AI agentic workflows that I believe will drive significant progress: Reflection, Tool use, Planning and Multi-agent collaboration. Instead of having an LLM generate its final output directly, an agentic workflow prompts the LLM multiple times, giving it opportunities to build step by step to higher-quality output. Here, I'd like to discuss Reflection. It's relatively quick to implement, and I've seen it lead to surprising performance gains. You may have had the experience of prompting ChatGPT/Claude/Gemini, receiving unsatisfactory output, delivering critical feedback to help the LLM improve its response, and then getting a better response. What if you automate the step of delivering critical feedback, so the model automatically criticizes its own output and improves its response? This is the crux of Reflection. Take the task of asking an LLM to write code. We can prompt it to generate the desired code directly to carry out some task X. Then, we can prompt it to reflect on its own output, perhaps as follows: Here’s code intended for task X: [previously generated code] Check the code carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it. Sometimes this causes the LLM to spot problems and come up with constructive suggestions. Next, we can prompt the LLM with context including (i) the previously generated code and (ii) the constructive feedback, and ask it to use the feedback to rewrite the code. This can lead to a better response. Repeating the criticism/rewrite process might yield further improvements. This self-reflection process allows the LLM to spot gaps and improve its output on a variety of tasks including producing code, writing text, and answering questions. And we can go beyond self-reflection by giving the LLM tools that help evaluate its output; for example, running its code through a few unit tests to check whether it generates correct results on test cases or searching the web to double-check text output. Then it can reflect on any errors it found and come up with ideas for improvement. Further, we can implement Reflection using a multi-agent framework. I've found it convenient to create two agents, one prompted to generate good outputs and the other prompted to give constructive criticism of the first agent's output. The resulting discussion between the two agents leads to improved responses. Reflection is a relatively basic type of agentic workflow, but I've been delighted by how much it improved my applications’ results. If you’re interested in learning more about reflection, I recommend: - Self-Refine: Iterative Refinement with Self-Feedback, by Madaan et al. (2023) - Reflexion: Language Agents with Verbal Reinforcement Learning, by Shinn et al. (2023) - CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing, by Gou et al. (2024) [Original text: https://lnkd.in/g4bTuWtU ]

Forget "Know-Feel-Do." Here’s why I’ve stopped designing workshops and programs around how I want people to feel. One of the first things I did in my design process was map out: - What I wanted participants to know - What I wanted them to feel - What I wanted them to do Maybe you do this too? And for 'feel' I would say things like Empowered Connected Inspired. But here’s the problem… 💡 You can’t make people feel anything. People walk into a room (or log into Zoom) with their baggage. Some might be excited, others sceptical, and some just exhausted from their day. Ot have biggest concerns happening in their lives. But what we can do is... • Create a safe space to support them. • Decide how we show up as facilitators. • Design experiences that meet their needs But we can’t control their feelings. So instead of saying: ❌ "I want them to feel empowered." ✅ I focus on: "They’ve mapped out a game plan and linked it to what they want in their lives." Instead of: ❌ "I want them to feel connected." ✅ I design an exercise where each participant discusses their hopes and fears with at least three others. Outcomes, not emotions. And a byproduct... When people take action, feelings follow. Confidence comes from greater clarity. Empowerment comes from progress. And if it doesn't, that's also ok! Our role isn't to impose emotions but to foster spaces where genuine growth happens. The transformation will take care of itself. ~~ ♻️ Share if this shift makes sense to you! ✍️ Have you ever rethought how you design your workshops? Drop your thoughts below!

🔬 UX Concept Testing. How to test your UX design without spending too much time and effort polishing mock-ups and prototypes ↓ ✅ Concept testing is an early real-world check of design ideas. ✅ It happens before a new product/feature is designed and built. ✅ It helps you find an idea that will meet user and business needs. ✅ Always low-fidelity, always pre-launch, always involves real users. 🚫 Testing, not validation: ideas are not confirmed, but evaluated. ✅ What people think, do, say and feel are often very different things. ✅ You’ll need 5 users per feature or a group of features. ✅ You will discover 85% of usability problems with 5 users. ✅ You will discover 100% of UX problems with 20–40 users. 🚫 Poor surveys are a dangerous, unreliable tool to assess design. 🚫 Never ask users if they prefer one design over the other. ✅ Ask what adjectives or qualities they connect with a design. ✅ Tree testing: ask users to find content in your navigation tree. ✅ Kano model survey: get user’s sentiment about new features. ✅ First impression test: ask to rate a concept against your keywords. ✅ Preference test: ask to pick a concept that better conveys keywords. ✅ Competitive testing: like preference test, but with competitor’s design. ✅ 5-sec test: show for 5 secs, then ask questions to answer from memory. ✅ Monadic testing: segment users, test concepts in-depth per segment. ✅ Concept testing isn’t one-off, but a continuous part of the UX process. In design process, we often speak about “validation” of the new design. Yet as Kara Pernice rightfully noted, the word is confusing and introduces bias. It suggests that we know it works, and are looking for data to prove that. Instead, test, study, watch how people use it, see where the design succeeds and fails. We don’t need polished mock-ups or advanced prototypes to test UX concepts. The earlier you bring your work to actual users, the less time you’ll spend on designing and building a solution that doesn’t meet user needs and doesn’t have a market fit. And that’s where concept testing can be extremely valuable. Useful resources: Concept Testing 101, by Jenny L. https://lnkd.in/egAiKreK A Guide To Concept Testing in UX, by Maze https://lnkd.in/eawUR-AM Concept Testing In Product Design, by Victor Yocco, PhD https://lnkd.in/egs-cyap How To Test A Design Concept For Effectiveness, by Paul Boag https://lnkd.in/e7wre6E4 The Perfect UX Research Midway Method, by Gabriella Campagna Lanning https://lnkd.in/e-iA3Wkn Don’t “Validate” Designs; Test Them, by Kara Pernice https://lnkd.in/eeHhG77j UX Research Methods Cheat Sheet, by Allison Grayce Marshall https://lnkd.in/eyKW8nSu #ux #testing

Silence isn't agreement. It's talent slowly checking out mentally. You hire brilliant people. They arrive engaged and excited. Then something shifts. The sharp questions stop coming. Bold ideas disappear. Energy fades. They still show up. They still deliver. But their best thinking stays locked away. This is psychological safety in reverse. Psychological safety means people can speak up, ask questions, and share ideas without fear of punishment or humiliation. When it's missing, talented people don't leave physically. They leave mentally first. What creates psychological unsafety: - Dismissing ideas without discussion.  - Interrupting or talking over people.  - Punishing mistakes publicly.  - Taking credit for others' work.  - Making people feel stupid for asking questions. These behaviors trigger something primal in the human brain. Here's what happens when people feel unsafe: - The threat system fires. Fight-or-flight kicks in. Creative thinking stops. - The brain prioritizes survival over innovation. Status protection becomes everything. Mental withdrawal follows. 5 ways to create psychological safety: 1. Replace "Why didn't you speak up?" with "What would help you share more?" ↳ The first creates shame. The second creates safety. 2. Respond to ideas with curiosity first. ↳ Ask questions before you evaluate. Even bad ideas deserve exploration. 3. Make mistakes learning opportunities. ↳ When someone fails, ask "What did we learn?" not "Who's to blame?" 4. Admit when you're wrong. ↳ When leaders change their mind, it gives others permission to take risks. 5. Ask for input before decisions. ↳ "What am I missing?" signals you value different perspectives. The cost of psychological unsafety is massive. Every day, brilliant insights die in meetings. Game-changing solutions stay hidden. Your best people start planning their exit. But the opposite is also true. When people feel safe to speak, everything changes. Ideas flow. Innovation accelerates. Talent stays. The choice is yours. What's one conversation you've been avoiding that might unlock breakthrough thinking? ♻️ Repost to help your network keep their best talent engaged 🔔 Follow Jyoti Patel for more Business & Psychology Insights

If your team’s not speaking up… you’ve already lost. Not ideas. Not productivity. Trust. And once trust is gone? Innovation stalls. Collaboration dies. People check out—or walk out. The fix? Not another tool. Not another policy. But something far more powerful: Psychological safety. It’s not a “nice to have.” It’s the hidden engine behind every high-performing team. Here’s how you build it—one conversation, one decision, one moment at a time 👇🏼 1. Lead with curiosity, not judgment. ↳ “Help me understand…” beats “Why’d you do that?” 2. Admit your own mistakes. ↳ Model the safety you want others to feel. 3. Give credit generously. ↳ Shine the light on others—often and publicly. 4. Respond, don’t react. ↳ Let people tell the truth without fear of fallout. 5. Invite pushback. ↳ Ask: “What am I missing?” 6. Remove silent punishments. ↳ Reward honesty, not just agreement. 7. Normalize “I don’t know.” ↳ That’s how real learning starts. 8. Make feedback feel safe. ↳ Correct with care. Aim for growth, not shame. 9. Start meetings with check-ins. ↳ Connection before conversation. 10. Celebrate courage, not just results. ↳ Applaud the voice, not just the victory. Because when people feel safe, they don’t hold back. They contribute. They challenge. They soar. If you want your team to rise—safety comes first. Which one of these 10 will you lead with this week? ♻️ Share this with your network if it resonates. ☝️ And follow Stuart Andrews for more insights like this.

The demo took a weekend. Shipping it to production took six months. Every AI team learns this the hard way. A working prototype and a trusted production system are separated by a dozen unglamorous steps nobody posts about. Here's the full journey of how companies ship AI systems 👇 → Problem Framing - identify the business problem & users → Requirements & Success Metrics - define behavior, quality, latency, cost, safety → Data Sourcing & Cleaning - gather, clean, and label representative data → Evaluation Set - build benchmarks before you build the system → Model Selection & Prompt Design - pick the model, tools, and orchestration → Development - implement logic and integrations → Automated Evals + Human Review - score outputs, inspect edge cases → Safety Guardrails & Release Gates - moderation, fallbacks, eval-gated releases → Deployment & Monitoring - track latency, cost, and drift → Feedback Loop - learn from usage and improve The pattern: evals and feedback bookend everything. That's the discipline most demos skip. Save this. Share it with someone shipping AI. Which step is hardest for your team?

Are you generating enough value for users net of the value to your company? Business value can only be created when you create so much value for users, that you can “tax” that value and take some for yourself as a business. If you don’t create any value for your users, then you can’t create value for your business. Ed Biden explains how to solve this in this week's guest post: Whilst there are many ways to understand what your users will value, two techniques in particular are incredibly valuable, especially if you’re working on a tight timeframe: 1. Jobs To Be Done 2. Customer Journey Mapping 𝟭. 𝗝𝗼𝗯𝘀 𝗧𝗼 𝗕𝗲 𝗗𝗼𝗻𝗲 (𝗝𝗧𝗕𝗗) “People don’t simply buy products or services, they ‘hire’ them to make progress in specific circumstances.”  – Clayton Christensen The core JTBD concept is that rather than buying a product for its features, customers “hire” a product to get a job done for them … and will ”fire” it for a better solution just as quickly. In practice, JTBD provides a series of lenses for understanding what your customers want, what progress looks like, and what they’ll pay for. This is a powerful way of understanding your users, because their needs are stable and it forces you to think from a user-centric point of view. This allows you to think about more radical solutions, and really focus on where you’re creating value. To use Jobs To Be Done to understand your customers, think through five key steps: 1. Use case – what is the outcome that people want? 2. Alternatives – what solutions are people using now? 3. Progress – where are people blocked? What does a better solution look like? 4. Value Proposition – why would they use your product over the alternatives? 5. Price – what would a customer pay for progress against this problem? 𝟮. 𝗖𝘂𝘀𝘁𝗼𝗺𝗲𝗿 𝗝𝗼𝘂𝗿𝗻𝗲𝘆 𝗠𝗮𝗽𝗽𝗶𝗻𝗴 Customer journey mapping is an effective way to visualize your customer’s experience as they try to reach one of their goals. In basic terms, a customer journey map breaks the user journey down into steps, and then for each step describes what touchpoints the customer has with your product, and how this makes them feel. The touch points are any interaction that the customer has with your company as they go through this flow: • Website and app screens • Notifications and emails • Customer service calls • Account management / sales touch points • Physically interacting with goods (e.g. Amazon), services (e.g. Airbnb) or hardware (e.g. Lime) Users’ feelings can be visualized by noting down: • What they like or feel good about at this step • What they dislike, find frustrating or confusing at this step • How they feel overall By mapping the customer’s subjective experience to the nuts and bolts of what’s going on, and then laying this out in a visual way, you can easily see where you can have the most impact, and align stakeholders on the critical problems to solve.

One of my client companies recently made a bold shift: They replaced their Engagement KPI with a Trust KPI. And it’s one of the smartest moves I’ve seen. Why? Because trust is not a byproduct of engagement - it’s the precondition. 📚 Research backs this up: A meta-analysis by De Jong et al. (2016) found that team trust is a strong predictor of performance, especially in high-interdependence teams. Yet we treat trust like something we either have or don’t. 👉But trust isn’t a mood but rather a design decision. To start with, we need to understand 3 types of trust: 1. Cognitive 2. Affective 3. Swift Most leaders focus on cognitive or affective trust - built over time. But there’s a third type they don’t know about: Swift Trust. 📍Swift Trust forms quickly in temporary, remote, or fast-moving teams. It doesn’t require deep familiarity, it requires structure. And here’s how leaders can engineer it: ✔️ Start with clearly defined roles and expectations ✔️ Align fast around shared goals and purpose ✔️ Create quick wins that build early credibility ✔️ Model openness and ask for input from day one ✔️ Name the importance of trust explicitly In other words, trust isn’t “earned slowly” in every context. It can be catalyzed intentionally if you know how. That’s what I’m helping this client do: not just educate about trust but build it inside the team with psychological safety and my method, one behavior and ritual at a time. Because when trust becomes a designed feature, not an accidental outcome - performance, inclusion, and engagement follow. P.S.: Which type of trust is most alive in your team right now?

Psychologically Safe Spaces – A Paradox? I’m often in front of a room with underrepresented people, or those who serve them, facilitating days designed to disrupt thinking, attitudes, and behaviours. This can trigger 'disturbances', unexpected, uncomfortable emotions. These feelings need to be welcomed, held, tended to. People often ask me how I 'create 'psychologically safe spaces.' This phrase makes me shudder, and I’d like to share why. Psychological safety is complex. It’s not something I, as a host, can simply declare. Safety is felt and grows as people sense they’re genuinely seen. It requires more than intention; it’s about holding, a space where participants are invited to notice themselves, self-regulate, and co-regulate with others. Language has power. Phrases like 'safe space' can be troubling for those who’ve experienced oppression, coercion, or abuse. This risk is acute for people with valid reasons not to trust authority. Hosting spaces where historical, systemic, and cultural wounds exist isn’t about harmony or sanitising discomfort. It’s about making space for real dialogue, welcoming feelings that may be suppressed by phrases like 'this is a safe space.' As Fred Moten and Stefano Harney describe in The Undercommons, hosting in these contexts is akin to 'fugitive planning'; supporting people’s agency and fostering connections that might not otherwise flourish. For me, it involves trauma-informed, ethical principles, that recognise the complexity of groups. My laws: 1. Own my power: I’m mindful of the privilege I hold. 2. Invite difference: I welcome perspectives, even if uncomfortable. 3. Frame context: Communicate purpose and set expectations. 4. Share decision-making: Co-create the space with participants. Be prepared to flex. 5. Look for consent: Check in so participants feel able to say “yes” or “no.” 6. Give time and check in: Safety isn’t static; it needs care and attention. Hosting is a living, breathing process. The paradox of psychologically safe spaces is that they cannot be claimed into existence. Instead, they’re grown through presence, responsiveness, and a collective willingness to engage, even when it’s uncomfortable. My latest podcast Phil Pearce Shoba Ram explores some of these themes; the conditions that help us thrive, that can also harm... Renate 艶子 Rohlfing Lucy Mullins Lucia Cesaroni Elisha Griswood-Morris Cathy SHUTT Catherine Fisher Alan Hudson Toby Lowe Joe Redston Steffi Bednarek Ruth Bradley Ellie Reed Anna Glinski Andrew Reece Liz Mack Peter Merrifield Alexander Lyons John Davison Zoe Rice Robert Wanalo Jonathan Goldsmith Katy Taylor Kira Jade Cooper Steve Cook NC ❤️Kathleen Reilly Nicola Wendel Steph French

Breakthrough results happen in safe spaces. Not the manufactured, corporate-speak version of safe spaces. The real kind, where people can actually be vulnerable. Here's the difference between saying it and actually putting measures in place to make it real. "This might get sensitive, but you know what? We got each other." That's how our facilitators start every session with executives facing major change. It's one of the most powerful moments. They don't just say "this is a safe space" and hope for the best. They create a container with actual commitments. Here's what we commit to in each session: Making a space for others to share and be heard.  Engaging and participating in exercises to the best of their ability.  Learning at least one new thing about themselves.  Learning at least one new thing about fellow participants. Taking risks.  Maintaining confidentiality.  Minimizing distractions.  Staying curious.  Having fun. It's a commitment that they all take to get vulnerable, to take a risk, and have each other's back. An actual framework. Not just theory. And here's what's powerful about it: We break the fourth wall. You can use this framework in your own meetings, one-on-ones, conversations and discussions. When you create space for executives to talk about their emotions, give them language for it and give them a productive framework to move through it, magic happens. This isn't “soft skill” coaching. This is practical, business-critical work. Because leaders who can't process their own emotions about change can't lead others through it. And those emotions come out in resistance, disengagement, and culture decay. In our sessions, executives talk about big things, like potentially losing their jobs in an acquisition. They name the fear. They explore the opportunity. All because we created a container where it was safe to be human. What would change if all of your meetings started with commitments like these?