Task Feedback Loop Design

𝗜𝗳 𝗬𝗼𝘂’𝗿𝗲 𝗡𝗼𝘁 𝗗𝗲𝘀𝗶𝗴𝗻𝗶𝗻𝗴 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝗟𝗼𝗼𝗽𝘀, 𝗬𝗼𝘂’𝗿𝗲 𝗡𝗼𝘁 𝗗𝗲𝘀𝗶𝗴𝗻𝗶𝗻𝗴 𝗦𝘆𝘀𝘁𝗲𝗺𝘀, 𝗬𝗼𝘂’𝗿𝗲 𝗚𝘂𝗲𝘀𝘀𝗶𝗻𝗴. In service design and journey management, we talk a lot about touchpoints, channels, and experiences. 𝗛𝗲𝗿𝗲’𝘀 𝘁𝗵𝗲 𝘁𝗿𝘂𝘁𝗵: - No journey gets better without feedback. - No system evolves without learning loops. A feedback loop is the engine that turns friction into insight, and insight into action. In great systems, feedback loops are: 1. 𝗩𝗶𝘀𝗶𝗯𝗹𝗲 – Customers, brokers, employees can see the impact of their feedback 2. 𝗧𝗶𝗺𝗲𝗹𝘆 – Data isn’t stuck in a quarterly report, it’s now 3. 𝗔𝗰𝘁𝗶𝗼𝗻𝗮𝗯𝗹𝗲 – It doesn’t just inform, it drives change 4. 𝗖𝗹𝗼𝘀𝗲𝗱 – People know they’ve been heard 𝗜𝗻 𝗯𝗿𝗼𝗸𝗲𝗻 𝘀𝘆𝘀𝘁𝗲𝗺𝘀, 𝗳𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝗱𝗶𝗲𝘀 𝗶𝗻:  🚫 Static maps and surveys nobody reads  🚫 Call logs without analysis  🚫 Dashboards with no ownership  🚫 “That’s just how the process works” 𝗧𝗵𝗶𝗻𝗸 𝗮𝗯𝗼𝘂𝘁 𝗶𝘁: - If a customer hits the same billing error twice, that’s not bad luck, it’s a broken loop. - If frontline staff keep hacks and workarounds to themselves, that’s a missed loop. - If leadership only hears what’s escalated, that’s a distorted loop. 𝗦𝗲𝗿𝘃𝗶𝗰𝗲 𝗱𝗲𝘀𝗶𝗴𝗻 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝗳𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝗶𝘀 𝗷𝘂𝘀𝘁 𝘁𝗵𝗲𝗮𝘁𝗲𝗿. 𝗦𝘆𝘀𝘁𝗲𝗺𝘀 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝗳𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝗮𝗿𝗲 𝗱𝗲𝘀𝘁𝗶𝗻𝗲𝗱 𝘁𝗼 𝗳𝗮𝗶𝗹. 𝗪𝗵𝗮𝘁 𝗰𝗮𝗻 𝘆𝗼𝘂 𝗱𝗼 𝘁𝗼𝗱𝗮𝘆? ✅ Embed feedback into your journeys—not after them ✅ Make insights operational, not optional ✅ Connect customer data to employee experience ✅ Design loops at every level—from micro-interactions to org-wide transformation 𝗬𝗼𝘂 𝗰𝗮𝗻’𝘁 𝗶𝗺𝗽𝗿𝗼𝘃𝗲 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝗱𝗼𝗻’𝘁 𝗹𝗶𝘀𝘁𝗲𝗻 𝘁𝗼. 𝗔𝗻𝗱 𝘆𝗼𝘂 𝗰𝗮𝗻’𝘁 𝗹𝗲𝗮𝗱 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝗰𝗮𝗻’𝘁 𝗹𝗲𝗮𝗿𝗻 𝗳𝗿𝗼𝗺. #ServiceDesign #OrganizationalDesign #BusinessDesign #SystemsDesign #Research

Why Delegation Fails When There’s No Feedback Loop Founders, hands up if you’ve ever delegated a task… and then assumed it was “handled” only to discover weeks later that the results are off, missing or worse, vanished into a black hole. You’re not alone. One of the top reasons delegation falls short is the absence of structured feedback and visibility. Without a feedback loop, delegation becomes risky guesswork... tasks disappear into an invisible workflow, and you’re left wondering what really happened. Here’s what successful founders do differently: - Set Clear Expectations: Document outcomes, not just activities. Make your standards visible and accessible. - Build Feedback Checkpoints: Schedule short, recurring updates where results and roadblocks are reviewed together. It’s not micromanagement, it’s vital course correction. - Enable Radical Visibility: Use tools like dashboards or shared trackers so everyone “sees” progress, not just completion. - Empower EAs to Improve the System: Great executive assistants don’t just do the work... they close the loop. They offer insights, propose improvements, and help refine your delegation process over time. Bottom line: Delegation only works when you design it as a closed loop. The best EAs aren’t just task-takers, they’re system optimizers and accountability partners. How are you closing the loop on delegation in your business? Have you implemented a feedback system that actually works? Share your tips or challenges in the comments below!

How do you actually know if your learning and development program is working? It’s a question that comes up all the time for AEC firms. There’s no shortage of ideas when it comes to training. New tools, new frameworks, new skills to build. The challenge isn’t generating ideas. It’s knowing which ones truly matter. In this clip, what stands out to me is a simple but powerful approach: let the project work itself tell you where to invest in L&D. At BWBR, their Landmark Learning program for emerging professionals is shaped directly by feedback from the people closest to the work—quality assurance and construction administration teams. These groups see, in real time, where projects succeed and where they struggle. They assign grades across key categories—like how well teams are handing off projects—and look for patterns. If project teams are consistently underperforming in a certain area, that becomes a priority for training. That same system creates a feedback loop. The issues identified by QA and CA teams inform the training agenda. Then over time, those same metrics show whether performance is actually improving. You’re not guessing if the training worked because you can see it in the data, in the field, in the way projects are delivered. And that’s where learning and development starts to feel less like a set of programs and more like a system for continuously improving how a firm operates. Because the goal of learning and development isn’t to deliver training. It’s to improve the performance of individuals, of project teams, and of the firm as a whole. When you design a feedback loop like this in your L&D efforts, prioritization becomes clearer. Investment becomes more intentional. And learning becomes directly connected to outcomes. This clip comes from “Redesigning Learning for the Next Generation of AEC Talent | Dan Hottinger and Kari Shonblom of BWBR”, episode 1 of the Smarter by Design podcast. 📺 🎧 Watch or listen to the full episode here: https://lnkd.in/gevBva5y #AEC #KnowledgeManagement #ModernLearningOrganizations #SmarterByDesign

𝗚𝘂𝗮𝗿𝗱𝗿𝗮𝗶𝗹𝘀 aren't an afterthought or extra credit anymore - they're core architectural patterns that determine whether your agentic system is safe to deploy. So here are four different workflow patterns that we've seen implemented in production systems: 1️⃣ 𝗔𝗱𝗮𝗽𝘁𝗶𝘃𝗲 𝗙𝗲𝗲𝗱𝗯𝗮𝗰𝗸 𝗟𝗼𝗼𝗽𝘀 Worker agents execute tasks → Supervisor evaluates → Rewards Service updates policies → Guidelines adjust → Workers improve over time. This creates a continuous learning cycle where the system reinforces effective behaviors and discourages risky ones. It's reward-driven learning that improves with iteration. 2️⃣ 𝗖𝗼𝗿𝗿𝗲𝗰𝘁𝗶𝘃𝗲 𝗔𝗰𝘁𝗶𝗼𝗻 The centralized Supervisor assigns tasks, compares outputs against application guidelines, and if errors are detected, engages alternative workers. The best validated result gets returned. This prevents bad outputs from ever reaching users. 3️⃣ 𝗛𝘂𝗺𝗮𝗻 𝗶𝗻 𝘁𝗵𝗲 𝗟𝗼𝗼𝗽 For sensitive domains (medical diagnosis, legal review, financial approvals), agents generate preliminary responses but humans validate before execution. The workflow automatically pauses for expert review, then resumes once approved. 4️⃣ 𝗘𝗺𝗲𝗿𝗴𝗲𝗻𝗰𝘆 𝗦𝘁𝗼𝗽 Critical for high-risk environments like trading systems. Agent 1 collects market data → LLM processes signals → Agent 2 evaluates conditions → if anomalies or risks detected, execution halts immediately. Consider a trading bot with access to a volatility API showing VIX at 42 (extreme market stress). Even if the bot generates an aggressive trade recommendation, the evaluator independently verifies: "Given current volatility, does this make sense?" If not, it blocks the action entirely. 𝗕𝗲𝗵𝗮𝘃𝗶𝗼𝗿 𝗦𝗵𝗮𝗽𝗶𝗻𝗴 is the underlying philosophy here - a three-step loop of scoring, feedback, and correction. The evaluator doesn't just measure performance after the fact. It actively intervenes: triggering rollbacks for bad transactions, halting workflows propagating incorrect data, or routing edge cases to human reviewers. This is especially important when agents interact with volatile external states - market conditions, API health, system load. The evaluator provides a sanity check to ensure the model correctly interpreted the signals it was given, not just that it generated understandable text. The goal isn't catching every possible failure upfront (impossible). It's building systems that detect problems as they happen, understand what went wrong, and automatically correct course before damage propagates. Inspired by our most recent ebook we did with StackAI and Weaviate: https://lnkd.in/dKt9SVya

🔥 One skill changed how we work with Claude Code on dbt projects. It runs automatically after every modeling session. We called it /retro. My colleague Fernando Jimenez implemented it. Here's what it does: 1. Triggers a retro — what went well, what didn't, what needs improvement 2. Reviews the current skills and workflows we have in place 3. Updates them based on what it just learned That's it. Three steps. But the compound effect is wild. Week 1: Claude keeps making the same formatting mistakes in our staging models. Week 2: It stopped. Because the retro caught it, updated the skill, and the pattern never repeated. Week 1: We manually remind it about our naming conventions every session. Week 3: It just knows. The workflow adapted. This is what a positive feedback loop looks like in practice. You don't need a complex AI strategy. You need a system that learns from its own work. Every dbt session we run makes the next one better. We're not getting better at prompting. The setup itself is getting better at building.