Planning Sprints Effectively

As the leader of an Intelligent Automation CoE, I’ve had the privilege of guiding enterprise teams in their evolution from RPA and low-code platforms to AI-driven decisioning and orchestration. Across industries, a few core principles consistently enable scalable, precise, and impactful automation. Here are five principles I’ve seen consistently deliver results: ✔️ Start with a high-impact use case: Identify a process with clear ROI and measurable outcomes. Automate it end-to-end before expanding. ✔️ Iterate fast, automate faster: Build automation in agile sprints. Test early, deploy often, and refine based on real user feedback. ✔️ Don’t fear manual effort early on: Use low-code tools, RPA, and human-in-the-loop models to validate automation before scaling. Doing things that don’t scale helps you learn what will. ✔️ Embed automation into existing workflows: Design bots and AI agents to integrate seamlessly with enterprise systems (ERP, CRM, ITSM). Automation should feel like an enhancement, not a disruption. ✔️ Build a strong automation foundation: Hire engineers and architects who understand both business processes and automation platforms. Early talent sets the tone for scalability and governance. These principles can help you move from isolated wins to enterprise-wide impact. Whether you're just starting or scaling your automation journey, these fundamentals hold true. What worked (or not) in your automation journey? 🎯 Follow my AI & IA - Art of the Possible newsletter for insights: https://lnkd.in/g5TkS8pv #IntelligentAutomation #AutomationCoE #DigitalTransformation #AI #RPA #EnterpriseAutomation #Leadership #AgileAutomation P.S. The content of this post reflects my personal viewpoints, not those of my employer.

AI adoption doesn’t happen through slide decks or when leaders buy subscriptions to a copilot—it happens when people feel the impact in their own work. 𝐈𝐧𝐭𝐞𝐫𝐧𝐚𝐥 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 𝐃𝐞𝐬𝐢𝐠𝐧 𝐒𝐩𝐫𝐢𝐧𝐭 At a recent company offsite, we ran an automation design sprint using n8n to help our departments eliminate repetitive tasks, free up time for high-impact work, and get hands-on with AI. We are definitely biased, but it seems like it was a solid success. 𝐒𝐞𝐭𝐭𝐢𝐧𝐠 𝐭𝐡𝐞 𝐒𝐭𝐚𝐠𝐞 • Focused on one tool – People are overwhelmed by the speed of AI and all the tools and capabilities. We did the research, chose n8n as our automation platform (others include Make, Zapier), and simplified the choice for them. • Assigned an Automation Lead – Gave them time to ramp up, set up preconfigured APIs, and prep the environment. • Pre-reads & videos – Our automation leader met with departments in advance and shared primers so teams weren’t starting cold. 𝐄𝐱𝐞𝐜𝐮𝐭𝐢𝐨𝐧: 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 𝐢𝐧 𝐀𝐜𝐭𝐢𝐨𝐧 • Breakout sessions – Departments identified pain points and mapped potential automations. Each team had an assigned engineer to help execute or clear roadblocks. • Rapid prototyping – 1-hour workflow design → timeboxed builds. • Show & tell – Teams presented their automations, the "why" behind them, and their progress. Many were fully functional by the end. 𝐊𝐞𝐞𝐩𝐢𝐧𝐠 𝐭𝐡𝐞 𝐌𝐨𝐦𝐞𝐧𝐭𝐮𝐦  A month later, live automations are running across all teams—with more in the pipeline. And to make automation stick, we put an initial structure in place: • Automation Lead role formalized. • Department-level automation roadmaps created. • Engineering leads assigned until teams are self-sufficient. • Focus on training team members in each department. • Regular check-ins between teams and automation leads. • “Automation of the Week” updates to highlight wins. We’ll share more on what’s working (and what’s not) as we scale this. I am curious what other teams are doing on this front and how they are executing. Would love to hear in the comments or directly from folks.  

Daily Scrum Meetings aka Daily Standups — these 15-minute huddles are crucial for identifying roadblocks, aligning Sprint Goals, and fostering team synergy. But let’s face the reality: Daily Standups don’t always go as planned. Discussions sometimes derail, blockers remain vague, or team members hold back crucial details. This is where AI can help! Let’s explore how AI can elevate Daily standups to the next level. ✅ Streamlining Pre-Standup Preparation: ‣ AI tools like TeamRetro and Standuply gather sprint data, highlight blockers, and generate insights. ‣ Review key metrics, spot trends, and personalize follow-ups for better focus. ✅ Guiding Team Discussions: ‣ Clockwise or Slack bots keep the agenda on track. ‣ AI suggests actionable questions and manages time, ensuring discussions stay brief. ✅ Improving Blocker Resolution: ‣ Pattern recognition flags recurring blockers. ‣ AI provides context and recommendations for faster resolution. ✅ Enhancing Remote Standups: ‣ Tools like Geekbot gather updates asynchronously. ‣ AI performs sentiment analysis and provides auto-translation for global teams. ✅ Automating Follow-Ups: ‣ AI tools like Otter.ai create summaries, updates Atlassian Jira/Trello, and send reminders for accountability. With these automated processes, Scrum Masters can focus on coaching and leadership instead of chasing mundane updates. #agile #scrum #scrummaster #productowner #ai #agiletools PremierAgile

I spend $4,600/month on tools. And it’s the best investment I’ve made for Leadle. Not to replace people, but to reorganize how we build and sell. Most B2B teams I work with aren’t losing deals from poor products or pitch. They’re losing them from friction, lag, and bad system design - and that’s what we aim to solve for. To go beyond their pipelines to fix the system that can make or break their revenue. Here are a few patterns I’ve observed working with teams with 50+ employees and how we helped them get better results: 1. Design workflows, not just roles. Map every motion in your GTM: • Lead capture • Routing • Qualification • Research • Proposal delivery • Follow-up • Deal monitoring Then ask: 👉 “Should this be owned by a human, or a system?” If it’s repeatable → it’s programmable. So, don’t assign a role. Build a flow. 2. Restructure your sales pod like this: → One Operator Owns pipeline strategy, playbooks, and automation logic. (Think: GTM engineer, not just a “manager.”) → One Closer Deep conversations, enterprise deals, internal buying groups. → A stack of micro-agents • Research assistants • Auto-sequencers • Proposal builders • No-show detectors • Risk signal monitors These agents save time + protect revenue. 3. Run weekly GTM sprints. Every Friday, review: ✅ Agent performance (triggers fired, actions completed) ✅ Reps’ blockers (What’s not automated yet?) ✅ Drop-offs in pipeline velocity Your RevOps team becomes the product team for your GTM engine. Result? Fewer handoffs, faster cycles, and cleaner decisions. But that doesn’t happen unless you rethink org design at the workflow level. TL;DR: Go beyond automating sales to building systems where humans + agents actually sell better together. If you want to see how this model will look like for you, DM me. Happy to share insights! #agenticsystems #aiinsales #salesautomation #aiworkflows 

Most teams are implementing AI in Scrum backwards. They're using it to automate the easy parts. Generating material for user stories, summarizing meetings, tracking throughput. But they're keeping the hard parts manual, the uncomfortable conversations, the difficult trade-offs, the moments where teams can learn. Experimenting with AI in the Scrum events has taught me that, usually, the real issue teams run into isn't efficiency it's avoidance and a reluctance to rock the boat. Scrum events fail because teams systematically avoid the conversations that matter. Sprint Planning becomes feature Tetris instead of value negotiation. Daily Scrums become status updates instead of problem-solving. Reviews become demo parties instead of outcome validation. Retros become complaint sessions instead of improvement engines. But AI can make these difficult conversations unavoidable. In Sprint Planning AI surfaces value tensions. Analysing customer behaviour data, technical debt patterns, and market signals to present conflicting priorities. For example: "Customer usage data suggests Feature A, but technical debt analysis suggests refactoring will deliver 3x more capacity for Feature B next quarter." AI focuses attention on what does value mean to us right now? In Daily Scrums AI flags collaboration patterns, handoff delays, and knowledge gaps in real-time to surface friction. For example: "Three stories are blocked on review, but I can see a lot of external meetings today, and similar patterns happened in 4 of the last 6 sprints." AI forces the conversation on how is our system of work working? In Sprint Reviews AI presents leading indicators of feature success/failure alongside the demo, forcing teams to align on what working product means. For example: "Feature demo looks good, but early CX data shows 40% of users need support to complete the workflow, and it's increased our support ticket volume by 15%." AI raises transparency on how we validate and build confidence if this sprint actually delivered value? In Retrospectives AI surfaces patterns across sprints that teams may miss or side-step. For example: "Sprints with >25% carryover consistently happen after weeks with 3+ urgent stakeholder requests." AI brings team attention to the systemic changes that could be made to move the needle. The counter-intuitive angle here is these teams have more conflict, not less. But it's far more productive conflict about the things that matter. Although using AI this way really helps I'd encourage you not to start by adding AI to all of your events at once. Start with the event your team avoids or plays safe in the most (usually it's Retros). And lastly a question you can ask to gauge if your team are ready to leverage AI this way: Can your team have a 30-minute argument about priorities and still respect each other afterward? If not, fix your psychological safety first. AI will likely amplify whatever dynamics already exist.

In Agile, automation should be driven by risk, stability, and feedback value not sprint pressure. Not every story should be automated but every story should go through an automation feasibility check. Instead of asking: “Can we automate this quickly?” A better technical question is: “Will automating this give us reliable, repeatable feedback in our pipeline?” When deciding whether a scenario belongs in the automation suite below evaluation shall be checked : - Stability : Is the feature/API/UI likely to change next sprint? - Repeatability : Will this scenario run across builds, environments, or data sets? - Business criticality : Does failure block core user journeys? - Regression impact :Could future changes easily break this flow? - Data variation : Does it benefit from parameterized runs? (DDT) - Execution cost : Is manual execution expensive or time-consuming? Then comes the technical placement decision: • Can this be validated at API/service layer instead of UI? • Does it fit fast CI smoke or deeper regression ? • Will the failure signal be clear and diagnosable? Good Agile automation is not about maximizing script count it’s about maximizing trustworthy feedback per build. How do you decide what gets automated each sprint?

Most teams focus on automating regression tests, ensuring that existing functionality remains intact. But if automation is only catching regressions, it’s always a step behind. By the time issues are found, they’re already in the product. What if automation worked with development instead of trailing behind? In sprint automation changes the game. Instead of waiting for a full sprint to end before automating, why not start automating as features are being built? • Automate unit and component level checks while development is in progress. • Build testability into the code from the start, reducing late-stage defects. • Catch issues early, making fixes cheaper and faster. • Reduce dependency on heavy post sprint regression cycles. • Align automation efforts with business goals, not just code changes. Automation isn’t just about efficiency, it’s about impact. When testing and automation are embedded in the sprint, teams move faster, ship with confidence, and deliver real value. Regression automation is necessary. But relying only on it is like wearing a seatbelt after a crash. Shift left, automate smart, and make testing a continuous process. #softwaretesting #softwareengineering #testautomation #agile #qualityengineering #brijeshsays

In-Sprint Test Automation is this a reality or a pipe dream? For most team achievement of a scalable, reliable and trustworthy #testautomation itself is a big deal. Getting to the stage of In-Sprint Test Automation seems like a dream. We have been able to achieve this dream in our projects in the last few years and this is what helped - Establish a clear objective & a common understanding of test automation among all stakeholders - Create a test automation framework which is easy to understand and performs the majority of tasks which a test automation engineer would require as a boiler plate implementation - Understand the architecture of the application and align your automation closely to it - Every test case documented shouldn't be considered for automation. Have a clear understanding and guideline on which ones to automate - Involve testers in automation and keep #testing as the primary focus in automation - Adopt API first test automation approach wherever possible - Work with developers to define standards for locator definition and validate/audit to ensure everyone is following it. Log defects if someone isn't doing it - Build as automation what you can run in the pipeline - Build automation execution infrastructure which can scale and support multiple parallel threads - Own your test data, don't have dependency on others for it Share your experiences and practices you have followed to achieve In-Sprint Test Automation

𝗠𝗼𝘀𝘁 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝘀 𝗮𝗿𝗲 𝗲𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗔𝗜… 𝗯𝘂𝘁 𝘃𝗲𝗿𝘆 𝗳𝗲𝘄 𝗵𝗮𝘃𝗲 𝘁𝗿𝘂𝗹𝘆 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝗮𝗹𝗶𝘇𝗲𝗱 𝗶𝘁. The result? ➡️ Fragmented usage ➡️ No clear ownership ➡️ Limited governance ➡️ Little to no measurable productivity gains Most delivery teams are still aligned to sprint commitments — not optimization. 𝗧𝗵𝗲 𝗿𝗲𝗮𝗹 𝗴𝗮𝗽 𝗶𝘀𝗻’𝘁 𝘁𝗼𝗼𝗹𝘀. 𝗜𝘁’𝘀 𝗮𝗻 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹. At Royal Cyber, we are addressing this through a structured approach we call: 👉 𝗔𝗜 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝘃𝗶𝘁𝘆 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 (𝗔𝗣𝗘) This is not just another role. It’s a new delivery construct embedded directly into sprint teams — focused on: ✔️ Identifying automation opportunities across the SDLC ✔️ Accelerating development, testing, and documentation ✔️ Establishing measurable productivity baselines ✔️ Driving continuous optimization through AI Our approach is simple, but powerful: 1️⃣ 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆 Map the current delivery lifecycle, identify repetitive workflows, and define high-impact AI opportunities. 2️⃣ 𝗣𝗶𝗹𝗼𝘁 𝗜𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 Embed an AI Productivity Engineer into a live sprint team to deliver quick, measurable wins. 3️⃣ 𝗦𝗰𝗮𝗹𝗲 & 𝗚𝗼𝘃𝗲𝗿𝗻 Standardize reusable patterns, introduce governance, and build a scalable AI-enabled operating model. 𝗧𝗵𝗶𝘀 𝗶𝘀 𝘁𝗵𝗲 𝘀𝗵𝗶𝗳𝘁 𝘄𝗲’𝗿𝗲 𝘀𝗲𝗲𝗶𝗻𝗴: From AI experimentation → AI operationalization → AI-driven delivery transformation We’re already applying this approach across enterprise programs — and the early impact is clear. 🎥 Watch the video to see how we are bringing this to life. If you’re exploring how to make AI practical, scalable, and measurable within your delivery teams — let’s connect. #AI #GenAI #DigitalTransformation #SoftwareDelivery #SDLC #EngineeringExcellence #RoyalCyber #EnterpriseAI #TechLeadership