<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Doug DeFranco on Medium]]></title>
        <description><![CDATA[Stories by Doug DeFranco on Medium]]></description>
        <link>https://medium.com/@dougdefranco?source=rss-8718785d6240------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*xGkKwBN88tpdeD5lW0qYlg@2x.jpeg</url>
            <title>Stories by Doug DeFranco on Medium</title>
            <link>https://medium.com/@dougdefranco?source=rss-8718785d6240------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 05:54:03 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dougdefranco/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Persistence in Adversity, Humility in Prosperity: My Grandfather’s Leadership Code]]></title>
            <link>https://medium.com/flow-over-theater/persistence-in-adversity-humility-in-prosperity-my-grandfathers-leadership-code-81e41e789b94?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/81e41e789b94</guid>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[agile-coaching]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Mon, 09 Nov 2026 23:37:03 GMT</pubDate>
            <atom:updated>2026-11-09T23:37:03.961Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>Persistence in Adversity, Humility in Prosperity: My Grandfather’s Leadership Code</p><p>My grandfather was a doctor. He never spent a day in tech. His one sentence has carried me through ten years of blocked sprints and green dashboards.</p><p>My grandfather was a doctor.</p><p>He never ran a standup. He never saw a burndown chart, never sat through a PI Planning, never heard the words &quot;Agile transformation&quot; in his life.</p><p>He still taught me everything I know about leading teams.</p><p>His saying — the one our family lives by — goes like this:</p><blockquote>&quot;Persistence and determination in all adversity, humility and goodness in all prosperity.&quot; — V. James deFranco, M.D.</blockquote><p>Read it again, slowly, because it&#39;s really two instructions wearing one coat. The first half is for when everything is on fire. The second half is for when everything is going well — and in my experience, the second half is the one leaders fail.</p><p>I&#39;ve spent 10 years as a Scrum Master and Agile Coach, running 4–5 teams at a time across banking, financial services, and healthcare. The frameworks changed. The tools changed. The companies changed. That sentence never has, because every sprint I&#39;ve ever worked falls into one of his two categories — and each one demands something different from the person leading the room.</p><h3>The adversity sprint: persistence is a tone before it’s a plan</h3><p>You know this sprint. The release is blocked. The PI is wobbling three weeks in. A dependency that was &quot;handled&quot; turns out to be handled by nobody. The team is still showing up, still moving sticky notes, but you can feel the confidence leaking out of the room — that quiet moment where good people start to privately quit on a goal they publicly committed to.</p><p>This is where most leaders reach for pressure. Louder status requests. More check-ins. A tone that says I am disappointed before it says anything useful.</p><p>My grandfather&#39;s first half taught me the opposite. Persistence is not noise. Persistence is the Scrum Master walking into that room calm, naming the real problem out loud — the actual one, not the comfortable one — and then doing the unglamorous work of removing it. Determination is a tone before it&#39;s a plan. A team takes its emotional cues from whoever leads the room, and a leader who panics gives everyone permission to panic. A leader who stays steady gives them something better: permission to think.</p><p>So in the adversity sprint, my job gets very specific. No blame — blame is just fear with a target, and it teaches teams to hide the next problem instead of surfacing it. Instead: which blocker is costing us the most, who actually owns it, and what does it take to move it today? Then follow up. And follow up again. Half the blockers I&#39;ve removed in ten years were never my teams&#39; to solve — they lived with other teams, other approvals, other priorities — and they moved because somebody refused to let them sit quietly in a status report.</p><p>That&#39;s not a heroic story. It&#39;s a discipline. The gains people like to quote from my work — throughput up 80%, committed work up 50% — didn&#39;t come from pushing teams harder in bad weeks. They came from what got removed in bad weeks, sprint after sprint, by somebody who treated persistence as a practice instead of a poster.</p><p>Adversity doesn&#39;t ask a leader to be brilliant. It asks them to still be there, calm and specific, on the fourth hard day — not just the first.</p><h3>The prosperity sprint: don’t believe your own dashboard</h3><p>Now the other sprint. The one nobody warns you about.</p><p>Everything ships. The metrics are green. Leadership is happy, the demo got applause, and your dashboard looks like a golf course. This is the sprint where my grandfather&#39;s second half earns its keep — because prosperity is where leadership quietly goes bad.</p><p>Success does something sneaky to a Scrum Master. You start believing the dashboard. You start believing the green is you — your process, your facilitation, your brilliance — instead of what it actually is: a team of people who did hard things well. And the moment you believe that, you stop listening. The retro gets shorter. The risks get waved through. The one person trying to say &quot;this number isn&#39;t as real as it looks&quot; learns to stop saying it.</p><p>I&#39;ve written elsewhere that the most honest meeting in Agile is the one teams fake when they stop believing the truth matters. Prosperity is where that faking starts — not in the bad weeks, when everyone knows there&#39;s a problem, but in the good ones, when nobody wants to be the person who questions a green chart.</p><p>Humility, in my grandfather&#39;s sense, is the discipline of the green sprint. The team did that. Not you. Credit flows down — by name, in public, to the people who actually carried the work. Protection flows up — the Scrum Master absorbs the noise, the escalations, the politics, so the team&#39;s focus survives its own success. And goodness means success never becomes an excuse to stop listening: you run the honest retro anyway, you ask what the dashboard isn&#39;t showing, you check the thing everybody assumes is fine.</p><p>A leader who can only lead in a crisis is half a leader. The green weeks test your character more than the red ones, because nothing forces you to be humble. You have to choose it.</p><h3>What I’d hand you for your worst week — and your best one</h3><p>If you lead teams — Scrum Master, coach, RTE, or the person everyone looks at when the room goes quiet — here is the whole code, translated into the two weeks you&#39;re guaranteed to get:</p><p><strong>In your worst week:</strong> Be the calmest person in the room, on purpose. Name the one real blocker, out loud, without blame. Route it to whoever actually owns it, with the cost stated plainly. Then follow up until it moves. Persistence is follow-through nobody applauds.</p><p><strong>In your best week:</strong> Give the credit away by name. Ask what the green metrics are hiding. Protect the team&#39;s time like the success was fragile — because it is. Humility is follow-through nobody demands.</p><p>Two kinds of sprint. Two kinds of discipline. One sentence from a doctor who never heard of Scrum, and it has outlasted every framework I&#39;ve ever been certified in.</p><p>That&#39;s not an accident. Frameworks tell you how to run the meeting. A code tells you who to be in it.</p><p>Scrum Masters, coaches, and RTEs — what&#39;s the best leadership advice you ever got from someone completely outside this industry? I read every answer, and the best ones tend to come from grandparents, coaches, drill sergeants, and first bosses. Tell me yours.</p><p><strong>If this was useful, follow along. </strong>I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don&#39;t miss the next one. One useful email when there&#39;s something worth reading. No noise.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=81e41e789b94" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/persistence-in-adversity-humility-in-prosperity-my-grandfathers-leadership-code-81e41e789b94">Persistence in Adversity, Humility in Prosperity: My Grandfather’s Leadership Code</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[500+ Workloads Taught Me This: Dependencies Are the Real Backlog]]></title>
            <link>https://medium.com/flow-over-theater/500-workloads-taught-me-this-dependencies-are-the-real-backlog-8b28fff1ec70?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/8b28fff1ec70</guid>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[cloud-migration]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Sun, 08 Nov 2026 17:43:15 GMT</pubDate>
            <atom:updated>2026-11-08T17:43:15.751Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>Migrating Small Business Card systems to AWS at Capital One with 4–5 teams, the work was never the hard part. The waiting was. Here’s the lesson I now carry into every engagement.</p><p>At Capital One, I ran 4–5 Scrum teams migrating Small Business Card systems to AWS — more than 500 workloads, moved piece by piece out of legacy infrastructure and into the cloud.</p><p>People assume the hard part of a job like that is the technology. It isn’t. The technology was challenging, sure — but it was <em>knowable</em>. It could be designed, tested, reviewed, and validated; across that program we held a 95% pre-production validation rate, which tells you the engineering discipline was real.</p><p>The hard part was everything the work was waiting on.</p><p>Waiting on another team’s deliverable. Waiting on an environment. Waiting on an approval, a security review, a window, a person who was in three other migrations at the same time. Five hundred workloads, and the single largest category of “work” in the whole program was work standing still.</p><p>That’s where I learned the lesson I now carry everywhere: <strong>dependencies are the real backlog.</strong> The stories in Jira are the backlog everyone manages. The dependencies are the backlog that manages you.</p><h4>The dependency nobody wrote down</h4><p>Here’s how it usually went wrong.</p><p>A team’s sprint would be planned clean. The stories were refined, estimated, ready. Then, mid-sprint, the truth would surface: this workload can’t move until that platform change lands. That change belongs to another team. That team has it scheduled for next sprint — or hasn’t scheduled it at all, because nobody told them we were counting on it.</p><p>Nothing in that story is incompetence. Every person in it did something reasonable. The failure was structural: the dependency existed in the hallway, in somebody’s head, in a thread from six weeks ago — everywhere except the one place it needed to be, which was <em>visible, early, next to the work it controlled</em>.</p><p>Multiply that by five teams and five hundred workloads and you don’t have a coordination problem. You have an invisible second backlog, larger than the real one, deciding your dates while everyone watches the burndown and wonders why it’s lying.</p><h4>Discovering them live is the most expensive way to learn</h4><p>The worst place to learn about a dependency is during PI Planning, in front of the whole train — or worse, mid-sprint, when the cost of the discovery is highest and the options are fewest.</p><p>We learned to pull the discovery forward. Before PI Planning, and before sprint planning, teams had to surface what their work depended on: which other teams, which platforms, which approvals, which dates. Not as a form to fill in — as a standing question with a name attached. Who owns the thing we’re waiting on? Do they know we’re waiting? Is their date a commitment or a hope?</p><p>Those three questions, asked early and asked routinely, did more for the migration’s predictability than any process change we made. A dependency found six weeks out is a conversation between teams. The same dependency found at go-live is an escalation, a slip, and a room full of people explaining a date nobody can defend anymore.</p><h4>Flow over theater, applied to the invisible backlog</h4><p>This is where my whole “flow over theater” conviction got forged, honestly.</p><p>Theater, in a migration program, looks like green status on work that is quietly blocked — boards full of in-progress stories, standups full of activity, and dates slipping anyway. Flow looks like a much less comfortable board: dependencies drawn in the open, blocked items visibly blocked, waiting time measured instead of hidden.</p><p>When we started treating dependencies as first-class work — listed, owned, dated, and reviewed like stories — a few things changed. Blocked time became a number we could argue with productively instead of a feeling. Escalations got earlier and calmer, because “we’ve known about this for five weeks and here’s the paper trail” lands very differently than “we just found out.” And throughput followed: across my teams in financial-services delivery, that discipline was a core part of how throughput rose 80% and committed work rose 50% — not by working harder, but by spending less time standing still.</p><h4>The practice I’d hand to anyone running multi-team work</h4><p>If you take one thing from 500+ workloads, take this:</p><ol><li><strong>Name the dependency when you name the work</strong>.<strong> </strong>Every significant story gets its external dependencies listed at refinement — or an honest “none that we know of,” which is itself useful information.</li><li><strong>Give every dependency an owner and a date. “</strong>The platform team” is not an owner. A person, a team, and a date they have actually agreed to — that’s an owner.</li><li><strong>Make waiting visible.</strong> Track blocked time the way you track completed work. What gets measured gets managed; what gets hidden gets discovered at the worst moment.</li><li><strong>Surface it before planning, not during.</strong> The dependency conversation belongs in the weeks before PI Planning and sprint planning, while there are still options. In the event, it’s too late to be cheap.</li></ol><p>Do that consistently and the real backlog shrinks to the size of the visible one — which is the only backlog you can actually manage.</p><h4>The work is rarely the work</h4><p>Five hundred workloads taught me that the stories are the easy part to see and the dependencies are the part that decides your outcomes. Every multi-team program I’ve run since, I’ve managed two backlogs: the one in the tool, and the one made of other people’s work, approvals, and dates.</p><p>Most Scrum Masters only manage the first one.</p><p>Manage both, and you stop being surprised by your own program.</p><p>Scrum Masters, RTEs, and delivery leads — what’s the dependency that taught you this lesson the hard way? I read every answer, and the best ones will shape a future article.</p><p><strong>If this was useful, follow along.</strong> I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don’t miss the next one. One useful email when there’s something worth reading. No noise.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8b28fff1ec70" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/500-workloads-taught-me-this-dependencies-are-the-real-backlog-8b28fff1ec70">500+ Workloads Taught Me This: Dependencies Are the Real Backlog</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How We Raised Throughput 80% Without Adding a Single Person]]></title>
            <link>https://medium.com/flow-over-theater/how-we-raised-throughput-80-without-adding-a-single-person-3215a34e62a6?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/3215a34e62a6</guid>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[agile-coaching]]></category>
            <category><![CDATA[leadership]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[scrum-master]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Sat, 07 Nov 2026 00:33:40 GMT</pubDate>
            <atom:updated>2026-11-08T19:04:24.311Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>The teams weren’t lazy and they weren’t understaffed. The work was stuck. Here’s what we changed — in the order we changed it — during a large AWS migration in financial services.</p><p>When I started leading 4–5 Scrum teams through a large AWS cloud migration in financial services, the teams were busy.</p><p>Busy is not the same as delivering.</p><p>Stories carried over sprint after sprint. Work in progress was everywhere — every board crowded, every person juggling three things, every standup a progress report on work that never seemed to land. And committed work, the work the teams had looked leadership in the eye and promised, kept slipping.</p><p>The instinct in that situation is almost always the same: we need more people. More hands, more capacity, more throughput. It’s an understandable instinct and, in my experience, it’s usually wrong. The teams weren’t failing because they lacked effort or headcount. They were failing because the work was stuck — and stuck work doesn’t care how many people you throw at it.</p><p>So we changed how the work flowed, not how hard people worked. Here’s what we did, in the order we did it — including the part that didn’t work first.</p><h4>What didn’t work first: pushing harder</h4><p>Before the flow changes, there was a phase of pressure. I won’t pretend I was above it; when commitments slip, the temperature rises, and the first lever everyone reaches for is intensity. More status checks. More urgency. More “can we pull this one in too?”</p><p>It produced motion, not delivery. Teams started more work to demonstrate effort. Boards got fuller. Carry-over got worse. Pressure applied to a system with too much work in progress doesn’t speed the system up — it just makes the traffic jam louder.</p><p>That failure taught me the real diagnosis: the constraint wasn’t effort. It was flow. So we stopped pushing and started unblocking.</p><h4>Change 1: Cut work in progress</h4><p>This was the highest-leverage change and the hardest sell.</p><p>Every team was working on too many things at once, which meant everything took longer, which meant more pressure to start the next thing early, which meant everything took even longer. Classic trap.</p><p>So we set explicit limits: fewer things started, more things finished. A story couldn’t be pulled in just because someone was free for an afternoon; the board had to earn the space. It felt wrong for about two sprints — people accustomed to being busy on five fronts felt exposed being focused on one.</p><p>Then the first stories started actually finishing, end to end, inside the sprint. Finishing is addictive. The teams didn’t need convincing after they felt it.</p><h4>Change 2: Make every dependency visible before sprint planning</h4><p>The second change attacked the carry-over problem at its source.</p><p>Stories were entering sprints with hidden dependencies — work owned by other teams, environments that weren’t ready, approvals nobody had requested yet. The dependency would surface mid-sprint, the story would stall, and the stall would be recorded as the team’s miss.</p><p>We moved the discovery earlier. Before planning, dependencies got named: who owns this, when does it land, what happens to our sprint if it doesn’t? Not bureaucracy — a fifteen-minute habit of asking “what could stop this story?” while there’s still time to do something about it.</p><p>Blocked work didn’t disappear. But it stopped being a surprise, and surprises were what were killing the commitments.</p><h4>Change 3: Tighten refinement so stories enter ready</h4><p>A story that enters a sprint fuzzy will leave it unfinished.</p><p>We tightened backlog refinement until “ready” meant something specific: the story is clear enough that the team can start without a meeting about the story, estimated by the people who’ll do the work, and free of hidden blockers we’ve already identified. If it wasn’t ready, it didn’t enter — and that rule held even when the sprint looked light.</p><p>This is where a lot of organizations flinch, because a sprint that starts with fewer, truly-ready stories looks less ambitious than one stuffed with maybes. It is less ambitious. It’s also the reason the work started landing.</p><h4>Change 4: Track flow, not just velocity</h4><p>Finally, we changed what we watched.</p><p>Velocity told us how much we were completing in story points — a number that can be inflated, argued with, and gamed, and that says nothing about where work waits. So alongside it we tracked the flow questions: Where is work waiting right now? How long has it been waiting? Why?</p><p>Those questions changed the standup. Instead of a tour of everyone’s activity, it became a scan for stuck work — and stuck work, named early, is work a Scrum Master can actually do something about. That’s the job: treat blockers as the work, not as an interruption to it.</p><h4>The result</h4><p>Throughput up 80%. Committed-work delivery up 50%.</p><p>Same teams. Same people. No heroics, no overtime culture, no weekend war rooms.</p><p>I keep coming back to that when leaders ask what a Scrum Master actually does for delivery numbers. We didn’t add capacity. We removed what was in the way of the capacity that was already there — the excess work in progress, the surprise dependencies, the unready stories, and the metric that hid all of it.</p><p>The people were never the constraint. The flow was.</p><p>If your teams are busy and your commitments keep slipping, resist the headcount conversation for one sprint and try the flow conversation instead: What are we finishing? What’s waiting, and why? What’s the one blockage whose removal would move everything behind it?</p><p>Start there. You might be surprised how much delivery was already on your payroll.</p><p>Scrum Masters and Agile Coaches — what’s the one flow change that made the biggest difference for your teams? I read every answer, and the best ones will shape a future article.</p><p><strong>If this was useful, follow along.</strong> I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don’t miss the next one. One useful email when there’s something worth reading. No noise.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on Linked</em>I<em>n.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3215a34e62a6" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/how-we-raised-throughput-80-without-adding-a-single-person-3215a34e62a6">How We Raised Throughput 80% Without Adding a Single Person</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Quality of Your PI Planning Is Decided Weeks Before the Event]]></title>
            <link>https://medium.com/flow-over-theater/the-quality-of-your-pi-planning-is-decided-weeks-before-the-event-41052d4443f9?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/41052d4443f9</guid>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[pi-planning]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[safe]]></category>
            <category><![CDATA[scrum]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Sat, 07 Nov 2026 00:20:18 GMT</pubDate>
            <atom:updated>2026-11-08T16:16:10.724Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>The Quality of Your PI Planning Is Decided Weeks Before the Event</p><p>I earned my SAFe RTE certification after years of supporting RTEs through PI Planning. Here’s what the view from the Scrum Master seat taught me about the weeks that actually decide the event.</p><p>I earned my SAFe Release Train Engineer certification after spending years on the other side of PI Planning — supporting RTEs, not holding the title.</p><p>I want to be precise about that, because honesty about experience is kind of the whole point of this publication. I assisted RTEs through PI Planning events. I prepared teams for them, fed work into them, and lived with what came out of them. The certification came after the reps, not instead of them.</p><p>And that order taught me more than the certification did. Because when you’re the Scrum Master feeding work into an Agile Release Train, you learn — sometimes painfully — what a good PI Planning actually needs. And most of it has nothing to do with the two days of the event itself.</p><p>By the time the room fills up and the RTE opens the session, the quality of your PI is already decided. It was decided in the weeks before: in the refinement sessions, the dependency conversations, and the risks somebody either named early or saved as a surprise. Here are the three things that decide it.</p><h4>Lesson 1: Dependencies should be named before they’re discovered live</h4><p>There is a specific kind of silence that falls over a planning room when a team announces a dependency nobody knew about — in front of a hundred people, with leadership watching and the schedule already drawn.</p><p>I’ve been in that silence. It is expensive.</p><p>The dependency itself usually isn’t the problem. Dependencies are normal; teams build on each other’s work, and in a big migration or a multi-team release, they’re the water everyone swims in. The problem is the <em>timing of the discovery</em>. A dependency found six weeks early is a conversation. The same dependency found live, mid-event, is a crisis — because now it has to be solved in public, under time pressure, by people who weren’t prepared to solve it.</p><p>The fix isn’t heroic. It’s a habit: in the weeks before the event, every team should be able to answer two questions. What do we need from other teams during this PI? And what are other teams expecting from us? If a team can’t answer both, that gap <em>is</em> the finding — and there’s still time to close it.</p><p>A good PI Planning confirms dependencies. A bad one discovers them. The difference is decided in refinement, long before anyone books a room.</p><h4>Lesson 2: Risks should be stated plainly enough for leadership to decide something</h4><p>Every PI has risks. The question is whether they’re written in a language anyone can act on.</p><p>I’ve read risk statements that were technically present and practically useless — vague worries, padded with qualifiers, engineered so that no one could be blamed for them later. They fill the ROAR board (risks that are resolved, owned, accepted, or mitigated — the framework gives us the categories) without giving leadership a single real decision to make.</p><p>A risk stated plainly sounds different. It names the exposure, the date it starts costing something, and what decision would change it. “If the environment isn’t ready by the second iteration, three teams lose their integration window and the feature slips a full PI — we need a call on resourcing this week.” That sentence makes people uncomfortable. It’s also the only kind of sentence leadership can actually use.</p><p>Here’s the uncomfortable truth I learned from the supporting seat: teams soften risks when they think honesty will be punished or ignored. The RTE can run the room beautifully, but if the weeks before the event taught teams to sand down their risks, the event will ratify a fantasy. Plain risks are a culture product — and culture is built before the event, not during it.</p><h4>Lesson 3: Know your team’s real capacity, not the capacity you wish you had</h4><p>Every team I’ve worked with has two capacities. There’s the honest one — what this team, with these people, their leave schedules, their support duties, and their actual track record, can deliver. And there’s the aspirational one — what the team would commit to if everything went right and nobody got pulled into anything.</p><p>PI Planning pressure pushes teams toward the aspirational number. The room is optimistic. Leadership is present. Nobody wants to be the team that commits to less. So teams load the plan with the version of themselves that has never once shown up.</p><p>Then the PI starts, reality arrives on schedule, and the train spends ten weeks managing the gap between the plan and the truth.</p><p>A Scrum Master who knows their team’s real capacity — because they watched the last three PIs, because they track what actually got finished instead of what got started — is worth more in that room than any facilitation technique. Real capacity isn’t pessimism. It’s the raw material of a plan that survives contact with the first sprint.</p><h4>The weeks-before checklist</h4><p>Pulling it together, here’s what I now push for in the run-up to any PI Planning, from the Scrum Master seat:</p><ol><li><strong>Dependency sweep.</strong> Every team lists what it needs and what it owes. Gaps get closed in conversation, not in front of the whole train.</li><li><strong>Risk honesty pass.</strong> Each major risk rewritten until it names the exposure, the cost date, and the decision needed. If leadership can’t decide anything from it, rewrite it again.</li><li><strong>Capacity reality check.</strong> Commitments sized against demonstrated delivery, with leave, support work, and the known unknowns counted — not wished away.</li><li><strong>Refinement with teeth.</strong> Stories entering the PI already clear, estimated, and stripped of hidden blockers wherever possible. Refinement is where PI quality is actually manufactured.</li></ol><p>None of this is glamorous. All of it happens before the event, in rooms nobody photographs. That’s exactly why it works.</p><h4>The event is the receipt, not the work</h4><p>The RTE can run the room. A great RTE makes the two days smooth, focused, and genuinely useful. But the RTE is working with whatever the preceding weeks produced — named or unnamed dependencies, plain or padded risks, real or wished-for capacity.</p><p>The quality of a PI is decided in the weeks before it, in the refinement sessions and honest conversations nobody sees. The event just prints the receipt.</p><p>So if your PI Planning keeps disappointing you, don’t start by fixing the event. Start six weeks earlier.</p><p>RTEs and Scrum Masters — what’s the one thing you wish teams would do <em>before</em> PI Planning that would make the whole event work better? I read every answer, and the best ones will shape a future article.</p><p><strong>If this was useful, follow along.</strong> I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don’t miss the next one. One useful email when there’s something worth reading. No noise.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=41052d4443f9" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/the-quality-of-your-pi-planning-is-decided-weeks-before-the-event-41052d4443f9">The Quality of Your PI Planning Is Decided Weeks Before the Event</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Won’t Replace the Scrum Master. It Will Expose the Ones Who Only Run Meetings.]]></title>
            <link>https://medium.com/flow-over-theater/ai-wont-replace-the-scrum-master-it-will-expose-the-ones-who-only-run-meetings-2a9906a57e13?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/2a9906a57e13</guid>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[agile-coaching]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Sat, 07 Nov 2026 00:02:53 GMT</pubDate>
            <atom:updated>2026-11-08T16:17:11.745Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>The three-part split I use on real teams: what AI should do, what humans must keep, and how to adopt it without breaking a regulated environment.</p><p>Everywhere I look right now, the question is the same: is AI going to replace the Scrum Master?</p><p>I don’t think so. I think it’s going to do something more uncomfortable. It’s going to expose which Scrum Masters were only running meetings — and make the rest of us more valuable.</p><p>That distinction matters, because the two kinds of Scrum Master look identical on a calendar. Both book the standup. Both facilitate the retro. Both update the board. From the outside, the job looks like a series of well-run ceremonies, and ceremonies are exactly the kind of thing software is getting very good at.</p><p>But the ceremonies were never the job. They were the visible ten percent of it. The job is what’s underneath: judgment, trust, escalation, and the willingness to say the true thing in a room that would rather not hear it. AI is about to make the visible ten percent nearly free. When that happens, everyone will finally see who was doing the other ninety.</p><p>Here’s how I see it working on real teams — especially in regulated banking and healthcare, where you can’t just plug in a tool and hope.</p><h4>Let AI do the routine work</h4><p>There’s a layer of Scrum Master work that is genuinely mechanical, and I’ve never understood the romance around doing it by hand.</p><p>Summarizing standup notes. Scanning cycle time data for patterns a human eye gets tired of looking for. Flagging the blocker that has quietly shown up sprint after sprint, wearing a slightly different disguise each time. Counting, sorting, watching, reminding. This is work a machine should do — not because it’s beneath us, but because it’s <em>exhausting</em>, and exhaustion is where things get missed.</p><p>I run 4–5 Scrum teams at a time. At that scale, the routine layer isn’t a nuisance; it’s a flood. The standup where three teams mention the same dependency in three different ways — a human can catch that. A human can also miss it on a Friday afternoon in the middle of a migration. A tool that watches the patterns doesn’t get tired, doesn’t get distracted, and doesn’t have a favorite team.</p><p>So let it watch. Let it summarize. Let it raise its hand and say: this blocker has appeared four times in six weeks, and nobody has named it as a pattern yet. That’s not replacing the Scrum Master. That’s giving the Scrum Master better eyes.</p><h4>Keep the human work human</h4><p>Here’s where the replacement conversation goes wrong. The people predicting our extinction are describing the Scrum Master they can <em>see —</em> the facilitator with the timer. They’re not describing the actual work.</p><p>The actual work sounds like this: coaching a team through a hard dependency conversation without it turning into blame. Building enough trust that a retrospective produces the truth instead of the word “communication.” Looking at a dashboard glowing green and knowing — because you were in the room, because you heard the hesitation in the standup — that the metric is lying to you.</p><p>AI can surface a problem faster than I can. It cannot walk into a tense room and get five teams, their Product Owners, and their stakeholders to actually <em>fix</em> it. It cannot tell when a team is quiet because they’re focused and when they’re quiet because they’ve given up. It cannot make the escalation call to a leader two levels up and hold its ground with the cost stated plainly.</p><p>That work is not routine. It is the profession. And in my experience, the Scrum Masters who do it well have never been the ones threatened by a new tool — because a tool was never the source of their value.</p><h4>Adopt it like any other delivery change: small, visible, governed</h4><p>The third part is the one my industry can’t skip.</p><p>I work in banking, financial services, and healthcare. In these environments, “move fast and break things” is not a strategy — it’s an incident report. Data privacy comes first. Audit trails matter. When a tool touches real delivery data, somebody has to be able to answer: what does it see, where does it go, and who is accountable for what it produces?</p><p>So adopt AI the way you’d adopt any delivery change worth making. Start small — one team, one use case, one visible problem. Make the results observable instead of assumed. Govern it like it matters, because in a regulated environment it does. The teams I’ve seen get real value from new tooling are never the ones who rolled it out everywhere on day one. They’re the ones who proved it in one place and let the evidence do the persuading.</p><p>That’s not caution for its own sake. That’s just delivery discipline applied to a new kind of change.</p><h4>The split is the strategy</h4><p>Step back and the whole argument resolves into a simple split:</p><ul><li><strong>Routine work goes to the machine.</strong> Notes, patterns, counts, flags, reminders. Faster and more tireless than any of us.</li><li><strong>Human work stays human.</strong> Coaching, trust, judgment, escalation, truth-telling. The work that only exists because people are complicated.</li><li><strong>Adoption gets governed.</strong> Small, visible, accountable — especially where regulators are watching, which in my world is everywhere.</li></ul><p>Scrum Masters who built their identity on the first category should be nervous, and honestly, they should have been nervous before AI arrived — running meetings was never a defensible career. The rest of us get a better toolkit for the work we were already doing: flow over theater, blockers treated as the job, now with better tools for seeing the blockers sooner.</p><p>AI won’t replace the Scrum Master. It will replace the <em>performance</em> of being one. For those of us actually doing the work, that’s not a threat. That’s a promotion.</p><p>Scrum Masters, Agile Coaches, and RTEs — where has AI actually helped your teams deliver, and where has it just added noise? I read every answer, and the best ones will shape a future article.</p><p><strong>If this was useful, follow along.</strong> I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don’t miss the next one. One useful email when there’s something worth reading. No noise.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2a9906a57e13" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/ai-wont-replace-the-scrum-master-it-will-expose-the-ones-who-only-run-meetings-2a9906a57e13">AI Won’t Replace the Scrum Master. It Will Expose the Ones Who Only Run Meetings.</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Start Here: Who I Am and What Flow Over Theater Is About]]></title>
            <link>https://medium.com/flow-over-theater/start-here-who-i-am-and-what-flow-over-theater-is-about-ddccaa4f01a5?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/ddccaa4f01a5</guid>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[agile-coaching]]></category>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[about]]></category>
            <category><![CDATA[agile]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Fri, 06 Nov 2026 21:47:45 GMT</pubDate>
            <atom:updated>2026-11-09T16:31:58.988Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>I’m Doug DeFranco — Senior Scrum Master and Agile Coach. Ten years, three industries, one belief: flow over theater.</p><p>I’m Doug DeFranco. For the past 10 years I’ve worked as a Senior Scrum Master and Agile Coach in banking, financial services, and healthcare — including Capital One, First Citizens Bank, and Anthem/Elevance Health — running as many as 4–5 Scrum teams at a time, including large-scale migrations of card systems to AWS (500+ workloads).</p><p>My credentials, for those who check: <strong>Senior Scrum Master | SAFe Release Train Engineer (RTE) | CSM, PMP | Agile Coach (ICP-ACC, ICP-ATF)</strong> — also certified in SAFe 6.0, 4.5 &amp; 4.0, AWS Certified Cloud Practitioner, and Cybersecurity Fundamentals. The full, verifiable record is on my LinkedIn.</p><p>But credentials aren’t why this publication exists.</p><h4>Why “Flow Over Theater”?</h4><p>Because most of what gets called Agile in big organizations is theater: standups that are status reports in costume, retrospectives where everyone says “communication” and nothing changes, metrics dashboards that glow green while the work quietly drowns.</p><p>Flow is the opposite of that. Flow is work actually moving — blockers named out loud and removed, dependencies made visible before they explode, teams telling the truth in the retro because they’ve watched the truth change something.</p><p>Everything I publish here comes from rooms I was actually in:</p><ul><li>Throughput up 80% and committed work up 50% — not by adding people, but by removing what was in the way.</li><li>What supporting Release Train Engineers through PI Planning taught me before I ever certified as an RTE.</li><li>Why I stopped reporting velocity, and the flow metrics I watch instead.</li><li>The retrospective rules that finally made my teams say the real thing.</li></ul><h4>How to read this publication</h4><p>New articles land <strong>Monday, Wednesday, and Friday</strong>, alongside my LinkedIn posts — the post is the short version; the article here is the full story. Start with the first one:</p><p>👉 <a href="https://medium.com/flow-over-theater/the-most-honest-meeting-in-agile-is-the-one-your-teams-are-faking-5e9684e901b5"><strong>The Most Honest Meeting in Agile Is the One Your Teams Are Faking</strong></a> — why your retrospectives keep producing the word “communication,” and the three rules that fixed mine.</p><p>If you run teams, coach them, or lead the people who do — follow along. This is Agile the way it actually happens: flow over theater.</p><p>— Doug DeFranco</p><p>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach</p><p><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ddccaa4f01a5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/start-here-who-i-am-and-what-flow-over-theater-is-about-ddccaa4f01a5">Start Here: Who I Am and What Flow Over Theater Is About</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Most Honest Meeting in Agile Is the One Your Teams Are Faking]]></title>
            <link>https://medium.com/flow-over-theater/the-most-honest-meeting-in-agile-is-the-one-your-teams-are-faking-5e9684e901b5?source=rss-8718785d6240------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e9684e901b5</guid>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[scrum-master]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[retrospectives]]></category>
            <category><![CDATA[agile-coaching]]></category>
            <dc:creator><![CDATA[Doug DeFranco]]></dc:creator>
            <pubDate>Fri, 06 Nov 2026 19:55:30 GMT</pubDate>
            <atom:updated>2026-11-08T16:17:35.506Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GD7hiyMminVzB_F8n_PtMQ.png" /></figure><p>Why your retrospectives keep producing the word “communication” — and the three rules that finally made mine tell the truth.</p><p>The most honest meeting in Agile is the one most teams are faking.</p><p>The retrospective.</p><p>I’ve sat in retros where the same word came up six sprints in a row: “communication.”</p><p>Six. Sprints. In a row.</p><p>Nobody in those rooms was lying. That’s the part people miss. The sticky notes were full, the board looked busy, everybody participated — and nobody said the real thing. Not once. Because the real thing was uncomfortable. Or it pointed at someone outside the room. Or — and this is the one that does the real damage — the last time somebody raised it, nothing happened.</p><p>So the team learned. They learned that “communication” is a safe word. It sounds like insight. It commits nobody to anything. It lets the meeting end on time with everybody’s relationships intact. And it guarantees that the actual problem — the approval that takes nine days, the dependency that shows up late every single sprint, the decision that keeps getting un-decided — walks out of the room with you and into the next sprint, untouched.</p><p>A retrospective where nothing changes teaches the team one lesson: stop being honest.</p><h3>Why teams fake it</h3><p>It took me longer than I’d like to admit to understand that a quiet retro is rarely a lazy team. It’s a rational one.</p><p>People tell the truth in a retrospective when they believe two things: that it’s safe to say it, and that saying it matters. Most teams have watched at least one of those beliefs die. Somebody named a real problem once — out loud, in front of everyone — and it got awkward, or it got defended, or it got written down and never mentioned again. The team didn’t become cynical after that. They became accurate.</p><p>You can’t fix that with a new retro format. I’ve tried the formats. Sailboats, starfish, timelines, mad-sad-glad — they’re fine tools, but a format can’t repair trust. What repairs trust is evidence: the team watches a problem they named actually get fixed. Do that a few times and the sticky notes start telling the truth on their own.</p><p>Here’s what actually works, after 10 years of running these rooms.</p><h3>Rule 1: One real problem beats five safe ones</h3><p>Depth over breadth, every time.</p><p>If your retro ends with seven action items, you almost certainly have zero. A long list is how a team spreads responsibility so thin that nobody is accountable for anything — and it usually means the conversation stayed on the surface, collecting small, comfortable complaints instead of following one thread down to where it actually hurts.</p><p>So I push for the opposite: what is the <em>one</em> problem that, if we fixed it, would make the next sprint genuinely better? Then we stay on it. We ask why it’s happening, and then why that’s happening, until the room gets a little uncomfortable — because discomfort is usually the sign you’ve arrived at the real work.</p><p>Here’s my test: if the retro ends and nobody felt a moment of discomfort, you probably didn’t get to the real work. Not conflict. Not blame. Just the brief, healthy discomfort of saying a true thing out loud that everyone already knew and nobody had said.</p><p>One problem, owned by name, with a next step somebody can actually do. That’s a retro. Everything else is a meeting about a meeting.</p><h3>Rule 2: If the problem lives outside the team, say so — out loud</h3><p>This is the rule that changed my career more than any framework ever did.</p><p>Watch a typical retro and count how many of the real blockers are actually inside the team’s control. In my experience, roughly half the blockers I’ve removed over the years were never the team’s to solve. The dependency owned by another group. The environment that isn’t ready. The approval sitting in someone’s inbox two levels up. The vendor deliverable that slipped three weeks ago and quietly rearranged everybody’s sprint.</p><p>Teams know this. But there’s a strange guilt in Scrum culture that says a good team should absorb everything — so the outside problem gets written down as the team’s failure, the team “commits” to fixing something it doesn’t control, and two sprints later it comes back, heavier, with a layer of shame on it.</p><p>Name it honestly instead: <em>this one isn’t ours.</em> Then do the unglamorous work that is actually the Scrum Master’s job — get it in front of the people who <em>can</em> fix it, with the cost stated plainly. Not “we’re blocked,” but “here is the date this starts costing us, and here is what moves if it lands.” Say it in the room, say it in the open, and then follow up until it moves.</p><p>Half the value I’ve delivered in ten years has come from that paragraph. Not from facilitating a prettier standup. From refusing to let a team carry a problem it never owned.</p><h3>Rule 3: Treat retro actions like sprint work</h3><p>This is the trust engine, and it’s embarrassingly simple.</p><p>Retro actions get an owner. They go somewhere visible — the same board, the same backlog, the same light of day as the “real” work. And at the <em>next</em> retrospective, before anyone says one word about the new sprint, we check them. First thing. Every time.</p><p>Done? Good — say so out loud, because the team needs to hear that the system worked. Not done? Also fine, but now it’s a real conversation: what got in the way, is it still worth doing, does it need a different owner or a bigger escalation?</p><p>What you cannot do is skip the check. The moment a retro action disappears into the void, you’ve confirmed every suspicion the team brought into the room. And the moment a team watches last sprint’s problem actually get fixed — the approval that now takes two days instead of nine, the dependency that now shows up on time — something physical changes in that room. Shoulders drop. The sticky notes get braver. The word “communication” starts retiring, one sprint at a time.</p><p>Trust in a retrospective isn’t built by the facilitator’s warmth. It’s built by follow-through the team can see.</p><h3>What to do Monday morning</h3><p>If your retros have gone quiet, don’t redesign them. Start smaller than that:</p><p>1. <strong>Open with last time.</strong> Before any new discussion, put the previous retro’s actions on the screen and check them — done, not done, and why. Five minutes. Non-negotiable. This single habit does more for honesty than any format you’ll ever download.<br>2. <strong>Force the choice of one.</strong> When the board fills up, ask: “If we could only fix one of these before next sprint, which one is costing us the most?” Take that one seriously. Park the rest without guilt.<br>3. <strong>Sort by ownership.</strong> For the chosen problem, ask out loud: “Is this ours, or does this live somewhere else?” If it lives somewhere else, your action item isn’t “try harder” — it’s a name, a person or team who owns it, and a date you’ll follow up.<br>4. <strong>End with a name and a date.</strong> Every action leaves the room with a person’s name on it and a check-in point. “The team will communicate better” is not an action. It’s a wish.</p><p>Do those four things for three sprints and watch the room change.</p><h3>The dashboard won’t save you</h3><p>Here’s the thing I keep coming back to. Every organization I know is hunting for delivery gains — in dashboards, in tooling, in new frameworks, in AI summaries of meetings nobody was honest in. But the biggest gains I’ve ever seen were hiding in a small room, in the last hour of a sprint, where a team either believed the truth was welcome or had learned it wasn’t.</p><p>That’s where the delivery gains nobody can find in a dashboard are hiding. Not in a new tool. In a room where the team believes telling the truth changes something.</p><p>Build that room, and the metrics everyone is chasing tend to take care of themselves.</p><p>Scrum Masters and Agile Coaches — what’s the most honest thing a team ever said in one of your retrospectives, and what actually changed after? I read every answer, and the best ones will shape a future article.</p><p><strong>If this was useful, follow along.</strong> I write about Agile the way it actually happens — flow over theater, blockers treated as the work — drawn from 10 years running 4–5 Scrum teams at a time in banking, financial services, and healthcare. Follow this publication so you don’t miss the next one, and if you want the free Retro Action Tracker template I use to run Rule 3 (retro actions owned, visible, and checked first thing), join the email list — the link is in my profile. One useful email when there’s something worth reading. No noise.</p><p>— Doug DeFranco<br>Senior Scrum Master | RTE, CSM, AC, ATF, PMP | Agile Coach<br><em>Also certified: SAFe 6.0, 4.5 &amp; 4.0 · AWS Certified Cloud Practitioner · Cybersecurity Fundamentals. Full certification record on LinkedIn.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5e9684e901b5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/flow-over-theater/the-most-honest-meeting-in-agile-is-the-one-your-teams-are-faking-5e9684e901b5">The Most Honest Meeting in Agile Is the One Your Teams Are Faking</a> was originally published in <a href="https://medium.com/flow-over-theater">Flow Over Theater</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>