Drunk Agile: Recent Episodes
Dan Vacanti & Prateek Singh
Dan Vacanti and Prateek Singh drink whisk(e)y and discuss various facets of agile software development. From scotch to bourbon and from metrics to human behaviour, we cover it all.
Dan and Prateek have thoughts on empiricism. How it is not all that it is cracked up to be. Find out more in this episode of Drunk Agile.
Have you been given the advice - "Slice stories so they fit within the sprint length"? There are numerous issues with this advice. Prateek and Dan take these on in the latest episode of Drunk Agile.
Should you use a Cycle Time Scatterplot of Epics or a Monte Carlo simulation of stories to figure out when an Epic would be done? Daniel Vacanti and Prateek Singh discuss this in the latest episode of #DrunkAgile.
Ever planned a sprint using sub-tasks? Dan and Prateek discuss how this is an anti-pattern that seems to show up often on Scrum teams. The latest episode of Drunk Agile also discusses this problem at Scale.
Dan and Prateek have a couple of questions for you. What was your favourite episode in 2024?What content would you like to see in 2025?Please respond on Youtube.
Nisha, Dan, and Prateek discuss what the 'Done' column on your board means and how it is not what most people think it is.
Enjoy the latest release of DrunkAgile where Dan and Prateek cover an example of the Process Behaviour Chart.This episode was a request from multiple viewers. Please keep sending these our way. They give us material to record and excuses to drink whisky!
Our contract negotiations with Nisha were successful. She is back and has some opinions about sprints and other fixed-length timeboxes. Daniel Vacanti and I might have a couple of things to say about this as well.
We are in the midst of contract negotiations with #Nisha. #Spartacus subs in as Dan and Prateek talk Control Charts in the latest episode of #DrunkAgile.
Nisha believes that the type of prioritization most Agile frameworks tell people to spend time on is the least important. It is also the most wasteful. Dan and Prateek wholeheartedly agree.
How do Nisha, Dan, and Prateek decide the content for the next episode of hashtag#DrunkAgile? Come find out the answer to the question in the next episode of our Prioritization series. We are on to Level II of prioritization.
We finally prioritized(or did we) doing the episodes. Nisha, Dan, and Prateek talk about the three levels of prioritization, starting with Level I - Prioritization of Active Work.
Does your board have an 'In Progress' column? #Nisha thinks that is not a great idea. Find out why on the latest episode of DrunkAgile.
Nisha, Dan, and Prateek are back talking about product and flow.How the most common Agile frameworks have misguided people in product roles and had them focus on the wrong things. Find out more in the latest episode of DrunkAgile.
"Why do you hate Agile?"Nisha loves Agile, but do Dan and Prateek?Find out in the latest episode of #DrunkAgile.
Colleen Johnson visits Nisha, Dan, and Prateek to discuss what is coming up for ProKanban.org. They talk about the various roles like Scrum Mastesr, managers, and team leads who can be better equipped to help teams succeed. As Agile roles keep getting put on the chopping block, how can learning more about kanban and flow help you stand out?
Nisha, Dan, and Prateek wrap up 2023 by looking back at our most-watched episodes of 2023 (released and watched in 2023). Which ones were your favourites? Were there topics you were expecting to see that we did not cover?
What happens when you have reduced cycle time to the point that it matches your customer's needs? Should you try to reduce cycle time further?#Nisha, Daniel Vacanti, and Prateek Singh discuss this in the latest episode of #DrunkAgile.
Nisha, Dan, and Prateek explain the ideas of the Voice of the Process and the Voice of the Customer. Is your Sprint length VoP or VoC?
Nisha, Dan, and Prateek talk about all the problems with dependency management. How focusing on dependency elimination instead, can produce better results. All that, with a good dose of rambling thrown in.
Which prioritization technique do you use? Does it even matter? Is ordering your backlog waste (Yes, it is)? What if you could save all the time you spend prioritizing and still get better results?#Nisha, Dan, and Prateek discuss how Just in Time decisions and limiting WIP are far superior to having a prioritized backlog, in the latest #DrunkAgile episode.
Nisha, Dan, and Prateek cover their preference for creating stable systems over spending time in planning sessions. What does planning have to do with Dante's circles of hell? If Agile is about continuously learning, can you really plan in advance what you will get done?
Chris McDermott stops by to talk to Nisha, Dan, and Prateek about Change Mapping. How do you look at your practices in context to figure out what to keep, improve, or discard? Come find out in this episode of #DrunkAgile. Also: Happy Birthday Nisha!
Nisha, Dan, and Prateek discuss the dogma that agile teams have to be small. They discuss their own experiences of working on large teams that were very effective, efficient, and predictable.
Dan and Prateek stick to the same whiskey and the same topic from the previous episode - Timeboxes. This time though we go deeper into the risk and complexity associated with them. Particularly in #Agile environments.#Nisha is as active as ever, on this latest episode of #DrunkAgile.
Nisha, Dan, and Prateek are back after a bit of a break. The latest episode of #DrunkAgile is brought to you by the word 'Arbitrary'.From the OED: 'arbitrary, adj. & n. - Derived from mere opinion or preference; not based on the nature of things'
Does variability directly imply that your data is distributed in some manner?How important is it to figure out which distribution your data follows?Are you even analyzing your data if you are not using a cubic spline?Nisha, Dan, and Prateek discuss.
Does your team get done with work early? Or does your team get done with work a few days late? Are these problems to be addressed? Is it really a problem that your #scrum team got done with the #Sprint goal on day 10 of a 14-day sprint? What if they look like they will get done on day 16?
Do you know what are the 'Two Mistakes' according to Shewhart?
Nisha, Daniel Vacanti and I talk about these in this week's episode of #DrunkAgile.
Episode 70 of #DrunkAgile is a start of a 3-part(at least) series on Variability. #Nisha, Dan, and Prateek start off with a high-level introduction to what the word variability means in our context.
In episode 69 of #DrunkAgile, #Nisha, Dan, and Prateek talk through a replacement for INVEST as work-item splitting criteria. Re-emphasizing our preference for right-sizing and feedback over value and estimation.
Nisha, Dan, and Prateek answer an audience question about how to deal with blockers that are outside the team's control. They take a little detour into Little's Law as well.
What do WIP Limits have to do with a team's capacity... turns out, everything. #Nisha, Dan, and Prateek dive deeper into this question in the latest #DrunkAgile episode. Enjoy!
You asked for Blocker policies and #Nisha delivered. In this week's #DrunkAgile episode, Dan, and Prateek discuss how to set some policies around dealing with blocked work.
Should you collect Blocker data? If yes, how do you use it? If not, what is an alternative?#Nisha, Dan, and Prateek discuss this in the third episode recorded on the same night. #DrunkAgile tending towards #PlasteredAgile.
Nisha, Dan, and Prateek answer an audience question about #MonteCarlo and the predicted date ending on a weekend. What does this mean in terms of #risk? Does the risk change based on how far out the forecast is?
Nisha, Dan, and Prateek talk through the effect weekends have on cycle time. Why they are included and what would be the effect if we did exclude them? Note : This episode includes the use of visuals and might be better experienced on Youtube.
Nisha, Dan, and Prateek talk through the applicability of Flow Metrics in Scrum. Also, they are 'Flow Metrics', not 'Kanban Metrics'. ProKanabn.Org's new AFMS course covers this as well.
In this episode, Dan and Prateek inspect the results of Monte-Carlo forecasts for when in the season would LeBron James break Kareem Abdul-Jabbar's record. Nisha does her best airbud impression.
The last episode in the Flaw of Averages series. Nisha, Dan, and Prateek answer a distillation of multiple audience questions - What is the problem with single-point answers?
The Flaw of Averages Part II. In this sequel, Dan and Prateek take on data distributions and shapes. #Nisha calls out the bull in Weibull. It is normal for your data to not be normal.
Nisha, Dan, and Prateek discuss Dr. Sam Savage's Flaw of Averages in this episode of #DrunkAgile. Have you been thinking of the world in terms of averages? If yes, hopefully, this episode will cure that ailment.
ProKanban.org hosts Nisha, Dan, and Prateek for a live edition of Drunk Agile. In this Holiday Special, the Drunk Agile crew takes on a variety of questions from the audience.
Nisha, Dan, and Prateek cover an audience question - I have a new feature, not yet started, when will that feature be completed? Come find out the answer.
Everything you have been told about Scaling is wrong! Nisha, Dan, and Prateek cover what is so different about Applying Scaled Portfolio Kanban.
Nisha, Dan, and Prateek talk about being Agile when implementing Agile. They discuss the pitfalls of Maturity Models and more...
Nisha, Dan, and Prateek are back with some whisky and revisiting thoughts around Ex-Ante vs Ex-Post in order to answer an audience question looking for "guidance on finding the sweet spot to achieve meaningful metrics without too much waste in the form of ex-Ante Analysis."
Should #Kanban boards have a "Ready" column? If they do, what's the difference between that column and a "Backlog"? Dan, Prateek, and #Nisha discuss this on this week's episode. A special thanks to Andy for the question!
Dan and Prateek (with a Nisha slideshow) take the stage at Lean Agile London to answer questions from a live audience.
Is each pull request a separate work item? How do Dan, Prateek and Nisha suggest we correlate the two? Hint - It is all about your flow policies.
Nisha, Dan and Prateek talk about observations of flow outside of product development. They discuss how flow management policies have a great effect on the performance of any kind of process.
Nisha, Dan and Prateek discuss the merits and drawbacks of Product Backlogs. Can a team or and organization actually be successful and achieve results without a Product Backlog?
Dan and Prateek answer an audience question - Is Kanban Better Suited For Mature Agile Teams? Nisha helps as usual.
James Scrimshire joins Nisha, Dan and Prateek to discuss the problems with trying to follow a big scaling recipe. Are you seeing problems with one of the popular scaling frameworks? You are not alone. Listen in to find out what might be causing those issues.
Potentially the most important concept in Agile that no one talks about is the difference between ex ante and ex post. Much of Agile assumes ex ante knowledge that you just don't have. #Nisha, Prateek, and Dan discuss on this week's episode of Drunk Agile.
Want to forecast a release but don't know exactly how many items are in your backlog? Want to know when a bunch of features will finish but haven't broken them down yet? Find out how in this week's episode of Drunk Agile with Nisha, Dan, and Prateek!
Most #agile frameworks love to talk about "done" and most forecasting is centered around done. But how do we forecast when something will begin? (spoiler alert: or do we?) Join Nisha, Dan, and Prateek to find out in this week's episode of Drunk Agile.
Dan, Prateek and Nisha discuss how to answer the question of "When will it be done" for an item that is still in the backlog. Can Monte Carlo help?
Jose Casal joins Nisha, Dan and Prateek for this week's edition of Drunk Agile where we talk about OKRs. Well, Jose and Prateek talk and Dan just channels his inner Demming.
Nisha, Dan and Prateek answer an audience question about when is it appropriate to use Cycle Time Scatterplot vs Monte Carlo for predictions.
Dan, Nisha and Prateek host Chad and Jeff as they talk about the shipping problems of the world and of course... aging.
Johanna Rothman comes back to visit Nisha, Dan and Prateek to talk about more aspects of aging. More whisky, more insights, more bigger aging.
Karl Scotland joins Nisha, Dan and Prateek as they explore definition of predictability and the metrics that could represent it.
Nisha, Dan and Prateek discuss leading and lagging indicators, the difference between the two and whether that difference is important or not. There is a decent dose of Little's Law thrown in for good measure.
Julia Wester stops by to talk about Board Design antipatterns with Nisha, Dan and Prateek. Come find out why your board design might be critical to how work is or is not flowing. Also... Scottish Gin!
Nisha, Dan and Prateek discuss the answer to "How much data is needed to generate reliable forecasts?". Spoiler alert - It is less than you think... and you might be asking the wrong question.
Dan and Prateek channel Ted Lasso's Roy Kent and Coach Beard. They get pedantic about Cycle Time and Lead Time. Ted Lasso(#Nisha) mostly stays out of it in the latest #DrunkAgile episode. What is your opinion on the distinction(or lack thereof) between these two terms? #metrics
Nisha, Dan and Prateek are back to their curmudgeon-y best as they discuss the metrics of flow as described by SAFe. The whisky helped quite a bit on this one.
Nisha, Dan and Prateek have a conversation with Yuval Yeret around Safe, Kanban, Scrum and the intersection of all these.
Dan, Nisha and Prateek discuss applying flow metrics to Features, especially the concepts of Right Sizing and Aging. As usual, whiskey!
Dan and Prateek discuss the common myth that items on a kanban board dont move backwards.
Daniel and Prateek discuss myths of #Scrum and #Kanban as they taste 3 (high proof) bourbons. Which of these agile related myths do you believe?
Mark Kilby joins Dan, Prateek and Nisha to talk about remote Scrum, hybrid teams and the future of remote work.
Dan and Prateek have a conversation with Johanna Rothman about management, teams, flow, agile and averages. Find out why managers need to watch their own cycle times!
Dan and Prateek talk about (of all things)...people. Please join us as we share some of our thoughts on teams and teamwork. At the very least, join them for your daily dose of #Nisha.
Dan and Prateek talk about the use of flow metrics during Scrum events. They dive into how metrics, usually associated with Kanban are very useful for Scrum teams and can help teams level up their agile implementation.
Probably the most common question asked during or after #kanban training is - How do we set WIP limits? Dan, Nisha and Prateek attempt to answer it in our latest #drunkAgile episode.
Dan and Prateek talk to Greg Boswell on how kanban evolved at Corbis and how the journey has continued since.
Dan and Prateek wrap up the kanban series by talking about improving your workflow. They dive into how kanban teams can utilize what they are learning from their systems to improve their process.
Why are people who let people cut in line the worst? In the follow up to Episode 17 - Defining Your Workflow, Dan and Prateek talk about the answer to that question and how to effectively operate your workflow. Part 2 of the kanban series form DrunkAgile.
Dan and Prateek dive into some common "math" issues with interpreting Cycle Time data. Learn why histograms are mostly a terrible way to display time-series data. What are alternatives that might provide better answers.
Dan and Prateek start a series of conversations about what constitutes a Kanban system. In this episode, step 1 - Defining your Workflow.
Dan and Prateek discuss CFDs. When should you use them and when should you not? What is the correct way to construct one? What are the common pitfalls? Also, shout out the Muppets with Statler and Waldorf.
Like most things, there's probably more myths than truths about Little's Law out there on the interwebs. How many myths do you know about? Join Dan and Prateek on the next episode of #DrunkAgile to find out!
Dan and Prateek take a deep-ish dive into Little's Law. They cover the theorem and the assumptions behind it and why is it important to understand them. They explain how paying attention to Little's Law and its assumptions helps create a stable, predictable system.
Dan and Prateek discuss the basic metrics of flow. WIP, Cycle Time, Throughput and an important one that you probably haven't heard of.
What does Seth Stephens-Davidowitz's 'Everybody Lies' tell us about predicting value based on customer feedback? Coke vs Facebook - Which model might work better for you? The conclusion of the Inept and Inapt series.
Dan and Prateek talk about the pitfalls of using Cost of Delay and CD3 for prioritization. They talk about Mike Ditka, Tom Brady and the NFL draft as an example of misjudgements of value. Monte Carlo simulations incorporating value are used to highlight the problems with CD3. We go all the way back to Episode 8 to find an alternative approach, ie. Right Sizing.
Note - Some parts of this podcast use visual aids. For the full experience, please refer to the Drunk Agile youtube channel - www.youtube.com/channel/UC758reHaPAeEixmCjWIbsOA
Dan and Prateek continue their series on #IneptAndInapt. Dan does a deep dive and explains #CostOfDelay and CD3. Learn why Cost of Delay and especially CD3 are dangerous to your health. Will #Nisha stay awake for this whole episode? #flow #agile Note - Some parts of this podcast use visual aids. For the full experience, please refer to the Drunk Agile youtube channel - www.youtube.com/channel/UC758reHaPAeEixmCjWIbsOA
Dan and Prateek take a closer look at the story of Wilt Chamberlain's 100 point game. What can we learn from the decisions he made? How does this tie in to the theme of #IneptAndInapt that we are exploring. If you have the motivation, the data and the outcomes all available would you make the right decision? What Would Wilt Do? Conversations about how individuals and teams fail with #InspectAndAdapt and are often #IneptAndInapt when interpreting data. Biases, data, whisky and much more on the latest episode of #DrunkAgile.
Dan and Prateek talk about the not-often-talked-about concept of right sizing. They discuss how limiting your WIP is not enough in this first episode of the #IneptAndInapt series.
Colleen Johnson, CEO of ProKanban and ScatterSpoke joins Dan and Prateek for a drink. They talk retrospectives, data, ProKanban and creating an inclusive community. Colleen takes us through the journey of how working directly with users helped create ScatterSpoke and when to ignore user requests/feedback. Also - Do feelings matter?
Dan and Prateek field a couple of listener questions. Why use daily throughput instead of weekly throughput? Why not use feature level metrics to forecast features?
Note - Some parts of this podcast use visual aids. For the full experience, please refer to the Drunk Agile youtube channel - www.youtube.com/channel/UC758reHaPAeEixmCjWIbsOA
Dan and Prateek take on using Monte Carlo simulations to predict probability of features getting done. They talk about the overwhelming effect that Feature WIP (number of simultaneous features in progress) has on your chances to finish your highest priority features. Note - Some parts of this podcast use visual aids. For the full experience, please refer to the Drunk Agile youtube channel - www.youtube.com/channel/UC758reHaPAeEixmCjWIbsOA
Dan and Prateek talk about 'What not to do' when using Monte Carlo simulations to forecast work. They cover problems with using split rates, arrival rates and curve fitting. Discussions include a brief overview of the powerful technique of continuous forecasting. As always, more whisky and more #Nisha.
Dan and Prateek talk about forecasting in a probabilistic world. They demonstrate how to use Monte Carlo simulations to answer - "When will it be done?". How do we use what we know about the system to predict what we can get done using some probabilistic forecasting techniques.
Note - Some parts of this podcast use visual aids. For the full experience, please refer to the Drunk Agile youtube channel - https://www.youtube.com/channel/UC758reHaPAeEixmCjWIbsOA
Dan and Prateek draw the distinction between deterministic and probabilistic thinking. They use everyday examples to describe what forecasting looks like in a probabilistic world.
In episode 1 of Drunk Agile, Dan Vacanti and Prateek Singh explore the problems with using story points for estimation. We try some whisky and talk agile software development.