Kanban for Remote Teams

Given the significant move across a wide number of industries to remote work and working from home, we recently hosted a Webinar focusing on Kanban for Remote teams. A big thanks to everyone who attended as well as our 3 guests Sonya from Nave, Mahesh from Digité, and Dimitar from Kanbanize for giving us such great overviews of their products.

This post curates a few things from the Webinar, specifically capturing the Q&A, the video and the presentation slides.

Webinar Slides

Webinar Video

Webinar Q&A

Q: 1) how much effort should we invest in digitization of everything that was not digital? 2) any thoughts about dealing with  Continuous improvement during this pandemic time?3) what symptoms can you describe to realize that some cadences should be reviewed?

A: Answered in the video Q&A section timestamp 1hr,22m

Q: How to use the discount promo codes when signing up for the public training sessions? don’t see any option here: https://www.squirrelnorth.com/events/online-june-kanban-system-design-kmp-i

A: You enter the discount code in the “checkout” screen, after adding tickets to your cart, and when you’re ready to pay.

Q: What is the thinking behind Sonja’s statement, “You shouldn’t use your CFDs for future predictions?

A: Possible meaning: the bottom line of the CFD shows the trend based on the average throughput, actual outcomes may deviate significantly from that trend.

Q: Martin, Fernando, this looks like the same function as your spreadsheets to create the histogram, correct?

A: Technically true. This solution (Swift, Nave, Kanbanize) is more automated and allows for more advanced configuration.

Q: Re: empathy during the shift to working from home, this is doubley true when our business is impacted by reductions in force, a decision made for the business to survive. With multiple impacts to a given service's capability, what recommendations do you have for SDM/Delivery team's to educate stakeholders/customers that we have no predictability to deliver the new demand created to survive this new environment? Especially when the goal of the new demand is to support both our business and our customers survive in the current environment, causing unpredictability to be more uncomfortable than normal.

A: Answered in the video Q&A section timestamp 1hr,22m

Q: How many data points are required to have a meaningful Monte Carlo simulation?

A: Alexei: The number of data points is not the constraint. The key is a good, if imperfect, understanding of the real phenomenon we want to simulate with Monte Carlo. This is a complicated question, more appropriate for a consultation. Martin: In our Agile Metrics class, we demonstrate fairly accurate forecasting with just 5 data points. But as with all models, accuracy changes over time with more data improving your model. Best to think of it as "most accurate" vs "correct". We can have a longer chat about this if you want to go a bit deeper.

Q: How does Nave differentiate itself from Actionable Agile? Or is Actionable Agile the foundation of Nave?

A: Nave: We are a customer-centric company, and we have evolved our product based on our customers' feedback. We are fast, we provide affordable services, and outstanding customer support. My best suggestion would be to try it out and decide which tool is F4P.

Q: Are the three tools best for one function/feature or all they all comparable? Why would I use one over the others? Thanks!

A: Karen, you as a customer would select one of these products over others because it's a better fit by your fitness criteria (functional, non-functional quality, etc.) connected to your purpose for using the product. Martin: There is no best product, you need to find the one that matches your needs best.

Q: Hi everybody and thank you for this great webinar! Is there some discount or benefit for Accredited Kanban Trainers (Kanban University) or consultants for using and promoting these tools?

A: Jose, yes, please follow up with KU (Inga Lenz, Todd Little) as well as with the three tool vendors.

Q: Regarding Monte Carlo forecasting, using a sample of historical throughput per scenario makes sense and is straightforward to accomplish with simple tools (e.g. Python script). Is there an argument/benefit to using historical aging or lead-time data? A: Throughput forecasting is valid (and simplifies the math significantly) when the lead time of work items is thin-tailed (less variability than the Exponential distribution). Invalid otherwise.

Thanks again, we hope to see you at the next one!

The SquirrelNorth Team