BONUS: Unleashing Agile Experimentation, Accelerating Learning Cycles With 24h Experiments, With Vasco Duarte
Read the full Show Notes and search through the world’s largest audio library on Scrum directly on the Scrum Master Toolbox Podcast website: http://bit.ly/SMTP_ShowNotes.
Merry Christmas, everyone! In today's Christmas week BONUS episode, we're diving into the world of Agile experimentation, specifically exploring how to enable 24-hour experiments—a topic that promises to revolutionize the way we approach product development. As we explore this exciting subject, remember that each episode this week is a companion to the Coach Your Product Owner e-course, accessible at bit.ly/coachyourpo.
This week's episodes are designed to equip Scrum Masters with essential tools to assist their Product Owners in mastering five key areas. Today, we unravel the secrets of conducting rapid experiments that speed up your learning and development cycles.
First: Defining Experiments and Their Purpose in Product DevelopmentLet's start by revisiting the Lean Startup methodology and its famed Build-Measure-Learn cycle. Lean Startup emphasizes that the purpose of "building" is not merely to deliver software but to validate hypotheses. Before running an experiment, it's crucial to define what you're trying to achieve, as discussed in Monday's episode on how to define value.
Taking Amazon’s website as an example, suppose a team aims to boost the conversion rate for users who have added items to their cart. The hypothesis could be: "Removing shipping costs from the checkout overview will result in a higher conversion rate for people who've added an item to their cart." This hypothesis, or belief, becomes the trigger for experimentation.
Second: Designing the ExperimentOnce the hypothesis is clear, it's time to design the experiment. Think of it as crafting a story, but with a focus on experimentation. Here's a template to structure your experiment:
Title: Increase Conversion Rate for People Who've Added an Item to the Cart by Removing Shipping Costs from the Checkout Overview.
We Do: Specify the actions to set up the experiment.
We Expect: Clearly state the anticipated outcomes.
Data We Collect: List the crucial data points needed for analysis.
Conclusion: Outline the lessons learned and set a date for analysis.
This structured template ensures clarity and sets the stage for a comprehensive understanding of the experiment.
Third: Designing 24-Hour ExperimentsBefore we start brainstorming stories and features that would need to be implemented by the development team, we should think about running experiments in 24 hours or less. We do this, because we want to avoid one of the major sources of waste in our processes: doing work that would not need to be done at all in the first place!
As we explore the experiment above, we define the key question as, "Would a person not shown shipping costs be more likely to complete a purchase after adding an item to the cart?"
To achieve faster learning, consider alternatives to software changes. In this episode, we explore some concrete alternatives to writing and deploying software when it comes to answer the key question defined above.
One example we explore is: conducting usability tests using paper prototypes with a small group. Through qualitative analysis and user feedback, you can gain confidence in your hypothesis without investing time in extensive software modifications.
Running 24h Experiments: SummaryIn summary, today's episode tackled critical aspects of designing and running experiments, emphasizing the importance of learning. Key takeaways include:
Lean Startup's Build-Measure-Learn cycle focuses on validating hypotheses.
Experiment design involves specifying actions, expected outcomes, data collection, and drawing conclusions.
The concept of 24-hour experiments emphasizes the need for rapid learning.
All episodes this week complement the Coach Your PO e-course, available at bit.ly/coachyourpo. Related modules include:
Version 2, Module 8: Critical Documents for the PO Role – Utilize Impact Mapping to identify assumptions and expectations for designing experiments.
Version 2, Module 9: Collecting Feedback Faster – Explore methods for collecting feedback at various stages of product development, laying the groundwork for experiment-driven development.
In Version 1, Module 3, we delve into Actionable Metrics—an indispensable tool for anyone looking to run successful experiments.
For more details or to purchase the Coach Your PO e-course, visit bit.ly/coachyourpo.
If personalized coaching is your preference, contact us at coaching@oikosofy.com.
Continue your journey of learning, keep supporting your team, and join us for the final episode tomorrow!
See you tomorrow!
About Vasco Duarte
Vasco is a leading voice in the agile community, known for his contributions to the development of agile methodologies and practices. He is the co-founder of Agile Finland and the host of Scrum Master Toolbox Podcast, the most popular Agile podcast in the world, which has more than 10 000 000 unique downloads. He is also the author of “NoEstimates: A novel look at how Agile can transform software development, making it both more sustainable, as well as incredibly profitable.” Vasco is a keynote speaker at many conferences and events, sharing his knowledge and experience with the agile community. With his passion and expertise in agile, Vasco has made a significant impact on the way software development is done today, helping organizations to become more efficient, flexible, and responsive to changing requirements.lYou You can link with Vasco Duarte on LinkedIn and connect with Vasco Duarte on Twitter.
