Day 41–Agile series 3/7: “Lean”
Ever wonder where the concept of Minimum Viable Product (MVP) comes from? What’s the benefit of “running lean”? Is it suitable for non-startup environment? In this post, I’d discuss the key characteristics and application of Lean Startup & Lean Software Development (LSD) in the specifics:
- The popularity of “Lean” concept
- Cutting waste
- Before vs. After Lean
- How to adopt Lean
Lean is In
The concept of Lean in digital product development is one of the most prominent and influential movement in the 21st century. The popularity can be found from these trace:
- The concept of Minimum Viable Product (MVP) is adopted almost everywhere in the software industry
- The book that introduced MVP, “Lean Startup”, sold over one million copies and has bee translated into more than thirty languages.
- All other derived best sellers: Lean UX, Lean Enterprise, Lean Analytics, Running Lean, The Lean Product Playbook, and more.
So why is it so popular? What’s its secret? The key concept of Lean is to eliminate wastes.
Lean is No Wastes
The key concept of Lean is to eliminate waste. When the car manufacturer Toyota developed their production system TPS, their focus was to eliminate three main types of wastes in the car manufacturing process:
- Non value adding work (Japanese: Muda)
- Overburden (Japanese: Muri)
- Unevenness/ un-smooth flow (Japanese: Mura)
Toyota’s TPS system is highly effective in manufacturing world, and later Mary Poppendieck and Tom Poppendieck adopted the Lean concept & principles from TPS in their book “Lean Software Development” in 2003. The arise of Lean Software Development (LSD) consequently shaped the form of Agile development world, and also inspired the entrepreneur Eric Ries, who later wrote the influential book “Lean Startup”.
Applying to digital world
If we pick the main concept of “eliminating waste” into software development world, among all the “digital process waste”, I found the list below are the most important ones to eliminate:
- Extra features: Anything that’s not essential for the problems user are trying to solve. In other words, the notorious Scope Creep.
- Extra processes: Often time the blocker of product success. Whether it costs missing the time-to-market, or filters out innovative ideas, and quite often defeats the team morale.
- Waiting: The biggest enemy when competing in the fast pace, technology-driven digital world. When people can collaborate instantly, but they rather email back & forth, it becomes a waste of time, and also create potential confusion and misunderstanding.
- Defects: We should be mindful about quality. According to 1-10-100 rule, if the prevention cost is $1, then the correction cost is $10, and the failure cost is $100.
From Lean Startup to Lean Enterprise
Eric Ries’ book “Lean Startup” is a big hit, because in startup environment, funding is often limited, and they have to prove themselves to receive more funding from next round within a short time frame. The concept of Lean is a perfect match for that nature, and the “Build-Measure-Learn” loop is proven effective to validate if the proposed business model is repeatable & scalable.
In the enterprise world, during the time traditional enterprises were threaten by technology companies like Google and Facebook. They were all seeking for ways to transform in order to compete or survive, and the learning from dot-com bubble proved that large upfront investment doesn’t guarantee product success. The method from Lean Startup and the MVP strategy helped enterprise experiment faster with less investment and lower risk.
Before Lean vs. After Lean
Before
In software development field, the long going issues in the past few decades that block product success are:
- Lack of innovative solutions
- Long development cycle times
- Many redevelopment cycles
- High development costs
After
By adopting Lean, the cutting-wastes approach, both startup and enterprise benefit from the hyper-efficiency process, and the concept of MVP is popularized and the build-measure-learn cycle is fully embedded in many organizations.
Lean, but not Mean
There’s no perfect process in the world. Although Lean Startup & Lean Software Development (LSD) can be highly effective, there’s also associated price to pay. For example, the Build-Measure-Learn from Lean Startup skips the upfront analysis, which is great in a fast-pace, low-risk project. However, in the environment that is highly regulated, or associated with larger risk, in-which one mistake could cost billion-dollar lost, the upfront risk analysis is more important than speed of delivery.
In order to maximize the benefit of Lean Startup or LSD, without exposing unwanted risk, in my experience, there are some general guidelines to help us wisely adopt Lean:
- cut unnecessary features: anything considered as good-to-have, or can wait till phase-two product release; for example. notification, customization, advance setting. Try this: Thinking about the success of Twitter. The 1st version twitter is extremely bare-bone, however, that’s the beauty of it, “Keep It Simple Stupid”.
- cut unnecessary process: There’s always certain approval, verification can be bypassed if we wisely manage team & stakeholder expectations; the core team should have autonomy on low-risk decisions.
- cut unnecessary waiting: If team members can collaborate quickly, we shouldn’t wait. Try this: Co-location team boost productivity. Encourage team to discuss problem in person instead of emailing back and forth.
- cut unnecessary bureaucracy: common issues are: too many cooks in the kitchen, or design by committee. Try this: create a clear RACI matrix with the mindset of just-enough-syndication.
- cut unnecessary documentation: for example, progress report. Try this: replace documentation by human interaction, limit it to minimum documentation, and organize them on a central knowledge management system (e.g. JIRA, Pivotal)
Lean to Fit
In order to make Lean Startup or LSD fit well in our team, project, or organization, here’s a list of considerations that I find helpful:
- Solution verification: There are many decisions that internal debate or HiPPO (highest paid person’s opinion) aren’t as helpful as testing in the market. In the “there’s only one way to find out” scenario, best way to do it is to get-it-out-of-building rather than stay in the stagnant stage.
- Lack of upfront research: Another perspective is that when we don’t have proper resource or method to conduct required analysis, using LSD can help us quickly map out the desired value, outcome, and test in the market. By doing so, we can strategically identify the lack of resource or analysis, and restructure the project in the right way
- Limited time frame/resource/ support: It’s probably not the best place to be, but we have to implement mandate type of project which has less resource or support, overly-investing in it might not please anyone. Adopting Lean helps you quickly propose solution and gather feedback.
- Skunk work or proof-of-concept: When dealing with skunk work or proof of concept where your goal is to test it to see if work, adopting Lean can gain us velocity so that we can learn fast and learn often
- Analysis paralysis: If we have done enough upfront analysis and want to get out of extra processes or analysis paralysis, adopting Build-Measure-Learn loop can help verify all the key assumptions we have.
- Digital Transformation: Lean Startup and LSD are also powerful tool to create a pilot project, in order to identify the existing Wastes that can be eliminate. By doing so, the project can serve as a case study to see what processes are fundamental, and what are simply bureaucratic and should be eliminate.
Conclusion
Lean is a powerful way to get result fast. It’s great for fast-pace teams and organizations, or whoever is striving to increase velocity. The associated price-to-pay is the increase of risk exposure. The best way to adopt it is to start with pilot project to help identify the how it can fit in and revolutionize the way we work.
Do you also have something to share about Lean? I’d like to learn from you!
ABC. Always be clappin’.
