Be More Agile with LeSS, Part 5: Lean Thinking
It’s all about out-learning competitors.
In this multi-part blog series, I will share useful lessons learned from the Large-Scale Scrum book and the four-days Certified LeSS Practitioner training.
- Part 1: Why Agile?
- Part 2: What is LeSS?
- Part 3: Why LeSS? Why Not Others?
- Part 4: Agility Requires ‘Respect for People’
- Part 5: Lean Thinking (this post)
Some of you might’ve already noticed that the “Respect for People” principle in the previous post isn’t listed on the LeSS website! Well, I wasn’t lying. “Respect for People” is, indeed, an official sacred piece of LeSS.
All you need to do is a little zoom in on this LeSS’ website page explaining one of its principles: Lean Thinking. You’ll find “Respect for People” as one of two Lean Thinking pillars there.
This post will start with Lean Thinking’s goal. Then, we go to its brief history and real practices. With a more complete view of Lean Thinking, I hope we’ll see its strong influence on LeSS as a whole (principles, rules, guides, and experimentations).
The Goal of Lean Thinking
…is to go from “concept to cash” or “order to cash” as fast as possible at a sustainable pace;
to quickly deliver things of value (to the customer and society) in shorter and shorter cycle times of all processes, while still achieving highest quality and morale levels.
You can already sense the “respect for people” pillar from these keywords: sustainable pace, highest quality, and morale level. I assume you already understand why putting respect for people is a very pragmatic move profit-wise (read the previous post if you disagree).
What interests me is the order to cash and concept to cash. Those are new terms for me. And this is how I visualized them:
Let’s explore the more tangible one first.
Fast, sustainable “order to cash”
Lean thinking was born in the car manufacturing industry. In their context, the goal is “whenever there’s a car order, there should be a high-quality car delivered, as soon as possible, at as low a cost as possible.”
Let’s get back to the goal of lean car manufacturing: “The ordered cars are delivered as fast & low cost as possible”. Of course, the buyer (probably a car dealer) would be very happy about the fast arrival. That goal is a complete no-brainer that everyone will agree with.
Then, what makes lean manufacturing different, other than its less inventory approach? Compared to the other existing manufacturing approaches, lean thinking puts a strong emphasis on these practices:
- The collective intelligence of everyone (manager as teacher, cross-functional collaboration).
- Quality comes first at every step of manufacturing (long-term thinking, no compromise on engineering practices, heavy investment in people development).
- An obsession with continuous improvement (simple visual management tool, ‘go see’ (gemba), ‘stop the line’ (jidoka), and frequent reflection).
Read this short article if you want a clearer comparison of lean manufacturing with conventional one.
Do you notice some resemblances between lean manufacturing with agile software development? Interesting, isn’t it? Don’t forget that their context is factory/manufacture, a duplication activity.
But what do they duplicate in the first place?
Fast, sustainable “concept to cash”
This is how they find a car that many customers would happily buy, which is worth duplicating: research and design (RnD).
Car companies have RnD people to investigate and decide on the next car model/series.
Similar to what we do in DKatalis, they (the RnD people in a car company) ideate, build, and validate the hypothetical value to the user. In the eyes of lean thinking, all the activities of car RnD can also be as lean as possible, just like the manufacturing counterpart. I can imagine that it’s a valid enough statement.
However, there’s one missing puzzle…
Rethinking fast, sustainable “concept to cash” in a digital business context
Below is the itch I felt the first time I heard the “concept to cash” phrase:
A concept (or ‘idea of feature’) is a hypothesis. As with any hypothesis, it could (very possibly) be wrong. If it’s wrong, where would the cash come from? For a bad car model, despite the low sales number, the cash was still coming. You can’t guarantee that in a software, where user engagement to a new feature often couldn’t be mapped directly to a revenue increase. Must we push every single idea until it ends with an immediate revenue increase? If not, why don’t we call it “concept to (hopefully) cash” so it wouldn’t be misleading?
For our digital business, it seems that the “concept to cash” isn’t as straightforward as the “order to cash.” So, let’s adjust our visualization to cover the case of failing to influence the revenue positively.
Hmmm…
Wait? Really? Wouldn’t it be any case where we perceived the knowledge of concept-X-does-not-seems-useful-for-users as something valuable? And what if we gain a better understanding of users' problems and generate a new, better idea in the process?
Well, if that’s very possible, let’s add it.
It begs another question: “What if we need deliberate thinking on filtering and interpreting how the users react to what we’ve just built (the learnings) so that we can choose which learning we’d act upon first?”
If that case is valid, shouldn’t we draw a ‘machine’ to visualize that activity instead of just passing the learning(s) like on the current drawing? I think so. Let’s redraw.
Okay, it looks better now.
But somehow, it seems strange that we have two separate ‘machines’ there. I think they should be merged. Here are the reasons:
- Shouldn’t lean thinking promote collective intelligence and avoid silos?
- Putting the lean thinking aside, don’t you think both ‘machines’ depend equally on users and stakeholders? No one can effectively build anything with just a concept/idea, nor learn about users’ problems, with only small interactions with users
- Noting that both parties are interacting with users, wouldn’t there be a strong possibility of misalignment between them if it was done separately? For example, they might disagree on the perception of users’ problems.
And to visualize the merge, we need a redraw.
Also, when the digital product is already up and being used by a few users, the need for initial concept/idea generation to ‘kickstart the machine’ is completely gone.
Equipped with a good understanding and empathy towards users, good concepts/ideas could be cheaply generated. They even need to be put in a queue due to their large number. Again, that’s assuming the people in that ‘learning machine’ are quite already close with the users in the first place.
So, to symbolize their eagerness to be close to users, let’s add a pair of eyes and ears.
Now, to grow a loyal userbase and generate as much revenue as possible, the CEO (who is the de facto owner of the ‘learning machine’) should do two things:
1. Let the ‘learning machine’ & the users be as close as possible.
Taking Lean Thinking as its basic principle, this closeness to users is also what LeSS is championing.
And even more profound, one of the four values of the Agile Manifesto itself.
2. Keep the ‘learning machine’ as lean as possible.
Structure-wise, which structure would have a shorter yet more fruitful communication and collaboration (a.k.a leaner), the pink or the green one?
Visually, the green one. Right?
But let me tell you a secret: some people will answer “the pink one.” Especially when there aren’t many people in the organization. I get the reason behind that answer: because it’s far easier!
- There is no need for extra effort to understand teammates’ work.
- The micro-management needed to orchestrate the whole thing in the trenches is still bearable — and not yet painful to the executors.
The silo structure still produces relatively good results, as everyone still personally knows everyone. The small number of people makes it feel lean.
But usually, the organization grows as the product and the funding grow. That ‘pink’ way of collaboration comes with many issues: miscommunications, misalignments, long waits, fights over a resource, behavioral complaints toward each other, and other problems. Fear slowly becomes the tool of control.
And it’s quite expected. Imagine the increase in PO’s stress level in the situation below.
The more a ‘learning machine’ busy with itself,
the less it understands users,
the less its survivability and the revenue increase will be.
A Short Story from DKatalis
What is the color of the DKatalis product development structure now? Pink? Nope.
Green? Not either (or not yet!)
Adopting LeSS (including its Lean Thinking) after your organization scale is difficult. Please note that last year (2022), due to the large size of our organization, we went directly to the LeSS Huge version — where there are requirement areas.
For sure, we’ve experienced the benefits of being leaner. For example, before the Flip (that‘s what we call the LeSS training and adoption kickoff), there were two teams in the Money Storage area. Each handling the account features on ‘two different software’: the web and the mobile app. Post-flip, those two teams turned into four smaller teams, and each is expected to bend both the web and the mobile app. Being a more cross-component team is not easy; it requires a lot of learning in the beginning. Yet, after a couple of months, the efforts paid off. They felt more able to produce more high-quality features compared to pre-Flip.
The goal of Lean Thinking… …is to quickly deliver things of value (to the customer and society) in shorter and shorter cycle times of all processes, while still achieving highest quality and morale levels.
The Money Storage story above is only a part of a leaner structure. Even structure is not the only domain to evaluate; processes and artefacts could also be made leaner and leaner. Increasing a team’s cross-functionalities and ability to bend more product components also takes a lot of time and effort, not to mention the impediments that are not really in the organization’s control (both technical and non-technical), rendering a perfect cross-component team impossible for now.
Just like with our body, to be a leaner organization (and to keep being lean) is a hard (and never-ending) endeavor.
However, unlike with our body, in the business of innovation race, not being lean enough might resulted to death (bankruptcy).
As users’ cost to switch between digital products is low to nothing and most feature ideas are free to copy, the digital business battle is all about out-learning competitors. The best learner wins and takes all.
Conclusion
This post aims to provide a better understanding of Lean Thinking so that we can see its influence on LeSS. In my opinion, Lean Thinking — with its respect for people pillar included — occupies a significant portion of LeSS’ core; and heavily influences many of LeSS’ parts. Here’s the mapping.
And also, can you notice the most obvious sign?
to be ‘lean’ is to have ‘less’ waste…
Find more way of working tips written by our Katalis here.
