In pursuit of quality
Reflecting on how my journey through quality and bottlenecks has shaped my work experiences.
Quality, bottleneck and variation
This post is high on personal anecdotes. The lessons from the topic I mention will follow in a standalone post. Keeping with the tradition of promising posts to come but uncertain when ensues. Yet , I wanted to share a trend of unintended learnings and how they shaped my work unknowingly.
To begin, let’s go back to 2014. Yours truly was in his 3rd year of college and started visiting the Anna Centenary library to escape campus and find a peaceful environment. It was my on the side thing at that time.
I stumbled upon Robert M Pirzig’s Zen and the art of motorcycle maintenance. It took me a long time to finish the book as I had to go to library to read it each time I wanted to get out. I made some notes during that time.
The book is hard to summarise because it masks itself as a travelogue when it is really about philosophy. I was introduced to “quality” and what it means while reading this book.
My life would have been simpler if I didn’t try finding quality in the work I went about doing as a startup founder.
Let’s go back even further, circa 2008. I was suppose to study for my 10th board exams during the month long holidays. My father just got done with reading "The Goal: A Process of Ongoing Improvement" by Eliyahu M. Goldratt and Jeff Cox” and the book was sitting on the drawing room table. So, instead of studying geography or history or any other class subjects , I spent my time reading this book. Once again it didn’t have an immediate outcome but my career would have been lot easier if I didn’t focus on bottlenecks everywhere at work.
Last year, I read the book Understanding Variation by Donald J Wheeler on the recommendation of Cedric Chin of commoncog. I hope you get the drift by now, my work would have been more peaceful in the past year if I didn’t rework the performance measurement of the product I manage.
This sort of connecting the dots is only possible retrospectively. So, let us start with systems, we can break any work into a system of workflows interacting with each other. These could be simple, compound and complex in nature as I covered in understanding system at work.
When we are checking for the overall throughput of the workflow, it is determined by the throughput of the bottleneck. The lesson over here is to identify the bottle neck process and only predict the potential output basing on the throughput of that bottleneck process.
To make it more concrete, think of prevalence of AI in developing products. If you map the entire system, you will realise developers adopting AI in their code deployment is essential for reaching the intended outcome. Its not how many lines of codes, how fast the code generates or how easy is to generate code. It comes down to how much of GenAI generated code is being deployed into product. So, the bottleneck over here is code testing and deployment. Increasing the throughput of code deployment instead of code generation should the focus of the designated product person managing it.
Next, every process has some amount of variation in its output. Workflows are bound to experience variation while executing. We cannot have most productive day at work everyday. So, as a team member, your own output varies. Understanding the variation of a workflow can be achieved by measuring it on a time series. This is where Statistical process control comes into picture and work of Donald J Wheeler.
I have written a lot about growth in the past, let us suppose we want to achieve a specific growth rate. Just fixating on the number will not help us achieve the target. We need to understand the process variation currently deployed for growth to possibly predict what needs to be done to hit the target. The lesson from this is management is form of prediction. And in order to make good predictions you need knowledge. Statistical process control charts provide a visual representation of knowledge form the underlying information we have by studying the variation of workflow.
Lastly, a process with least amount of variation can be termed as one having higher standards of quality. This is what even Mr. Pirsig says in his book. Quality is hard to define but it is only probable if you care about the work you do.
Systemic outline of workflows, data driven mapping of variation in workflows and relentless but composed work in reducing the variation to achieve quality. These are inputs to achieve outcomes that matter at the end.
Spoiler alert of trying to achieve this in your work : The burden of proof will rest on you and you will have to work 10x harder to substantiate it every time. All the reasons why work is demanding for yours truly even after 10 years of working.
Round up
Every company is fundamentally a system of input and outputs. Designing metric trees is the first step in visualising the system.
The second step is defining the Value Architecture which illustrates the concept of mapping value creation function. We identify the bottleneck once this architecture is mapped and then deduce steps to consider to improve the throughput.
Links that resonated
Maybe You're Not Actually Trying to Solve Your Problems
This post by Navin Kabra is a review of the post(following recommendation) of Cate Hall on agency.
…Solomon’s Paradox, named after King Solomon who was famous for giving others wise advice but often struggled with applying the same advice to his own problems.
Maybe you’re not Actually Trying
I think we are all like this. People are not just high-agency or low-agency in a global sense, across their entire lives. Instead, people are selectively agentic.
Read both the posts to get a grip on what living in high agency means.
I write this newsletter mainly for myself because it helps me treat my work as a client requiring guidance. And all writing that I indulge over here is advice of some sort to yours truly.
Sign off
This last couple of weeks were challenging at work.
Like the topic of this post, work has variation in it. The output capacity of product development function at work feels sub-optimal and requires unclogging of the bottleneck. Past few week, it feels like my work is the bottleneck.
If you have read confidence in product building issue, you are aware of my wavering confidence in my capabilities. The week on week change for this week was negative.
Signing off till next time,
Vivek, mapping metrics even in my dreams