“Apple suffered for several years from … lousy engineering management. I have to say it. And there were people that were going off in eighteen different directions, doing arguably interesting things in each one of them. Good engineers, lousy engineering management. The farm that’s been created with all these animals going in different directions and it dosen’t add up. The total is less than the sum of the parts.” — Steve Jobs, Apple WWDC 1997
Back in my Aussie days, my 1st graduate essay on management almost drove me craze 🤯 because I got back to school after making a living for 10 years. Even the working experience did help, but they also clouded my thought & effort on properly writing academically. Thanks to all my dear friends, especially Kay for kicking me out of procrastinating & Rich for fixing my broken English. Without your support, I won’t finish right before the deadline, literally.
As I survived, allow me to share the management 101: Plan, Organize, Lead & Control. Don’t worry you don’t need a MBA to see how important it’s for whatever business you really in. To keep it simple, let’s just use my inaccurate definition: Organizing is to form a group structure for people to create value. In this sense, an organizing effort is:
- Good: 1+1 > 2
- Bad/reality: 1+1 is about 2
- Ugly: 1+1 < 2
So what could be a better way to organize right? Let see what lesson learned may yield practical heuristics.
Focusing On The Road Ahead
“Almost any reorganization is designed to solve current problems people see. Over time, other problems come up.” — Michael A. Cusumano, a MIT professor
In 2013, ”Microsoft Overhauls, the Apple Way” was arguably a fair strategy to reform for one MS empire by unifying 8 feudal lands into 4 kingdoms. It did not turn MS into Apple or Google “killers”. A good strategy along changes nothing because “after culture eats strategy for breakfast 🤮, execution needs to clean up the mess”. Another more important question may be where you should really be. Because no one can turn to a new world by focusing on back mirrors only.
To be fair, it did put Satya Nadella in charger of the new MS cloud kingdom. Soon after in 2014, Satya Nadella was crowned as the MS CEO. Since that, he turns MS into a “cloud company” & makes MS super relevant again. For example, it’s the top challenger to the leader, Amazon Web Service. Even its market share, 20% is still behind AWS’ 32%, it grows faster, 50% vs AWS’ 28%. Very impressive, Nadella especially to walk-the-talk: “The question is whether you can amplify.” Even not yet an AWS “killer”, but Jeff Bezos is passing the torch of Amazon CEO to Andy Jassy, the CEO of AWS in Q3, 2021. What an interesting coincidence or just about time 😉.
Reverse Mirroring Hypothesis
“Any organization that designs a system (defined broadly) will produce a design whose structure is a copy of the organization’s communication structure.” — Melvin E. Conway
Back in my AJAR, a feature phone app framework days, we have often torn apart between two C++ standard libraries 🤹. Believe me, it’s not a happy case when 2 is not necessary better than 1 😭. There were many “good reasons”, but fundamentally there was a team UK & there was another team Denmark. Exactly as the prediction of Conway’s law, or the mirroring hypothesis more precisely. The MIT & Harvard working paper tested the “Mirroring” hypothesis & found that the structure of an organization does affect its software architecture. It has many examples of how open source products are significantly more modular than those commercial alternatives. Which is highly correlated to the ways developer organized between loosely-coupled communities vs tightly-coupled companies.
Ok you know it now, but how to use this knowledge to structure talents to maximize the collective outcome & individual potential? tl;dr
- Imagine the business you desire by Press Release & FAQ of the end system of a product or service should be in the view of the end-user.
- Reverse the mirroring hypothesis by applying the retrograde analysis. You can find out the best way to group talent & what should be the communication interfaces as the ideal system should be.
- Group talents accordingly & streamline communication interfaces for “Highly Aligned, Loosely Coupled” execution. This is the hard part because execution is the one actually clean up the mess 😉.
Remember the rate & scope of change in your context define how agile you need to be. And the level of uncertainty defines how quick you need to adapt.
Micro-Service-Oriented Architecture
“A small group of determined and like-minded people can change the course of history.” — Mahatma Gandhi
As the book, Creative Selection suggests many successful big Apple projects actually started from surprisedly small groups such as Safari Browser with 7–9 & iPhone with 25. Similarly Jeff Bezos’ 2 pizza rule & the SEAL team, they are just like micro-services: all small, autonomic, independent & decentralized as necessary with clear lines of communication & control to do one job well. So what could be practical heuristics to keep a team happy with 2 pizzas?
An answer may be in what Stevey shared of the first-hand observations on Amazon’s big reform to Service-Oriented Architecture from 2002. tl;dr All teams must expose their data & functionality by service interfaces, all communications must be done through service interfaces & nothing else. Which eventually turn Amazon from a “product company” to a platform company making its services “universally accessible”. Thank you, Stevey for a such great read & so many insights.
Building A Talent Cathedral
As technology advances, individual capability can be highly augmented & amplified. It’ll pay big for anyone to organize them properly to maximize the impact. Use these for your advantages on a journey of digital transformation no matter what business you are really in because software & service are talent-intensive businesses.
Go to build your talent cathedral. Sure, it takes time & dedication to build. Which’s hard for others to replicate. And as you build, it’ll be a great talent magnet 🧲. Remember if you want to play to stay in a long game, a few “SEAL teams” may be better than a big army because of the response time.
Full Disclosure
The opinions stated here are my own, not those of my company. They are mostly extrapolations from public information. I don’t have insider knowledge of those companies, nor a whatever expert.
