Amazing CTO | More happiness and success
🚀 66.3

by Stephan Schmidt

Happy 🌞 Sunday,

if you struggle with vision and strategy (or the second order effects of unalignment, friction and rewrites), here is my simple ZigZag model of strategy for you.

My ZigZag Model of Vision and Strategy is simple. There are three groups of vision and strategy: Business, Product and Technology. Each group has its own vision and strategy. But they depend on each other. You start with a business vision and come up with a business strategy to reach that vision. The business strategy then enforces a product vision. Then a product strategy is defined to achieve the product vision. From the product strategy follows a tech vision and from the tech vision - you’ve guessed it - a tech strategy. Zigzagging between vision and strategy - each one is defined by another one.

Details with an example: https://www.amazingcto.com/zig-zag-model-vision-strategy/

This week’s insights

• 📯 Don’t use A/B Tests in “Optimizing for Taste”
• 🧙‍♂️ Excellent ideas in “Measuring Developer Productivity: Real-World Examples”
• 🐘 How Much Architecture Is “Enough?”

Good reading, have a nice Sunday ❤️ and a great week,

Stephan
CTO-Coach and CTO-veteran

CTO Job Market

CW 5 January 2026
	US	DE	UK
Open Positions	69 -22	35 +1	33 +13
Min Year	69	34	20
Max Year	91	35	33
Salary Avg	$174,561	€111.910	£118,146
P10	$97,000	€71.000	£79,000
P90	$260,000	€204.000	£175,000

If you only read one thing

Optimizing for Taste (13 minute read)

“If you’re not familiar with A/B testing, it’s primarily used in performance marketing as a way to optimize for human behavior. [..] Now you might ask, why would I be opposed to such a thing within our product? [..] It fosters a culture of decision making without having an opinion, without having to put a stake in the ground. It fosters a culture where making a quick buck trumps a great product experience.” Or as I would put it, people without a vision or strategy use A/B testing. It says, “I have no clue what I’m doing”. And A/B testing might be a good thing, when you’re into the cash cow phase of your product (Prototype,MVP,PMF,Scaling,Cash Cow), and while it might be a good idea for landing pages, it lets you run in circles instead of implementing your vision. The article has many more good points on A/B testing - or rather against it.

https://cra.mr/optimizing-for-taste

Tweet of the week

Meta releases new Code Llama 70B model (2 minute read)

“Today we’re releasing Code Llama 70B: a new, more performant version of our LLM for code generation” Just you know that is the current state of models. Though I do hear good things from https://github.com/deepseek-ai/DeepSeek-Coder - at least it’s fast and it seems to have less coding bugs compared to ChatGPT 4 from my limited tests. Will try Lama 70B.

https://twitter.com/AIatMeta/status/1752013879532782075

Picture of the week

Measuring Developer Productivity: Real-World Examples (20 minute read)

A very in-depth analysis of what companies use for developer productivity metrics. A lot of them measure developer satisfaction, but many good ideas for your company.

https://newsletter.pragmaticengineer.com/p/measuring-developer-productivity-bae

Stories I’ve enjoyed this week

No, office mandates don’t help companies make more money, study finds (8 minute read)

So return to office mandates do not increase revenue. And why would they? If your CEO needs arguments for work-from-home, the article has them. “Now, new research from the Katz Graduate School of Business at the University of Pittsburgh suggests that office mandates may not help companies’ financial performances, but they can make workers less satisfied with their jobs and work-life balance.”

https://www.washingtonpost.com/technology/2024/01/24/return-to-office-mandates-company-performance/

Announcing TypeScript 5.4 Beta (16 minute read)

Quite some of my clients dropped TS for JS again. TypeScript is racing fast to it’s demise. Why? TypeScripts adds more and more features, it makes it more and more complicated to learn TS. It makes it more and more complicated for juniors to enter. So while it does cater to it’s power users, it neglects newcomers. Many came to TypeScript because of “Javascript with types”, but over time TS became so much more complicated. That beta contains c: NoInfer<C>. How many TS developers will know what this does? I feel like it’s on the same curve as Scala, “Java but FP”. Scala got more and more complicated and then people dropped out, once the king of hip languages, today forgotten. Let’s contrast TypeScript with Go. Go adds basically no features each release. Most work is done on the toolchain. Go 1.22 adds range for integers. So for i := range 10 works, something people assumed would have worked already. Adding features to make the language simpler, not more complicated. Because Go has a mission of being simple. TypeScript has lost it’s mission of “Javascript with Types”.

https://devblogs.microsoft.com/typescript/announcing-typescript-5-4-beta/

👨‍✈️ New GitHub Copilot Research Finds ‘Downward Pressure on Code Quality’ (5 minute read)

“We find disconcerting trends for maintainability. Code churn – the percentage of lines that are reverted or updated less than two weeks after being authored – is projected to double in 2024 compared to its 2021, pre-AI baseline.” Who would have thought that people copy and paste stupid things from AI. AI makes good developers better and bad developers worse. I use ChatGPT a lot for coding, but there are many bugs in the more complicated snippets. To use AI for coding - for now - you need to know what you’re doing. And from my decades long interviewing, 30-50% of developers can’t code. I wonder how they use ChatGTP?

https://visualstudiomagazine.com/articles/2024/01/25/copilot-research.aspx

🏛️ How Much Architecture Is “Enough?”: Balancing the MVP and MVA Helps You Make Better Decisions (27 minute read)

As most startups are over-architecting their code (often in the wrong way because of lack of experience and extra enthusiasm), I like the concept of minimal-viable-architecture (MVA). If you give things a name, you can discuss it. I don’t agree with the author though, “The MVA balances the MVP by making sure that the MVP is technically viable, sustainable, and extensible over time” - the MVP should not be extensible, because you don’t know at that point in time what “extensible” means in your context. Instead, race to PMF as fast as you can, start scaling and think about what “extensible” means for you, at that point. Also https://www.radicalsimpli.city/

https://www.infoq.com/articles/mva-enough-architecture/

Hell is other people: performance management at Big Tech (12 minute read)

For me, performance management always tries to solve the wrong problem. The problem is bad managers, who don’t help their reports, don’t support them, don’t care, don’t talk about expectations, don’t hold people accountable, don’t give feedback and don’t fire reports and fight hard for their promotions. So instead of getting better managers or training them, they force performance management on everyone.

https://lcamtuf.substack.com/p/hell-is-other-people-performance

12 Modern CSS One-Line Upgrades (17 minute read)

I guess most people are stuck with their concept of CSS at that point when they’ve learned CSS. Which for me is >25y ago. It’s like talking to a medical doctor. From what they tell you, you can find out when they have finished university. I was surprised by some cool CSS changes. Just make sure, your developers are not like those doctors!

https://moderncss.dev/12-modern-css-one-line-upgrades/

The Missing Apex of Maslow’s Hierarchy Could Save Us All (14 minute read)

“Maslow lived until 1970 [..] In his later years, he added a new apex to the pyramid: self-transcendence.” - sense of meaning and purpose. For some short time, we thought Gen Z would be this people to realize that level. Then studies after studies showed, what they are most interested in, is money and career. Just like everyone else. Looking at my clients, those without a purpose struggle in the CTO role after some time. Being a CTO is demanding and stressful, pressure from all directions. Have a purpose. Or talk to your coach to find it 😉

https://bigthink.com/personal-growth/the-missing-apex-of-maslows-hierarchy-could-save-us-all/

In ❤️ Loving Memory of Square Checkbox (3 minute read)

Most designers want to express themselves. They don’t want to stick to rules. Which is one of the challenges between developers and designers. Every design is new and different each week, depending on what blogs the designer has read most recently. So no more square checkboxes this week.

https://tonsky.me/blog/checkbox/

The first rolling-ball mouse 🐁 e-basteln (12 minute read)

So there was a mouse (in Germany) before XEROX. And (on of) the first microprocessor was not the 4004 but the flight computer of an F-14. And the first programming language was Plankalkuel by Zuse. Most of what we think about history is wrong.

https://www.e-basteln.de/computing/rollkugel/rollkugel/

📜 I looked through attacks in my access logs. Here’s what I found (28 minute read)

Just a gentle reminder, that as soon as you turn a server life, and it’s connected to the internet, it will be attacked. Even if only for 10 seconds.

https://nishtahir.com/i-looked-through-attacks-in-my-access-logs-heres-what-i-found/

[Video] 🚂 KURUKURU KURURI SENRO (99 minute read - Just joking!)

Just watch this for 30 seconds, it’s genius engineering. I love how the track is laid out to minimize space and the train uses the same track in both directions (if you follow the steel, there are two distinct circuits, no electrical shorting ;-) This is what engineers in your company need to come up with. This gets them kudos from peers. This makes them proud. This makes customers WOW. This prevents them fiddling with the newest JS framework just because they are bored.

https://www.youtube.com/watch?v=mFOrZ0s0J6s

🏎️ Optimize sprint points to get nowhere fast (5 minute read)

“The ultimate metric that we care about is: how quickly do we get to the final destination of features that work for the users? [..] So why don’t we measure progress toward our destination? Well, because we don’t know where that is until we get there.” The reason we don’t know where we need to go is, we don’t have a strategy. With a good strategy we know where we want to go to, we only don’t know how it looks there. Which is a crucial difference.

https://ntietz.com/blog/optimize-sprint-points-to-go-slow/