"The best CTO is not the one who knows the most... it is the one who connects the best."
There is a recurring debate in boards of directors, startups, and companies undergoing digital transformation: after all, does the CTO (Chief Technology Officer) need to remain deeply technical, or should they act purely as a management executive? The answer to this question has changed decisively in recent years. And the shift is much more subtle than choosing just one side.
The classic dilemma: the two extremes that break operations
In market practice, whenever a company pushes its technology leader toward just one of these poles, it hires a predictable problem:
When the CTO is purely technical (and runs away from strategy):
- Dives into code as if they were still the lonely senior developer and forgets the direction of the business;
- Makes all architectural decisions alone in the middle of the night;
- Accustoms the team to merely executing orders, without thinking about the customer problem;
- As the company grows, they turn into the biggest approval bottleneck in the company.
When the CTO is purely a spreadsheet manager (and has lost their technical foundation):
- Does not truly understand what their own engineers are saying in architectural discussions;
- Makes million-dollar decisions based on PowerPoint presentations from trendy vendors;
- Quickly loses the intellectual respect and trust of internal technical leadership;
- Lets product architecture drift without clear direction, accumulating invisible security and cost liabilities.
Gregor Hohpe's Elevator and Camille Fournier's synthesis
Two fundamental works of modern engineering literature explain how to overcome this false dichotomy.
In the book The Manager's Path (O'Reilly, 2017), Camille Fournier (former CTO of Rent the Runway and engineering executive with a background at Goldman Sachs) analyzes the painful transition of top technical specialists into executive seats. Fournier demonstrates that the leader who tries to maintain technical control by reviewing every line of code stifles the autonomy of their Staff Engineers and stalls the organization; on the other hand, the executive who completely abdicates technical fluency loses the ability to judge real risks.
Meanwhile, Gregor Hohpe (former chief architect at Google Cloud and author of The Software Architect Elevator, O'Reilly, 2020) created the ultimate metaphor for the role: the Technical Leader's Elevator.
In any modern company, Hohpe says, there are two distant floors. On the penthouse, the board of directors and executives discuss EBITDA margin, customer acquisition cost, market expansion, and cash flow. In the basement, in the engine room, engineers discuss database latency, microservice coupling, CI/CD pipelines, and AI token consumption. Failure happens when the technology leader chooses to live on just one of these floors.
The synthesis that works for the CTO of the future is to operate the elevator every day:
- They need to be technically fluent, even without programming on the front lines every day. They need to understand architectural trade-offs, ask the difficult questions in five minutes, and have a sharp nose for spotting empty promises (technological bullshit) from vendors or consultancies.
- Above all, they need to be a bilingual translator: going up to the board to translate engineering decisions into financial impact, risk, and commercial velocity, and going down to the technical team to translate business strategy into clean architecture and clear priorities.
The 5 competencies of the CTO through 2030
Looking at the transformation accelerated by artificial intelligence and pressure for capital efficiency, five competencies separate the modern CTO from the traditional manager:
1. Deep product vision and unit economics
Technology does not exist in a vacuum to win engineering awards; it exists to make the product run with a healthy margin. The CTO who does not understand the customer funnel and the unit economics of the operation ends up building technological cathedrals that are perfect and financially unviable.
2. Talent management and multiplication
With AI agents taking over a good portion of repetitive code generation, the relative value of engineers with systemic vision and critical thinking has skyrocketed. Attracting, challenging, developing, and retaining this technical elite is the direct responsibility of executive leadership.
3. Evolutionary architecture
The role of the CTO is not to draw the definitive architecture for the next ten years on a whiteboard, because the market changes in six months. It is to guarantee an evolutionary architecture: modular systems, with low coupling and high observability, where changing course is cheap and premature optimizations are avoided.
4. Practical governance of AI and data
With autonomous systems and generative models entering the heart of operations, who defines what an agent can or cannot execute alone? What are the limits of security, privacy, auditing, and cloud cost overruns? This governance is no longer an academic topic and now lives on the CTO's desk.
5. Executive communication without technobabble
Knowing how to explain to the CEO and CFO why it is worth reserving 20% of the quarter to pay off a specific technical debt, why a critical migration requires more time, or why an investment in security prevents a million-dollar loss. Always in the language of the business, never hidden behind acronyms.
Technical governance in multi-squad operations
There is also a structural challenge as a company scales: what works for a single six-person team breaks quickly when engineering expands across dozens of engineers organized into multiple squads. Without explicit domain boundaries, versioned service contracts, and living operating agreements (such as a concise Engineering Handbook and lightweight Architecture Decision Records), local autonomy degenerates into fragmentation.
That scale is where executive engineering leadership delivers its highest leverage: connecting evolutionary architecture, per-application FinOps, critical database reliability, and autonomous Tech Lead development so squads ship fast without accumulating hidden liabilities on the balance sheet.
The final question
If you already occupy a technical leadership seat (or are building a career to get there), it is worth pausing and answering honestly: are you investing your energy in the connection, architecture, and business competencies that will be worth gold over the next five years, or are you doubling down on operational tasks that automation is already commoditizing?
The future of technical leadership belongs to those who know how to navigate between code and cash without losing truth on either floor.
Continue deepening: also read Light Environment, Strong Team and Developing Good People is an Investment. Also discover the work in Executive Engineering Leadership and the Case Studies.
References and verifiable readings
- Hohpe, G. (2020). The Software Architect Elevator: Redefining the Architect's Role in the Digital Enterprise. O'Reilly Media.
- Fournier, C. (2017). The Manager's Path: A Guide for Tech Leaders Navigating Growth and Change. O'Reilly Media.
- Ford, N., Parsons, R., & Kua, P. (2017). Building Evolutionary Architectures: Support Constant Change. O'Reilly Media.
Good ideas keep the conversation going.
Share this article or publish the visual deck to your Stories and Feed.

Cesar Zanis
15+ years leading software architecture, database performance, FinOps, and applied AI across enterprise operations. Built and scaled an engineering organization from zero to 20+ developers.
Enjoyed this article?
I publish practical analyses on critical systems architecture, Cloud FinOps, data, AI, and engineering leadership directly on LinkedIn.
Continue Reading
Take this analysis further with data and sources
All visual cards of this article are ready above to share on your Stories, Feed, WhatsApp, or LinkedIn with full sources.



