The Engineering Leadership Pipeline is Broken: Are We Promoting Too Late?

eng_director_luis · March 28, 2026, 11:13am

I’ve been thinking a lot about something that keeps me up at night: where are the next generation of engineering leaders going to come from?

The data is stark. We’re looking at an 18% increase in demand for engineering executives by 2026. At the same time, 45% of current senior engineering leaders in the US are eligible for retirement within the next five years. Over 25% of working engineers plan to retire within five years.

But here’s what really troubles me: only 28% of engineering executives in major US corporations were promoted through a dedicated, internal technical leadership track. The average time from senior engineer to VP of Engineering? 15+ years. That’s too slow for where we are today.

My journey: 18 years from Intel to here

I started as an embedded systems engineer at Intel right out of college. Moved to software architecture at Adobe after 7 years. Now I lead engineering teams of 40+ at a major financial services company. It took me 18 years to get to Director level.

Looking back, I can see years where I was ready for more responsibility but didn’t get the opportunity. I can also see gaps in my development—times when I was thrown into leadership roles without proper preparation. I learned through trial and error, which meant my teams paid the price for my learning curve.

The skills mismatch is real

Here’s another data point that concerns me: 62% of US companies report a critical gap in leadership talent possessing expertise in next-generation digital technologies. We have seasoned leaders with deep legacy expertise, but the market is demanding fluency in AI, IoT, Cybersecurity, Quantum Computing, and other emerging technologies.

This isn’t about older leaders being “out of touch.” It’s about the pace of technological change outstripping our leadership development processes. We’re asking leaders who cut their teeth on one generation of technology to lead teams building something fundamentally different.

Three possible problems

I keep coming back to three questions, and I honestly don’t know which is the biggest issue:

1. Are we promoting too late?
If it takes 15+ years to reach VP level, we’re losing talented people who won’t wait that long. Compensation for top-tier engineering executives is increasing 10-15% year-over-year. Companies that can promote faster or bring in external talent at higher levels are winning the war for leadership talent.

2. Are we training wrong?
I see an overemphasis on generalist management experience at the expense of deep technical expertise and proven innovation capability. We teach people how to run meetings and manage performance reviews, but do we teach them how to make strategic technical bets? How to balance innovation with reliability? How to build technical cultures that attract and retain top talent?

3. Are we looking in the wrong places?
The pervasive lack of structured development paths means we’re often promoting the best individual contributor rather than the person with the best leadership potential. And when we do hire externally, are we just recycling the same leaders from the same big tech companies, missing talented leaders from smaller organizations?

What I’m trying to do differently

At our financial services company, I’m trying to be more intentional about leadership development. We’ve started:

Identifying high-potential engineers earlier (at Senior level, not just Staff+)
Creating technical leadership tracks that don’t require giving up coding
Pairing emerging leaders with executive mentors
Giving leadership opportunities on smaller projects before throwing people into the deep end

But I’ll be honest—it’s hard. We’re competing with startups that can promote faster and big tech companies with bigger compensation budgets. And I don’t know if we’re moving fast enough.

What are you seeing?

I’m curious what others are experiencing:

If you’re an engineering leader: How long did it take you to reach your current role? What do you wish had been different in your development?
If you’re building leadership pipelines: What’s working? What’s not working?
If you’re an IC considering leadership: What’s holding you back? What would you need to see to make that transition?

Are we promoting too late, training wrong, or looking in the wrong places? Or is it all three?

Looking forward to hearing your perspectives.

vp_eng_keisha · March 28, 2026, 11:14am

Luis, this resonates deeply—but I want to push back on the framing a bit.

You asked if we’re “promoting too late.” I don’t think that’s quite it. I think we’re starting development too late.

My path: 16 years, but only the last 5 were intentional

I spent 16 years getting from Google IC to VP at my current EdTech startup. But looking back, only the last 5 years involved structured leadership development. The first 11 years? I was an amazing individual contributor who occasionally got thrown into management roles without preparation.

At Google, I became a manager because my team needed one, not because anyone assessed my leadership potential. At Slack, I learned director-level skills by trial and error—exactly what you described. The real turning point was when I got an executive coach and started being deliberate about my development.

The retention data is compelling

You mentioned the 15+ year timeline. Here’s what the research shows: internal recruits have 41% higher retention and reach full productivity faster than external hires. That’s a massive advantage.

But here’s the catch: we only get those benefits if we’re developing people intentionally from early in their careers. Not waiting until they’re Staff+ to think about leadership potential.

What we’re doing: building bench strength early

I’m scaling from 25 to 80+ engineers right now. I’m being very deliberate about identifying potential leaders at the Senior engineer level—not because they’re asking for management roles, but because I see specific traits:

They naturally mentor and elevate others
They think about systems and impact, not just their own tasks
They ask good questions about strategy and priorities
They’re comfortable with ambiguity and change

Once I identify them, we start development immediately:

Pairing with senior leaders for shadowing and mentorship
Leading initiatives across teams (not just their own work)
Getting exposure to business context (customer meetings, board updates)
Practicing the skills they’ll need 2-3 years before they need them

The question isn’t WHEN, it’s HOW

Your three questions are all valid. But I think they all point to the same root cause: we treat leadership development as something that happens TO people rather than something we build WITH them systematically.

We wait until someone is “ready” for promotion, then promote them and hope they figure it out. Instead, we should be asking: who has leadership potential, and what do they need to be ready in 3-5 years?

The skills mismatch you mentioned—62% reporting gaps in next-gen tech expertise—that’s a development failure, not a promotion timing failure. We should have been developing that expertise in our existing leaders years ago.

Real talk: It’s still really hard

Even with intentional development, I’m competing with:

Startups that can promote faster (smaller organizations, flatter structures)
Big tech companies with massive comp packages
The appeal of IC roles at companies that pay Senior+ engineers $400k+

80% of engineering executives now prefer hybrid or remote work. That’s changed the game. Geographic boundaries don’t constrain talent the way they used to, which means everyone is competing for the same leadership talent pool.

What would change if we started earlier?

Imagine if we:

Identified leadership potential at Mid-level (4-6 years experience)
Gave them 5-7 years of structured development
Had them ready for Director roles at 10-12 years total experience instead of 15+
Built diverse pipelines because we’re looking at potential, not just proven track records

That’s a much healthier pipeline. And it addresses your “looking in the wrong places” concern—we’d be developing leaders internally who might otherwise leave because they don’t see a path.

My question back to you

You’re doing something really important at your financial services company. But are you starting early enough? You mentioned identifying potential at Senior level. What about earlier? What if some of your mid-level engineers could be VPs in 6-8 years with the right development?

I’d love to hear what others think about this timing. When should we start intentional leadership development?

cto_michelle · March 28, 2026, 11:15am

Both Luis and Keisha are raising critical points. After 25 years in this industry, I’ve seen this cycle repeat, and I want to add a third dimension to this conversation: the strategic value of external hiring.

The 70/30 rule that actually works

I generally operate on a 70/30 split: 70% of leadership roles filled internally, 30% externally. This isn’t arbitrary—it’s based on what I’ve observed actually working at scale.

Internal promotion gives you:

41% higher retention (as Keisha noted)
Faster time to productivity
Cultural continuity
Clear career progression signals to the team

External hiring gives you:

Fresh perspectives that challenge assumptions
Patterns and practices from other contexts
Skills or domain expertise you don’t have internally
Catalysts for necessary change

The key is knowing when you need each.

When external hires make strategic sense

I’ve brought in external leaders specifically when:

We’re entering new territory: When we started our cloud migration initiative, I hired a VP who had done it twice before. The learning curve was too expensive to navigate ourselves.
We need to challenge the status quo: Sometimes your internal culture becomes too insular. An external leader who’s seen different approaches can be incredibly valuable.
We have a specific expertise gap: I needed someone with deep experience in technical M&A. That’s not something you develop organically unless you’re doing acquisitions regularly.

The culture fit trap

Here’s what concerns me about external hiring at the executive level: culture fit is real, and the cost of a bad fit is enormous.

I’ve seen brilliant external hires fail not because they weren’t technically capable, but because they couldn’t adapt to the organization’s decision-making style, communication patterns, or risk tolerance. At the executive level, that misalignment creates organizational chaos.

When I hire externally now, I spend as much time assessing cultural adaptability as I do technical capability. Can they flex their style? Do they listen before they act? Are they curious about why things work the way they do before trying to change everything?

The measurement problem Luis raised

Luis asked if we’re “looking in the wrong places.” I think the deeper question is: Are we measuring leadership potential correctly?

Most organizations promote based on:

Technical excellence as an IC
Tenure and loyalty
Success in increasingly larger scope roles

But what actually predicts executive-level success?

Strategic thinking and business acumen
Ability to lead through influence, not just authority
Emotional intelligence and stakeholder management
Comfort with ambiguity and incomplete information
Capacity to make and own high-stakes decisions

These are fundamentally different skill sets. I’ve seen amazing Staff+ engineers struggle as Directors because we promoted technical excellence when we needed strategic leadership.

What I wish I’d known 15 years ago

When I was at Microsoft, I thought technical depth was the most important thing. As I moved through architecture roles and into R&D leadership, I learned that the transition from technical to strategic thinking is the hardest part of the journey—and we give people almost no help making it.

Keisha’s point about starting development earlier is crucial. But we also need to be honest about what we’re developing people for. Are we developing them to be better engineers? Better managers? Or better strategic leaders?

Those require different development paths, and we often confuse them.

The real question: What does your organization actually need?

Before we decide internal vs external, early promotion vs patient development, we need to ask: What leadership capabilities does your organization actually need right now?

Scaling 50→120 engineers (which I’m doing right now) requires different leadership than:

Turning around a struggling team
Leading a major technical pivot
Navigating a regulatory compliance overhaul
Building a new technical capability from scratch

Sometimes you have those capabilities internally. Sometimes you don’t. The mistake is having a one-size-fits-all approach.

Back to Luis’s three questions

Are we promoting too late? Sometimes yes, sometimes no. It depends on the individual and what the role requires.

Are we training wrong? Absolutely yes. We’re training for management when we need to train for strategic leadership.

Are we looking in the wrong places? Yes—we’re often looking for proven track records when we should be looking for potential and adaptability. But also yes in another way: we’re not looking broadly enough at mid-size companies and non-traditional backgrounds.

What I’d add to Keisha’s framework

Keisha’s leadership traits are excellent:

Natural mentoring
Systems thinking
Strategic questions
Comfort with ambiguity

I’d add one more: curiosity about the business, not just the technology.

The best engineering leaders I’ve worked with (and hired) understand that technology is a means to business outcomes. They ask about customer needs, market dynamics, competitive positioning, and financial constraints. They make technical decisions with business context.

That’s the mindset shift from engineering leader to executive, and it’s what the 62% skills gap is really about—not just next-gen tech expertise, but business-technology integration.

What do others think? Is this mix of internal development and selective external hiring workable, or am I being too optimistic about finding the balance?

maya_builds · March 28, 2026, 11:15am

Coming at this from a totally different angle—I’m not an engineering exec, but I crashed and burned trying to lead a startup, and this conversation hits close to home.

The mistake I made (and see engineering orgs making)

When my startup was scaling, I promoted our best engineer to engineering lead. He was brilliant—could solve any technical problem, wrote beautiful code, mentored juniors naturally.

He was miserable as a lead. And honestly, not great at it.

The problem wasn’t that he wasn’t ready. The problem was that being great at building ≠ being great at leading building.

I see engineering organizations make this exact mistake constantly: promote the best IC to manager because they’re excellent at IC work. But leadership requires completely different skills.

Luis asked if we’re “training wrong”—I think this is it

You all are talking about when to identify leaders and whether to hire externally. But what about what we’re training for?

From what I’ve seen (admittedly as an outsider), engineering leadership training tends to be:

How to run 1:1s
How to give feedback
How to manage performance
How to prioritize projects

These are management basics. They’re important. But they’re not strategic leadership.

When I think about what killed my startup, it wasn’t that I didn’t know how to run 1:1s. It was that I didn’t know:

How to make strategic bets with incomplete information
How to communicate vision in a way that got people excited
How to recognize when I was solving the wrong problem
How to balance what customers said they wanted vs what they actually needed

Those are fundamentally different skills from “how to be a good manager.”

Are we even testing for the right things?

Michelle’s point about measuring leadership potential really resonates. She listed traits like:

Strategic thinking
Leading through influence
Emotional intelligence
Comfort with ambiguity
Making high-stakes decisions

But how many organizations actually assess these before promoting someone? In my experience (limited, I’ll admit), we test:

Are they good at their current job?
Have they been here long enough?
Do people like working with them?

Those predict “will they be an okay manager.” They don’t predict “will they be an effective strategic leader.”

The conversation we should have first

Before “when do we promote” or “do we hire externally,” maybe we need: What does this leadership role actually require?

Leading a 10-person team is different from leading a 40-person org is different from being a VP who leads directors. But we often treat “engineering leadership” as one monolithic thing.

My engineer-turned-lead would have been amazing leading a small team with clear scope. He struggled when the role required setting vision, negotiating with other departments, and making calls with incomplete information.

We put him in the wrong role and called it a development opportunity. That was unfair to him and expensive for the company.

What I learned the hard way

When I think about what made me a terrible CEO but a decent design lead:

I’m good at:

Seeing patterns and connections
Empathizing with users and team members
Mentoring and elevating others
Solving concrete design problems

I’m terrible at:

Making strategic decisions with high uncertainty
Selling a vision when I’m not sure it’s right
Staying focused when things are ambiguous
Managing my own anxiety enough to lead during crisis

I wish someone had helped me see that earlier. Not to tell me “don’t be a founder,” but to help me understand what skills I’d need to develop and whether I actually wanted to.

Back to your conversation

Luis, you asked three questions. I think “are we training wrong” is the most important one. But I’d reframe it: Are we even clear on what we’re training people FOR?

Because “engineering leader” at different levels requires different skills. And some people who’d be amazing Directors will never want to be VPs, and vice versa.

Keisha’s framework about identifying potential early is great—but only if we’re clear about what potential we’re looking for. Potential for what, specifically?

Michelle’s 70/30 rule makes sense—but it requires knowing when you need which capabilities. And that requires being honest about what the role demands and what your internal candidates actually want and can develop.

One more thing

Something that bothers me in tech: we act like technical leadership is the only path to impact and compensation.

Some of your best engineers might not want to be leaders. They might want to be Staff+ engineers who solve hard technical problems. And that should be okay—and comparably compensated.

Maybe part of the pipeline problem is that we’re pushing people into leadership because it’s the only path to growth, not because they actually want it or are suited for it.

Just a thought from someone who learned this lesson expensively.