Focus vs. Diversification

The hidden cost we don’t teach

For decades we’ve been taught a clean idea: diversify.

Diversification reduces risk. It smooths volatility. It prevents catastrophic dependence on a single bet. In finance, it’s foundational. In strategy, it’s often prudent. In careers, it can create resilience.

All true.

But there’s a downside to diversification that we talk about far less.

And it matters even more in the AI era.

Diversification doesn’t just reduce risk.

It also dilutes commitment.

It dilutes ownership. It dilutes accountability. It dilutes learning. It dilutes the painful, necessary work of finishing.

And sometimes the downside of diversification is not a small tax; it’s the reason nothing transformative happens.

Two examples keep bringing me back to this tension.

One is Steve Jobs’ discipline of doing fewer things; especially when Apple needed to stop bleeding.

The other is a case I wrote on Titan Materials, which managed to bring online what it describes as the first cement plant to operate in a fully closed-loop system run by AI; not by spreading AI across a dozen pilots, but by focusing it on one mission-critical project until it worked end-to-end.

This post is about the tradeoff we rarely teach:

Diversification protects you from being wrong. Focus gives you a chance to be great.

---

Why diversification is so seductive

Diversification feels like sophistication.

• It looks like optionality.

• It keeps everyone happy.

• It avoids uncomfortable choices.

• It gives leaders a story: “We’re exploring multiple pathways.”

• If something fails, you can say: “That’s why we have a portfolio.”

But there’s a hidden behavioral effect:

When everything is a priority, nothing is owned.

Diversification is often how organizations avoid commitment while still appearing ambitious.

It’s also how accountability gets blurred:

• “Multiple workstreams” means no single throat to choke.

• “Cross-functional” can become “no one is responsible.”

• “Pilots” become a way to postpone the hard work of adoption.

• “Innovation” becomes a shield against outcomes.

You end up with activity and meetings, not performance and change.

---

The cost of diversification: dilution

Here’s what diversification tends to do inside real systems:

1) It weakens accountability

When you have five AI initiatives, each gets a part-time owner. When each has a part-time owner, none has real authority. When no one has authority, the last mile doesn’t happen.

The organization stays safe… and mediocre.

2) It prevents compounding learning

Deep learning comes from repetition:
same team, same workflow, tighter feedback loops, continuous iteration.

Diversification spreads attention thin. Learning becomes shallow and fragmented.

3) It increases coordination costs

Every additional initiative adds meetings, governance, prioritization conflicts, integration dependencies, and executive attention.

You hit a throughput limit long before you hit an idea limit.

4) It creates “permission to drift”

With multiple bets, it’s easy for any single bet to drift:

• timelines slip,

• the best people get pulled away,

• priorities change,

• and eventually the work dies quietly.

A diversified portfolio can become a graveyard of almost-finished initiatives.

---

Titan: focus as a strategy for high-stakes AI

This is why Titan’s approach is so instructive.

In heavy industry, “try a bunch of AI pilots” is often the wrong move. The constraint isn’t curiosity. It’s integration.

To make AI real in operations, you need:

• reliable data pipelines and sensors

• robust models under real conditions

• process redesign

• operator training and trust

• safety and quality assurance

• clear KPIs tied to performance

That doesn’t happen with scattered attention.

The case shows the power of choosing one mission-critical objective and committing the organization to making it work end-to-end—model to workflow to behavior to performance.

That’s focus as an execution strategy.

Focus as a structural commitment that makes accountability possible.

---

Steve Jobs: focus is the discipline of subtraction

The Jobs story isn’t just “simplicity” as a design aesthetic.

It’s focus as strategic triage.

When Apple was in trouble, the company didn’t need more ideas. It needed fewer bets that could be executed with excellence.

And Jobs’ defining move wasn’t announcing focus. It was enforcing it:
saying no to good products, good features, good teams because doing fewer things well was the only path to survival and differentiation.

This is the uncomfortable truth:

Diversification is often socially easy. Focus is socially expensive.

Focus requires telling people:

• “Not now.”

• “Not this.”

• “We are not doing your idea.”

• “You may not get headcount for that.”

That’s why focus is rare.

---

The right way to think about the tradeoff

The goal isn’t “focus always” or “diversify always.”

The goal is to diversify risk without diversifying responsibility.

Here’s a helpful distinction:

Diversify at the level of options; focus at the level of execution.

You can explore multiple possibilities early. But once you decide what matters, execution requires concentration.

In practice, that means:

• Many hypotheses, few commitments

• Many ideas, few funded bets

• Many experiments, one end-to-end build

Especially for AI, where value comes from embedding the tool into a workflow—not from showcasing a demo.

---

The modern twist: AI makes the downside of dilution worse

AI creates a tempting illusion: because the tool is versatile, you can apply it everywhere.

But AI is not value until it changes a workflow. And workflow change requires commitment—training, redesign, measurement, trust.

That’s why so many organizations end up with:

• dozens of pilots,

• lots of chatter,

• and very little performance impact.

In the AI era, focus isn’t just a virtue. It’s a prerequisite for transformation.

---

The balanced conclusion

Diversification protects you from being wrong.

But the downside of diversification, when it becomes a lifestyle, a strategy, or a culture, is that it protects you from being accountable.

And without accountability, you don’t get the compounding learning that turns an initiative into an advantage.

Titan’s closed-loop achievement is a reminder that real transformation often looks like a single big bet, deeply executed.

Jobs’ Apple is a reminder that focus is usually less about doing more and more about choosing what you will not do.

So the question for your team, or your own life, is not: “Are we focused or diversified?”

It’s:

Where should we diversify to reduce existential risk and where must we focus to create excellence?

Because some things should be hedged.

And some things should be owned.