Here's the thing — most organizations hand AI to their team and say "be more efficient." And then they're confused six months later when the budget spent doesn't show up in the bottom line.

I'll be honest with you. That's not where the money is.

The real ROI from AI comes from a very specific group of people. Not your individual contributors. Not your single-function specialists. Cross-functional leaders. PMO directors. Operations folks. People who touch quality, P&L, project management, engineering, and customer service simultaneously. These people see strategic opportunities across the entire business that someone siloed in one function simply can't see.

Efficiency is the floor. Strategy is the ceiling. And almost nobody's aiming for the ceiling.

Why Are Most Companies Getting Poor ROI From AI Investments?

According to recent industry data, 72% of companies report no meaningful ROI from their AI implementations despite spending hundreds of thousands on tools and training. That's a massive failure rate — and it's not because the AI is bad. It's because the organization is approaching it wrong.

Here's what's happening: executives distribute AI tools like they're distributing productivity software. "Everyone gets access to ChatGPT. Everyone learns how to use it. We'll be more efficient." And people do get faster at their individual tasks. Marketing writes copy 30% quicker. Engineers generate boilerplate code faster. Support agents respond to tickets in half the time.

But that's table stakes now. That's the cost of admission. You're buying an app that everyone else also bought.

The companies that are actually making money from AI — and I mean real, measurable money — aren't doing that. They're identifying cross-functional leaders who operate at the intersection of multiple business functions. Then they're equipping those people to see patterns and opportunities that single-function teams miss completely.

A PMO director doesn't just see "I can use AI to write better status reports." They see "AI can identify bottlenecks that are slowing down engineering and affecting customer delivery timelines at the same time." Then they use that insight to change the entire project structure. That's strategy. That's where the money is.

What Makes Cross-Functional Leaders Better At Using AI?

It comes down to perspective. These people don't think in silos. They think in systems.

A silo-based AI user asks: "How do I use AI to do my job faster?"

A cross-functional leader asks: "Where are the handoffs between my teams breaking down? Where is quality suffering because communication is fragmented? Where are we duplicating work across functions?"

Then they use AI not to go faster — but to see and fix those structural problems.

Does that make sense? It's a completely different mindset.

I'm living it right now. When I'm working with cross-functional leaders in the AI Mastery Academy, I see them start asking different questions almost immediately. They stop thinking about individual productivity tools and start thinking about information flow. They stop thinking about cutting hours and start thinking about changing decisions.

One PMO director I worked with realized that engineering and customer service were running two separate AI workflows for customer feedback analysis. They were asking slightly different questions. They got slightly different results. Customer problems were getting solved 3-5 days slower than they should have because the two teams couldn't see the full picture.

She designed a unified AI workflow that both teams used. Same input. Same analysis framework. One decision point instead of two.

That wasn't more efficiency. That was a strategic restructuring that happened because one person could see across functional boundaries.

Here's what that looked like on the P&L: they recovered about 15% of customer service costs while simultaneously improving engineering prioritization. Not by cutting people. By changing how information moved through the organization.

That's the difference between efficiency and strategy.

The Efficiency vs. Strategy Framework

Let me walk you through how to think about this.

Efficiency ROI is linear. You implement AI, people work faster, you measure time saved. You spend $50K on tools and training, you save maybe $75K in labor in the first year. That's a win — but it's also your ceiling. Efficiency has a hard upper limit. You can't save more than 100% of someone's time.

Strategy ROI is multiplicative. You implement AI across functional boundaries, you change how decisions get made, you prevent problems instead of just solving them faster. You spend $50K on tools and training. You recover $200K in wasted motion across three departments. You speed up revenue-generating cycles by 20%. You prevent a $500K quality issue from getting to customers.

One is a productivity app. One is a business transformation.

And here's the hard truth — most organizations never make that jump because it requires something that's hard to build: trust and alignment across functions. Siloed teams are easier to manage. Siloed teams have clear budgets and clear responsibilities. Cross-functional initiatives are messier. They require people to see the bigger picture.

But that's where the money is.

How To Start Building Cross-Functional AI Strategy

If you're leading a cross-functional team and you're trying to figure out where to start with AI, here's what I want you to think about:

Don't start with the tools. Start with the map.

Draw out your critical handoffs. Where does information pass from one team to another? Marketing hands off to sales. Sales hands off to engineering. Engineering hands off to customer service. Customer service feeds back to product. Where are those handoffs slow? Where do you lose information? Where do teams make the same decisions twice?

That's your opportunity zone.

Then ask: "Where can AI see across that handoff in a way the current process can't?"

A support ticket doesn't just need an answer. It needs to be analyzed in the context of current engineering priorities. A customer request doesn't just need to be recorded. It needs to be weighted against what other customers are asking for. A feature request doesn't just need to go into the backlog. It needs to be evaluated against business impact metrics from sales and support simultaneously.

AI can do all of that. But only if you've built a cross-functional decision framework first. The tools come second.

That's why I built RACEprompt — to help people think strategically about their prompts, not just type faster. RACE is Role, Action, Context, Expectation. It's a framework that forces you to think about who's using the information, what decision they need to make, and what context they need to make it well.

You can build better prompts individually. But when you're working across functions, you need a shared framework. Otherwise everyone's asking different questions and getting incompatible answers.

The Framework In Action

Here's what a cross-functional AI implementation looks like:

First, you identify the core functions that touch your critical business outcome. Let's say it's customer retention. That touches support, product, sales, and engineering.

Second, you map what each function needs to see and what decisions they need to make. Support needs to know if a customer is at risk. Product needs to know what's driving that risk. Sales needs to know what intervention might work. Engineering needs to know if it's a product problem or a usage problem.

Third, you build one AI system that feeds all those perspectives into one decision point instead of four separate ones.

That system is using AI more efficiently than any individual function could — but it's doing it in service of strategy, not speed.

Done is better than perfect. Start with one cross-functional workflow. Prove the ROI. Build from there.

Why This Matters Right Now

Here's the reality: your competitors either understand this already or they're about to. The companies that are going to outpace everyone else in the next 12-18 months aren't the ones with the smartest AI. They're the ones with the smartest organizational design.

AI amplifies whatever structure you already have. If you have siloed functions, AI makes them more siloed because each function optimizes for itself. If you have cross-functional clarity, AI multiplies that clarity.

The budget you're spending on AI this year? That's an organizational design decision, not a tool decision.

Make it strategically.

Frequently Asked Questions

Why do most companies fail to get ROI from AI?

Most companies treat AI as a tool for individual productivity rather than as a strategic system for organizational decision-making. They focus on efficiency gains without addressing the cross-functional handoffs where real money lives.

What's the difference between an efficiency gain and strategic ROI from AI?

Efficiency gains are linear — you save time. Strategic ROI is multiplicative — you change how decisions get made across the organization, prevent problems before they happen, and improve outcomes you can't measure in hours saved.

How do I know if my organization is ready for cross-functional AI implementation?

Look at your critical business outcomes. Can you map which functions touch that outcome? Can those functions currently share information easily? If the answer to either is no, you need to fix that first. The AI is the second step, not the first.

What should a cross-functional leader focus on when implementing AI?

Start with the handoffs. Identify where information flows between functions and where that flow breaks down. That's where AI creates the most value. Then build a unified framework — like RACE — that all functions use to generate and interpret AI outputs.

How long does it take to see ROI from a cross-functional AI strategy?

It depends on your organizational structure, but most teams see measurable impact within 60-90 days of implementing their first cross-functional workflow. The key is starting small, proving ROI, and then scaling. Don't try to do everything at once.

Ready To Build Your Strategic AI Advantage?

Here's what I know: you already have the budget. You probably have the tools. What you might not have is the framework for using AI strategically — and that makes all the difference.

That's why I built RACEprompt. It's not just a prompt builder. It's a way to think about every AI interaction as a strategic decision instead of a tactical task. When your whole team is using the same framework — Role, Action, Context, Expectation — you stop getting disconnected outputs and start building organizational intelligence.

Check out RACEprompt at https://www.drjonesy.com/raceprompt. It's available on iOS and Android. Grab it, run it through your first cross-functional workflow, and see what changes when everyone's asking the right questions.

Your competitor might already be doing this. Don't get left behind.

- Jonesy