Your Team Has the Tools. But Nothing Has Changed.

There is no shortage of content about which AI tool to use. Every week brings a new model, a new framework, a new workflow that promises to change everything. I've written about the tools, about organizational culture change, about what becomes possible when old constraints dissolve. All of that matters.

But the more people I talk to, the clearer it becomes that the most valuable thing I provide isn't showing people which tool to use or how to think about change. It's getting them unstuck.

The people I talk to are smart. They're accomplished. They run companies and product organizations. And many of them are quietly paralyzed. They feel the pressure. They see the opportunity. But they can't translate it into forward motion for their teams. They don't know what the destination looks like, so every step feels like a guess.

I help them see the future. And then I help them get there.

You're Not the Only One Feeling This

I've been having the same conversation over and over, with different people, in different companies, using different words. Here's what it sounds like.

A CPO told me recently that his team has been thinking about how to integrate AI into their product development process for four months but hasn't had the opportunity to start pushing on it. He said "we haven't truly defined the to-state. We know we want to be AI driven, but we haven't defined that destination. So how do you identify which problems are actually worth solving with AI versus just an expensive distraction?"

He went further. "If we knew where we wanted to go, I would happily give up two sprints and tell everybody: go work on this. I would take the hit. But the problem is, how do I get there? How do I ensure success?"

A CEO I spoke with put it more bluntly. "We gotta figure this out. Or we won't win."

Another described having their entire team do a week-long AI offsite. They introduced tools, had everyone play with different platforms, tried building some things. Her assessment afterwards? "Yeah, this is cute. It doesn't really do anything." Back to business as usual.

And it's not just tech companies. I spoke with someone running a clinical practice where report writing eats 8-12 hours per case. His entire team has commercial AI licenses. Every single one of them is individually copy-pasting into a chatbot, each prompting differently, no standardization, no shared approach. He has a thousand reports that represent the institutional knowledge of what great work looks like at his practice. Nobody on his team realized that knowledge could be leveraged as anything more than a chat window.

Meanwhile, his people are describing themselves as "burnt to a crisp." One told him "I'm done." He asked if that meant they quit. They said no. But the point landed.

These aren't people who lack ambition or intelligence. They have both. What they don't have is a picture of what success with these tools looks like. And without that, every initiative feels like motion without progress.

The data confirms what I'm hearing. 74% of CEOs fear losing their jobs over AI within two years. 95% of AI pilots fail to deliver ROI. Over 80% of organizations report no productivity gains despite 70% of employees using AI tools. The tools exist. But still, the results aren't evenly distributed. And the gap isn't access. It's direction.

Very Few Problems Are Purely Technical

The hardest part of this transition isn't technical. It's human.

There are complicated feelings that go into adopting these tools. Your team is grappling with what their jobs are becoming. This was not the job they interviewed for. This was not the dynamic when they joined the company. The designer who spent years mastering a craft is watching AI generate layouts in seconds. The engineer who prided themselves on writing elegant code is being asked to manage code they didn't write. The PM who built their career on careful prioritization is being told the constraints that made that skill valuable have changed.

One leader I work with described their team as feeling "so much pressure and maybe a bit of overwhelm." They'd just finished a nine-month re-platform, and now leadership was saying "okay, ship weekly." It wasn't just a pace change. It was an identity change.

In this time of tumultuous change we get a clear view into whether we've actually been living the values everybody talks about. Does your organization provide the psychological safety and encourage the risk taking we need more than ever? Have you hired for resume checkboxes or for curiosity and resilience and a focus on outcomes? Does our team feel that they're not going to get fired for trying something new and having it not work?

That's not a given. For many people, the calculus is simple. I'm just trying to do what's expected of me and stay out of trouble. Experimentation is risk. Risk is dangerous.

And here's the thing leadership often gets wrong. They issue the AI mandate but penny-pinch on token budgets. They tell the team to experiment but don't give them time away from sprint work. Your team was already stretched thin, and now leadership has mandated AI usage without a roadmap. Figuring this thing out and applying it is a whole other job that they don't even know where to start with.

The math then becomes, which of these jobs will I get fired for not doing? The job they know and think they can do well in the old way is an easy choice.

They say "be bold" but the signals say "don't take risks." The message and the incentives are pointing in opposite directions. And incentives are undefeated.

Helping an organization through this isn't just about showing them the tools. It's about creating the conditions where people can actually use them. Permission and purpose. I wrote about this months ago, and it's become even more central to the work I do now.

AI. Easy to Do. Hard to Do Well.

Most organizations I talk to are stuck in what I'd call transactional use. Someone asks Claude a question. Someone uses Cursor on a ticket. Someone generates a document. Individual tasks, individual optimizations, one at a time.

It's a great start. But it's not a transformation.

The difference between "using AI" and getting real results comes down to two things. Context and tools.

Most teams have neither. They're copying something into a chat window and hoping the model's general training is enough. Sometimes it is. But you're getting generic output from generic input. A team of ten people, each typing into their own chatbot, each prompting differently, none building on each other's work.

You can automate some of that. Set up a recurring review, a regular synthesis. But if it has no context about your specific work, you're automating mediocrity.

The jump happens when you bring in context. Instead of a generic code review, the AI knows your architecture, your conventions, your standards. Instead of drafting a document from nothing, it draws on your full repository of past work, your customer interviews, your previous deliverables. It has a notion of what great looks like here, at your company. The output goes from plausible to informed.

And then there's the real shift. Context combined with tools. The AI doesn't just know about your work. It can act within it. Outputs from one process become inputs to another. You're not asking what content to write for SEO. You've built a system that continuously monitors where you should be investing and what you should publish. You're not running quarterly customer research. Insights from CS tickets, interviews, analytics, and user behavior flow continuously into your product decisions. You stop managing individual steps and start focusing on outcomes. The organization isn't using AI. It's running on it.

Most teams can't see past where they are because they've never experienced what's next. They need someone to show them.

I Get In the Work and Show You

Every organization I've described is starting from a different place. The CPO who's been thinking about this for four months has different needs than the clinical practice drowning in reports. If we're charting a course to the future, the path depends on where you're starting from.

So I figure out where you actually are. I talk to your people. I watch how they work. Usually it means transactional use, with a few adventurous individuals starting to bring in context on their own. That gap between perception and reality is enormous. 76% of executives think their employees are excited about AI. Only 31% are.

Then we address the thing I keep coming back to. The signals. If leadership isn't modeling the behavior, isn't creating the space for people to try and fail, the rest doesn't matter. We fix the conditions before we fix the tools.

Once that space exists, I embed with the team and work on real outcomes. Not a workshop. Not a deck. I'm in the work alongside your people. And my presence is itself a signal from leadership that this behavior is prized.

Because there is an enormous gap between knowing how to use these tools and knowing how to apply them. Showing someone how to use Claude Code is easy. Connecting your Slack conversations, your CS tickets, your analytics, and your customer interviews to generate a prioritized roadmap in an afternoon. That's a different thing entirely. It's the kind of thing you have to see happen with your own data, in your own context, before you believe it's real. And once you see it, it's a real unlock. But until you see it, it's hard to imagine.

One objection I hear is that embedding someone new is a big investment. Onboarding takes time. Is it worth it for a short engagement? But that assumption is part of the old model. Coming up to speed is orders of magnitude faster when you can browse the codebase, read every ticket, and synthesize the context without someone spoon-feeding it to you. Even watching how quickly that happens is a demonstration of what these tools make possible.

The team sees all of it. They participate. That's how the pattern transfers. That's how the capability becomes theirs, not mine.

And when I leave, the capability stays. Environments set up for AI to be effective. Workflows the team can repeat and extend. Confidence that this is achievable, because they just did it. And the to-state that CPO spent four months unable to define? It's no longer abstract. The team just lived it.

This Is the Work I Do

I've been writing about tools and tactics for months. Those posts aren't going away. But I wanted to be clear about what I actually do for organizations, because it's bigger than any individual tool or technique.

I help companies move from anxiety to action. From transactional to systemic. From "we should be doing something with AI" to actually operating differently. The path is technical and tactical, but it starts with vision and runs through the messy human stuff that makes organizational change so hard.

Naming it helps. If any of this sounds familiar, you're not behind. You're at the starting line. And having someone who has navigated this before makes the difference.

I'm looking for my next engagement. If your company feels like the ones I've described, I'd love to talk. Especially if you're:

• A CEO or CPO with a mandate but no method. You've told the team to use AI. Nothing has changed. You need someone who can show the way, not just explain it.
• A leader whose team did the offsite, bought the licenses, and went back to business as usual. The tools are there. The transformation isn't. You need someone in the work, not on a stage.
• Running an organization where the work is burning people out and you know there's a better way but can't picture what it looks like.

And if you know someone this might help, I'd appreciate the introduction.