Making AI Adoption Work: A Practical Series - part 1

The 5 Critical Foundations (That Most Organizations Skip)

Welcome to FullStack HR, and an extra welcome to the 12 people who have signed up since last week.

If you haven’t yet subscribed, join the 10000+ smart, curious HR folks by subscribing here:

---

Happy Tuesday!

Today you are reading the first part in a 4-part series about AI adoption.
My aim here is to provide you with a toolkit to make AI adoption work in organisations. I know a lot of you struggle with this, hence why I’ve put this together.
It’s based on what I do myself as a consultant and also what I’ve seen actually work in organisations.

If you have questions, feedback or comments - let me know!

But let’s get to it.

---

How do you actually drive AI adoption in organizations? Not the theory. The practice.

This is Part 1 of the AI Adoption Series - for HR people who either have a mandate to make this happen, or who see the need but don’t have leadership buy-in yet.

Both situations need the same foundations. Let’s get into it.

The 5 Critical Foundations for AI Adoption (That Most Organizations Miss)

Here’s what challenges most people’s beliefs: Having leadership say “yes, let’s do AI” doesn’t mean you’ll succeed. I’ve been called in as rescue patrol too many times where leadership wanted it, budgets were allocated, but nothing moved.

This costs organizations thousands of wasted hours, burned budgets on unused tools, and competitive advantage lost while others move forward. And it happens whether you fought for buy-in or it was handed to you.

The difference between organizations where AI adoption works and where it dies? Five specific foundations that must be in place.

Why Good Intentions and Budget Aren’t Enough

Most organizations think: Get leadership to say yes, allocate budget, buy tools, done.

Doesn’t work like that.

The fundamental flaw is thinking adoption happens because you decided it should. Research shows AI will impact work, the jury is still out on exact scope, but there’s zero doubt it’s happening. But decision to adopt and actual adoption are completely different things.

I’ve seen it both ways. Organizations where the CEO pushed hard but nothing stuck. Organizations where HR had to fight for it but then made it happen. The pattern? Same foundations missing or present in both.

Foundation 1: Real Leadership Mandate (Not Just Permission)

This is the crucial foundation. Without it, everything else fails. But here’s where it gets interesting - you might think you have it when you don’t.

If you don’t have leadership buy-in yet:

You need it. Period. Not negotiable. (I thought differently about this a couple of months back but I’ve completely changed on this - you need that buy-in.)
Here’s how you get it, ranked from best to last resort:

• Spark genuine curiosity: Show them tools working practically. Let them feel it. ChatGPT demo, whatever’s relevant. Get them to that “oh shit” moment. This is the best path when you can take it.

• Use the data angle: Show research on actual impact. Show what your closest competitors are investing. “They’ve put X into this, seeing Y results.” Sometimes leadership needs to see they’re falling behind.

• Play the compliance card: AI literacy is legally required as of February through EU AI Act. Frame it as risk management and compliance. Not my favorite path, but it works when others don’t.

Meet them where they are, not where you wish they were.

If you already have leadership saying “yes”:

Good. Now verify it’s real. Real mandate looks like this:

• They understand AI’s impact on YOUR specific industry, not generic AI hype

• There’s actual budget and resources, not just “figure it out”

• It appears in OKRs, leadership meetings, actual priorities

• Decisions get made with new premises, not old playbooks

Warning sign you DON’T actually have it: Six months in and nothing’s moved. Tools sitting unused. “We’ll get to it next quarter.” That’s symbolic buy-in, not operational buy-in.

Real example: CEO said “we’re doing AI” but kept prioritizing everything the old way. We got it into actual OKRs and every leadership meeting agenda. Things moved within weeks. That’s the difference.

Foundation 2: Both Types of Knowledge (Technical AND Organizational)

You need two parallel competencies. Not one or the other. Both.

Industry-specific AI understanding: How does AI specifically reshape YOUR industry? Not AI in general. YOUR industry.

If you’re building from scratch: Bring in someone who knows both AI and your industry. Or consume industry podcasts, track industry moves, stay close to the conversation.

If leadership handed you this: Verify they’ve actually consumed industry-specific knowledge, not just generic AI content. Most leadership has read a lot about AI. Almost none of it helps them lead differently in their specific context.

Organizational context translation: Here’s where it gets rare. Plenty of people explain how AI works technically. Very few can explain how you lead into this. How you change culture. How you shift processes. How you set it in organizational context.

This is why I’ve been rescue patrol so many times after strategy consultants. They come in, talk high-level, maybe provocate, then leave zero practical tools. Leadership is inspired but paralyzed because they can’t connect it to tomorrow’s decisions.

Three questions to answer:

1. Can we connect AI possibilities to our actual work processes?

2. Do we understand organizational change, not just technology?

3. Are we thinking practically or just strategically?

If you can’t answer yes to all three, you’re missing something critical.

Foundation 3: Cross-Functional Structure with Clear Ownership

Once you have mandate and knowledge, you need structure. Fast.

If you’re building from zero:

Create a cross-functional team immediately. Not a committee that talks. A team that drives.

Who needs to be in: HR (you), IT, business representative, maybe finance. 3-7 people max. Meet monthly minimum with concrete progress checks.

If structure already exists:

Verify it’s actually functional. Does it meet regularly? Are there concrete deliverables? Does anyone know who’s actually responsible?

Here’s where I get controversial: Someone must own this. And it should be HR, not IT.

Why? This is a culture and organizational question far more than a tech question. I’ve come into organizations where IT drove for years without effect. When HR took ownership, things moved.

Important distinctions:

• Ownership doesn’t mean alone

• You don’t need to be an IT project manager

• You don’t need to know all the systems

• But someone must be accountable for forward movement

If nothing’s happening, someone needs to answer for it. That creates urgency.

First week metric: Does everyone in the organization know who’s driving this?

Foundation 4: Practical Learning, Not Just Theoretical

Leadership needs to learn by doing, not just by consuming content.

For different leadership types:

• Hands-on learners: Get them using tools in their actual work. Not demos. Real work. This week.

• Analytical types: Give them frameworks to evaluate use cases. Show them how to spot opportunities in their domain. Make it diagnostic, not prescriptive.

• Cautious leaders: Start with low-risk pilots that prove value fast. Build confidence through small wins that compound.

The common thread: They need to feel competent, not overwhelmed. If they’re paralyzed by not understanding, they’ll deprioritize it.

Unexpected benefit at this stage: When leadership learns practically, they start seeing opportunities you didn’t. They become advocates, not obstacles. That’s when adoption accelerates.

Foundation 5: Intentional Approach to Process Change

Here’s the thing most organizations miss: AI adoption means process change. And process change means people change.

You need to decide upfront: Are we replacing tasks or replacing people? Because if you’re not intentional about this, you’ll end up with neither adoption nor trust.

For situations where you’re starting fresh: Define this early. What’s the goal? Efficiency? New capabilities? Cost reduction? Different goals require different approaches.

For situations where you inherited this: Figure out what leadership actually wants, even if they haven’t articulated it. Their real goal versus stated goal. Because if those don’t align, you’ll hit resistance you can’t see coming.

Precise ways to measure impact:

• Track time saved on specific tasks (concrete, not estimated)

• Measure output quality changes (before/after comparisons)

• Monitor adoption rates across teams (who’s using it, who’s not, why)

Challenge for your first month: Pick one measurable metric that matters to your business. Track it religiously. Report it consistently.

When You Hit the Wall (And You Will)

Most common roadblock: Everything looks good on paper but nothing actually moves.

This means one of your five foundations isn’t actually solid. Usually it’s Foundation 1 - mandate is symbolic, not real.

Why it happens: Leadership said yes but didn’t change how they decide, prioritize, or measure. The “yes” was aspirational, not operational.

Recovery steps:

1. Go back to Foundation 1 - is mandate real or symbolic?

2. If symbolic, escalate using tactics from Foundation 1

3. If real, check Foundation 3 - is ownership clear?

4. If ownership is clear, check Foundation 4 - is leadership learning practically?

Reframe the resistance: If you’re hitting real walls, it means you’re asking for real change. That’s good. Means you’re not doing cosmetic bullshit.

Your Path Forward from Here

Here’s what to do next based on where you are:

If you don’t have leadership buy-in yet:

1. Assess where your leadership stands today (honestly!).

2. Choose tactic from Foundation 1 based on their drivers

3. Book a meeting this week with CEO or closest leadership person

4. Agenda: “AI’s impact on our organization - where do we stand?”

5. Bring concrete examples from your industry

If you have leadership saying “yes”:

1. Verify it’s real buy-in using Foundation 1 checklist

2. Audit Foundation 2 - do you have both types of knowledge?

3. Build Foundation 3 structure this week if it doesn’t exist

4. Get Foundation 4 moving - leadership learning practically

5. Define Foundation 5 - what’s the actual goal here?

Both paths converge here: Without these five foundations solid, adoption won’t stick. With them, it compounds faster than you expect.

Single most important action regardless of scenario: Get crystal clear on whether you have real mandate or symbolic mandate. Everything else depends on getting this right.

Don’t wait for the “right time.” That doesn’t exist. Start this week. Start now.


PS. Next week in part 2 we dive into the training approach that makes adoption stick - because you can have all the buy-in in the world and still fail if people don’t know what the hell they’re doing.