You Didn’t Stall. Your Org Chart Blocked You.
Let me describe your past year with AI rollouts, and you tell me how close I get.
You ran the pilots. More than one. They worked. You built the deck, you presented the results to leadership, and leadership said, “This is great.”
So now they’re back with the 2027 directive: increase AI usage across the organization, with a real budget attached this time.
This is where you start to panic.
Because you still have your AI Champions. You’re still running the office hours. You’re still answering the questions. People are using it. And usage is flat. The same people who were using it in March are the people using it now.
That’s where you are right now: sitting with a directive, a budget, a team of champions, and no idea where to go from here.
That’s not a failure.
A lot of people are quietly carrying this around like they blew it. You didn’t. Almost everyone who did the pilot work right is standing in the same spot, asking the same question. What’s next?
Here’s the thing. Nobody has a good answer because we’re not looking in the right place for one. We look at the tools. We look at the budget. We look at whatever model came out last week.
The starting place was always People. And “People” isn’t just who’s on your team. It’s how your people are arranged. Almost nobody thinks of that as an AI problem.
Your structure decides what you’re able to see
Every organization has a structure. Most of us inherited ours and stopped noticing it, the way you stop noticing the layout of your own house.
You only start noticing when problems show up. The stove is too small to cook for everybody at once. The laundry is in the basement when it really should be on the first floor. The bathroom is on a different level than the bedrooms. That one door swings out instead of in, so you can’t put anything behind it, and whoever designed it clearly never carried a laundry basket through it.
You don’t fix any of it because you don’t have the time or the budget to move a bathroom, so you get creative. You do laundry in shifts, you prep on the dining room table, and you get very good at working around a layout you’d never have chosen.
That’s organizational structure, especially the inflexible kind. It decides who talks to whom, who approves what, and (this is the part that matters) what you can see from where you sit.
Within every organization, there are four basic structures. Yours is one of them, or a mix of two.
Hierarchical. Chain of command, layers of management, information moves up and comes back down. This is the least flexible of the four, and it is what most enterprise-sized companies are working with. Not because they don’t know better. Because changing an organizational structure is a far bigger lift than learning to live with its limitations, so you learn to live with them.
The problem for AI is that everything moves vertically. Your best use case in customer service has no path to the person in finance doing the same task under a different name.
Functional. Grouped by specialty. Marketing here, finance there, HR, IT, ops. Deep expertise in every box. The problem is you scope AI by department, and AI opportunity isn’t department-shaped. It’s task-shaped. So five departments build five slightly different versions of “summarize this long document,” none of which knows the other four exist. Flat. Few layers, lots of autonomy, everyone starts their own work. Which sounds collaborative and isn’t. Everybody’s heads down on their own thing, so flat organizations get really good at optimizing what they already do and stay bad at inventing what they don’t.
In this type of organization, the pilots went well. It’s also why nothing came after them, because scaling requires somebody to say “we’re all doing it this way now,” and a flat structure has nobody whose job that is. Enthusiasm becomes your routing system; enthusiasm is not a routing system.
Matrix. Two reporting lines, usually function plus project or region. Matrix is the closest of the four to how AI work actually wants to move, because it already assumes people work across boxes.
But most companies shy away from it for reasons that have nothing to do with AI. Billing gets messy. Resource sharing gets contentious. When one person’s time is split between two bosses, somebody has to decide whose budget that hour comes out of, and that argument happens every week. (If you’re in a matrix and you just sighed, I see you.)
Notice what all four have in common. Every one of them organizes people by who they report to. Not one of them organizes people by what they actually do all day.
And what people do all day is the only thing AI can see.
We have data on that. We just published a research study (you can read the whole thing here) where we broke 434 occupations from the federal O*NET job database into individual tasks, then scored each one with our TRIPS Framework by Trust Insights: Time, Repetitiveness, Importance, Pain, and Sufficient data, the five things that tell you whether a task is a good candidate for AI.
AI-suited work showed up in every single category we looked at. Not clustered in the obvious ones. Spread across all of them. (Directional, not a leaderboard, and we sell AI enablement work, so read our numbers knowing that.)
Back to the house
Here’s why the workarounds stopped working.
A too-small stove isn’t a problem when you’re making dinner for your family. One burner at a time, one dish at a time, and it comes out fine.
It becomes a problem the day you host the holiday. Now the stove dictates the menu. Nothing about the house changed. What changed is the size of what you’re trying to do in it.
So it becomes a potluck. Everybody brings a dish they’re responsible for. That’s a smart answer to not having the space, and it’s also how you lose oversight of what ends up on the table. It might not be the meal you envisioned. It’s the meal somebody brought to the literal table.
That’s what your AI program is right now. Every champion brought a dish, all perfectly good and aligned with the goals, none of them planned together. That’s not a criticism. It’s the reasonable response to a kitchen that couldn’t handle more.
Eventually, you have to decide which problem you actually have. Either you work within the structure and live with the potluck, or the structure has to change.
And AI is no help with that call, because AI can’t see any of it. It doesn’t know your stove is small or that the laundry is in the basement. It hums along doing what it’s asked in whatever corner you put it. Every limitation in this house is obvious to you and invisible to it.
What to do Next
Look at your org chart and figure out which of the four you mostly are. Not perfectly. Most companies are one structure with a little of another mixed in, and “mostly” is good enough.
Then name the limitation that comes with it. Hierarchical, nothing travels sideways. Functional, you scoped AI by department when the opportunity is task-shaped. Flat, nobody has the authority to standardize. Matrix, the coordination cost is real and somebody has to be willing to pay it.
That’s your 2027 constraint. Not model choice, not budget, not whether people are bought in. Whatever your structure has always been bad at is exactly what scaling AI is about to ask you to do.
And whichever one it is, somebody has to own the meal. That role has to exist before the next pilot, not after it.
That’s the gap we built the AI Enablement Package for. Ninety days, fixed fee, mapping AI opportunity at the task level across your actual roles instead of by department, and handing you back a prioritized sequence instead of another list. It’s a diagnostic, not an implementation. If you want to see how it works, and view sample outputs, read more here.
The moral of the story
Your AI program didn’t stall because your people are resistant or because you picked the wrong tool. It stalled because it reached the edge of what your structure can see, and structures don’t reorganize themselves just because a pilot went well.
You can’t redraw the org chart, and I’m not suggesting you try. But you can give somebody the authority to move a proven workflow out of one box and into another. That’s what’s next.
What shape is your organization, and where does your best AI use case get stuck?
Reply to this email or join the conversation in our Free Slack community, Analytics for Marketers!
– Katie Robbert, CEO
|
Need help with your marketing AI and analytics? |
You might also enjoy: |
|
Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, INBOX INSIGHTS. Subscribe now for free; new issues every Wednesday! |
Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new episodes every Wednesday. |
Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.
Newsletter
Podcast
