Get Leaders Aligned on AI
Getting the Leadership Team Aligned on AI Strategy:
The Real Conversation Mid-Market Leaders Need to Have
Your CFO thinks AI is about cost reduction. Your VP Operations sees it as automation. Your CIO is focused on data infrastructure. Your CEO wants competitive advantage - yesterday.
This is the room I walk into almost every month. And it's the room where most AI initiatives either die or spin their wheels for years.
The problem isn't that you don't have smart people in the room. It's that you don't have a shared language for what you're trying to do with AI, why you're doing it, and how you'll know it worked.
Getting your leadership team aligned isn't about consensus on AI philosophy. It's about getting clear, specific agreement on the business problem you're solving first - and then collectively sizing up whether AI is part of the answer.
Here's what I've learned works. And what doesn't.
What's Really Going On: The Misalignment Trap
Most mid-market leadership teams approach AI backwards. They start with the technology ("Should we use machine learning?" "Do we need data scientists?") when they should start with the problem ("What metric is keeping us up at night, and why isn't it moving?").
That backward framing creates three predictable breakdowns:
Clear Owner. Everybody's for AI; nobody's responsible for it. Finance thinks IT will drive it. IT thinks Operations will prioritize it. Operations thinks the CEO will fund it. Nobody moves.
Conflicting Success Metrics. The CFO measures success by ROI and cost savings. Operations measures it by time saved or throughput gains. Customer Success measures it by satisfaction scores. A single AI project can "fail" in one lens while "succeeding" in another - and the team never resolves which lens matters first.
No Shared System View. When a metric stalls - say, customer churn keeps rising despite higher NPS scores - it's tempting to bolt on a solution: "We need better customer segmentation AI." But if you haven't looked at the whole system - why customers actually leave, where handoffs break down, which processes create friction - you're treating a symptom, not the root.
I worked with a dental insurance company that wanted to implement AI x-ray technology to accelerate claim processing. But when we zoomed out and mapped the entire system, we discovered opportunities to streamline interfaces with the practices along with smaller scale AI automations that would provide immediate cost savings while they worked on the more complex X-ray technology solution.
That discovery only happened because the leadership team got aligned on the problem before jumping to the solution.
A Practical Method: The Three Questions
Here's what works. In your next leadership meeting, answer these three questions - in order. Don't skip ahead. Don't let the room jump to solutions.
Question 1: What's Our Stuck Metric?
Pick one business metric that leadership genuinely cares about and that hasn't moved (or moved enough) in the last 6-12 months. This should be something tied to revenue, cost, risk, or customer/employee experience.
Examples:
Customer acquisition cost (CAC) is rising despite higher marketing spend.
Average time to resolution for support tickets is stuck at 48 hours.
Manufacturing line downtime is costing us $2M annually.
Sales cycle is 9 months; industry benchmark is 6.
Don't pick "We need to be AI-forward." That's not a business metric. Pick something that hurts if it doesn't move.
Question 2: What System Influences This Metric?
Now zoom out. Where does this metric live? What functions, data sources, decisions, and handoffs feed into it?
Use this simple framework:
Upstream: What decisions or processes create the conditions that affect this metric?
Core: Where is the actual work happening that directly impacts the metric?
Downstream: What happens after? Who depends on the output? Where does friction appear?
Example: If your stuck metric is "Average time to resolution for support tickets":
Upstream: How are tickets routed? What knowledge does the agent have? Is there a triage delay?
Core: How long does the agent spend researching, solving, and documenting?
Downstream: How often does the customer come back? What handoffs happen if escalation is needed?
Get the room to draw this. It's messy. That's good. Messiness reveals where the real levers are.
Question 3: Where Are the Friction Points That AI Might Actually Address?
Once you have the system mapped, look for three types of friction:
Handoffs: Where does work pass from one person or system to another? Where do things slow down or get lost?
Multi-Step Tasks: What repetitive processes are eating time that people shouldn't need to be doing?
Decision Bottlenecks: Where is decision-making slow because data is scattered, incomplete, or hard to access?
AI is good at certain kinds of friction. It can automate multi-step tasks (like document processing or data extraction). It can surface buried data to speed decisions. It can handle routine handoffs. It's not good at fixing broken human workflows or solving root-cause problems that need process redesign first.
An example: Financial Communications Creation Flow
A global financial services team I worked with was frustrated. Average document cycle time had crept over 4 months and stakeholders were complaining that materials were always lagging. Leadership suspected “AI accelerate the review process” but nobody had fully mapped and analyzed where the time was actually being lost.
The three questions
1. Stuck metric: Average document cycle time (target: 2 months; actual: 4+ months and rising).
2. System map: Projects were often launched 2 months before the communications team was even notified, quality-review handoffs stalled due to bottlenecks and difficulty spotting changes, and the design team received incomplete specs that required multiple follow-ups with content creators.
3. Friction points:
Resource allocation constantly lagged due to no up-front visibility into new projects.
Review cycles took weeks instead of days, every step considered ‘essential’ for quality.
Designers waited on missing details, dragging out the final step.
Where leadership aligned
Once everyone saw the system map, the leadership team aligned that “we need workflow visibility and faster handoffs,” not “we need AI to automate parts of the process.” The real leverage was fixing how work entered the system, moved across teams, and surfaced changes, with AI as a future focused transformation, not the “ROI now” centerpiece.
Quick wins first
Projects added at conception: New initiatives were entered into the project system as soon as they were conceived, giving the communication teams early visibility and allowing better resource planning.
SharePoint automations: Simpler, automated change-highlights and queue-prediction models focused reviewers on what mattered and cut handoff wait times.
Stronger design support and specs: Additional design support to create clearer spec templates reduced back-and-forth and trimmed days off the final stage.
Then AI becomes strategic
Only after these process and visibility fixes did the team explore an internal LLM, trained on millions of existing documents to consistently express the company’s brand voice. At that point, we had accelerated a healthy workflow instead of trying to rescue a broken one. By starting with the system, the team didn't just pick a "more interesting" AI project. They picked automation solutions that would actually move the metric.
By starting with the system, the team didn't just pick a "more interesting" AI project. They picked one that would actually move the metric.
Tool: The Leadership Alignment Worksheet
Use this worksheet in your next meeting. Assign someone to facilitate.
Key Takeaways
·Start with stuck metrics, not technology. "We need AI" is not a strategy; "We need to reduce churn" is.
·Map the system before designing the solution. Zoom out first. Most problems live at the intersection of process, data, and people.
·Identify friction types that AI can actually address. Not all problems are AI problems. Many are process-redesign problems. Know the difference.
·Get clear on who owns alignment. Don't leave it to "the team will figure it out." Name a person or small group responsible for maintaining that clarity.
A Question to Ask Your Team This Week
"What business metric has our leadership team been frustrated about for the last six months, and what system influences it?" If you can't answer that in 30 seconds of clear conversation, you're not ready to talk about AI projects yet. Get aligned on the problem first.
If you want help getting aligned and choosing the next right project, Book Your AI Leadership Alignment Workshop today. We'll get your team clarity on the stuck metric, the system that influences it, and whether AI is part of the answer.
Jeff Richardson, CEO, Empowered Alliances
