Over the last 18 months I’ve talked with hundreds of leaders at mid sized companies to talk about AI. Across those chats there’s been a pattern of 7 key questions that keep showing up in different ways no matter the industry, size or maturity.
This week I’m breaking down the 1st 4 questions and sharing some ways to tackle if you’re asking these same questions yourself. Use these as thought starters and ways to get rolling on AI before the end of the year so you’re in a good spot for 2026.
Stay tuned for next week’s newsletter where I’ll share the last 3 questions.
1. Where should we start & which use cases matter most?
Align on high-impact opportunities first
Start with a collaborative working session to define/align on a shared vision for AI (we call it an AI North Star). Use the session to identify the most valuable use cases and get stakeholder buy in. Focus on core business needs where AI can drive outsized impact, rather than chasing every idea. Prioritize by effort and impact and identify the quickest ways to get value (i.e., “quick wins”) that will be the lowest effort and have the highest impact.
Prioritize with data and speed-to-value in mind
Narrow your list to use cases that you can do quickly and where the data needed is easily accessible. Get a tangible win to build momentum and confidence for broader AI investment.
Leverage internal templates & tools
Maintain templates for use case backlogs and opportunity scoring to systematically rank ideas. Include criteria for ROI, feasibility, and strategic fit. A consistent way of scoring ensures you can stay organized and start with use cases that matter most to the business.
Putting this into action
Host a 3-hour AI North Star workshop with cross functional leaders. Capture goals, constraints, and success metrics.
Collect and score use cases by effort, impact, and strategic fit. Select one pilot and one backup.
Draft a one page brief per use case that lists owner, data needed, success metric, and risks.
✨ Special Offer ✨
We’re offering (2) AI North Star workshops free for mid-sized companies in Charlotte before the end of 2025.
If that’s you, simply reply to this email & we can quickly schedule time to discuss.
2. How do we actually get started & build momentum?
Start small, build to learn
Rather than lengthy analysis, kick off execution with a quick proof of concept or pilot on a high priority use case. This show-not-tell approach wins hearts and minds. A scrappy prototype can prove what is possible and energize your team. What used to take 6-12 months can now happen in a few weeks if you use AI tools and approach the development process differently. Start with what matters most and focus on core user needs. Build to learn! Speed-to-value and speed-to-learning beats polish every time.
Use rapid sprints & 90-day plans
Break the initiative into a 30/60/90-day plan with clear owners and milestones. In the first 30-days you might enable basic AI tools like enterprise GPT, Claude, Gemini, or Copilot. By 60-days you might pilot one or two use cases. At 90-days you’ll have tangible results and a validated/updated roadmap. This cadence creates sustained momentum through regular checkpoints and quick wins.
Accelerate with AI-native tools
Take advantage of AI tools to speed up delivery cycles. These fast track projects and build organizational confidence that “we can do this” which fuels appetite for the next one.
Putting this into action
Draft and share a simple 90-plan with owners, milestones, and success criteria for a pilot.
Do quick demos every 2-weeks. Decide to scale or sunset based on impact.
3. How should our business, products, & services evolve with AI?
Integrate AI into your strategy & vision
Make AI a core consideration in your business roadmap, not an afterthought. Regularly assess how emerging AI trends could disrupt your industry or change customer expectations. Lead future state ideation sessions on topics like AI in customer experience or AI-driven operations so leaders within your organization can envision what to build, transform, or sunset. The outcome might be a shift in your strategy or new service ideas enabled by AI.
Think along 3-horizons
Evolving with AI does not mean changing everything at once. Use the 3-horizons framework to balance immediate projects with longer term bets.
Horizon 1: Apply AI to streamline current processes and create efficiencies.
Horizon 2: Add AI features to products and services to differentiate.
Horizon 3: Explore new AI-driven business models to disrupt.
This approach ensures early wins while also thinking about bigger bets over time.
Pilot AI enhancements to core offerings
Identify parts of your customer or employee experience that AI can significantly improve and run small pilots there. Digital products might test an AI powered feature such as a recommendation engine to gauge impact on engagement. Service based businesses might experiment with AI driven personalization of service delivery. Use the results to guide broader roll out and treat each pilot as a learning opportunity. Scale what works and document any lessons learned.
Putting this into action
Map one critical journey and pinpoint where AI can help reduce effort or increase confidence for customers or your team.
Write out a 3-horizon view for things to do now, do next, and do later. Getting a little meta.. if you’re not sure where to start, try giving ChatGPT/Claude/Gemini some context about your business/customer/team and asking it to generate some ideas. That’s a great way to get past the blinking cursor/blank screen syndrome if you feel stuck 😉
4. What’s the business case?
Link AI to real outcomes
For every AI initiative it should be pretty clear how it will make or save money or otherwise improve key business metrics for your company. Tie projects to specific KPIs and baseline them up front and use simple ROI calculators to forecast value.
For example, if getting a new AI tool costs $30/month per user and you anticipate it will save each employee 3 hrs/wk, do a simple calculation of average hourly rate X # of employees X hours saved per week X # of employees impacted X 52 weeks = annual return on that investment.
Average hourly rate | $50/hr |
Savings per week | 3 hrs/wk |
# of employees impacted | 100 employees |
Estimated total savings per week | $15,000/wk |
Estimated total savings per year (52 weeks) | $780,000/yr |
Present the math in leadership language such as $’s, hours, and risk reduced.
Start with quick wins to build the case
It’s way easier to justify spending after your can prove success/results. Pilot a low cost project specifically to generate an internal case study. Automate a manual data entry task or improve a single workflow step. When the pilot yields a % time savings (define what makes the most sense for your org), you’ve got a proof point.
Multiply that pattern across processes and you have a credible case for larger investments. Collect before/after metrics from each experiment and compile them into an AI ROI portfolio. Include the cost of doing nothing and the competitive risk of inaction.
Manage expectations and deliver iteratively
Avoid overhyping AI as a magic bullet by setting realistic scope/expectations and emphasize you you’ll get value incrementally. End each iteration with a measurable result you can take to the CFO/board/executive leadership. Highlight financial ROI and qualitative benefits such as improved customer experience and better decision making. State the investment, expected return, timeline, and risks. Discipline earns trust and keeps focus on business value.
Putting this into action
Define KPIs and baselines before you build. Show the math in simple terms that leaders use.
Choose one quick win pilot that can deliver measurable savings within 90-days.
Build an AI ROI portfolio across use cases and track the cost of inaction.
Wrapping up
You don’t need a giant AI program to get started. You simply need to demonstrate value quickly in an areas of the business that matters, prove employees will use it, and a safe path to scale.
Honestly, the behavior and mental model shift matters more than the actual use case or AI tool. Build habits in your organization now that make you faster and more customer centered next quarter. Keep testing/delivering value where your teams already work. Measure results/outcomes that leaders actually care about.
If you need a sounding board, shoot me a note. Happy to jam on ideas together 😊
Onward & upward,
Drew