I’ve heard from hundreds of leaders at mid sized companies about AI over the last 18 months. Across those conversations seven questions kept showing up in different ways.
If you missed it last week, I shared the first 4 questions and gave you practical ways to get going before the end of the year. This week I’m picking up where I left off and going through the last 3 questions with tips/tricks to put this into action.
5. Are we ready from a data, process, & tooling standpoint?
Data readiness
AI efforts live and die by data so you’ll need to take inventory of where relevant data resides, how clean/accurate it is, and any gaps or issues with accessing. Start with processes where data is already pretty well organized and accessible/available. For example, if you want to streamline customer service/support but your knowledge base is a mess, fix hygiene first or choose a different use case.
Processes & governance
Process and policy is a huge part of being ready for AI. Setup a basic AI governance model early to address any compliance risks, ethical considerations, and usage guidelines. The most basic example here is giving a set of simple internal policies like not sharing/uploading customer sensitive data into public tools or only using tools that IT has permitted. My guess is you probably already have something like this in place, but it’s important to treat it like a living document and continue to iterate as things evolve.
Putting this into action
Run a quick data audit for the pilot. List systems and owners, who has access, how sensitive the data is, and how long should the data be retained/stored and accessible to AI tools.
Stand up secure access for your team. Also, make sure any external connectors/integrations are secure as well. Check permissions and add any “human-in-the-loop” steps if/when needed.
Draft/share a one-pager with your team that’s written in super plain english and outlines what acceptable AI tool use looks like, when human review is needed/required, and when to escalate a concern (and to whom).
Gif by rockymovie on Giphy
6. Will our team adopt this? How do we train them fast?
Cultivate AI champions
Top down mandates almost never work, unless you’re a dictator (p.s. don’t do this 😬). Try launching an AI ambassador program by finding early adopters and/or power users in each department to help lead by example. These folks can try new AI tools first, share success stories, and coach/guide their peers. This helps skeptics see real/practical benefits and creates a natural support network.
Hands-on training & upskilling
Training that meets your team where they’re at is critical. Focus on practical sessions and start with basics around asking good questions (aka “prompt engineering”), providing the right amount of context in the right structure (aka “context engineering”). Do this first then move to sessions that are focused on specific tools, use cases, and workflows.
Cheat sheets and short video demos are great resources to share after sessions for anyone struggling to internalize the change. Also, setting up internal comms channels like a Slack/Teams channel or hosting AI office hours for trouble shooting and sharing tips is a great place to start.
Remember, this is a massive paradigm shift for people and depending on your team/company’s current understanding and maturity, this will likely take some time.
Include people in the decision-making process & make AI feel approachable
People want things to be done with them, not to them. Involve people in the brainstorming and testing process so they feel a sense of ownership. If you’re rolling out a new tool or workflow, co-create the plan with key team members to help them become champions. You’ll get way better results this way 😉
A lot of folks are terrified of AI. It feels like this massive thing to learn the tools and incorporate into their daily work. You’ve got to spend a lot of time showing (not telling) people how the AI tools are easy and accessible to lower the perceived barrier to entry.
Putting this into action
Find and identify “ambassadors” and give them early access. Ideally this is in each department/team/function, but starting out it might be a single person that’s already dabbling. Include them in the process of deciding on which tools to use and how to rollout. Focus on appetite/interest/enthusiasm about using the tools over their title. Better to have someone who’s eager and can help you get some quick wins, than someone who’s high in the company and doesn’t make time to learn (or believe AI will help them).
Run working sessions/workshops focused on tackling real tasks.
Share a living library of simple checklists and short videos folks can reference easily.
Hold weekly office hours.
Gif by WBPictures on Giphy
7. Who should lead this & what roles do we need?
Exec sponsorship + a task force
Assign an executive champion (if you’re reading this newsletter, this might be you) to lead the charge and to be accountable for results. Form a cross functional AI task force with representatives from key functions. The goal is to drive the roadmap, allocate resources, and remove roadblocks. Visible support from exec leadership shows how important AI is (or should be) to the whole company.
Key roles to staff
There are some important skills/capabilities you’ll need to start building momentum:
AI Strategist/s or Product Manager/s: There’s a lot of debate on what to call this role, but ultimately you need someone who can bridge business needs/goals with technology. They should have a good handle on AI capabilities and deeply understand the business so they can drive use case prioritization, tool selection, and workflow optimization that deliver real value.
Fullstack Engineer/s: We call these folks “AI-forward engineers.” These are folks comfortable and familiar with AI capabilities and traditional fullstack engineering and can support building custom integrations and AI agents, once you’re ready to move past basic tools like Copilot, ChatGPT, Gemini, Claude, etc.
Depending on the size of your organization and maturity, it might not be feasible to have a dedicated team. If this is you, start by identifying internal people that show the right aptitudes in data, software, or analytics and give them dedicated time to focus on this.
Augment & upskill
Partner with outside experts to kickstart learning and initiatives while transferring knowledge to your team. Invest in training so your folks can take over sustainably. The goal is to spread AI skills across the company to help transform how the entire company operates.
Putting this into action
Name an exec sponsor. Publish a simple AI charter so everyone’s aligned on the north star for AI within your company.
Stand up a lean cross functional team with clear time allocation and a mandate to deliver value.
Pair with outside experts for the first wave while you upskill the internal team.
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 as you work through your approach for 2026, shoot me a note. Happy to jam on ideas together 😊
Onward & upward,
Drew