I’ve seen it. I’ve read about it. And the pattern is consistent: most companies jump into AI the same way they jumped into digital transformation a decade ago, i.e., fast, fragmented, and without a clear definition of what success looks like. That approach doesn’t just waste money but also creates performative progress, the kind that looks good in an all-hands presentation and falls apart in execution.
If you want AI to actually move the business forward, the quality of your upfront questions matters more than the sophistication of the technology.
Here’s a practical set of questions leaders should ask about their AI strategy before they commit to it.
1. What Problem Are We Actually Trying to Solve?
If this isn’t painfully clear, stop. It’s the first question that I have every leader ask before implementing any new strategy, technology, protocol, etc. or developing any new products.
- Is this a customer problem, an employee friction point, or an operational inefficiency?
- Are we solving something meaningful or just something measurable?
- Would fixing this without AI still be meaningful or relevant?
Reality check: If AI disappeared tomorrow, would this initiative still be worth doing? If not, it’s a gimmick.
2. Where Does This Sit: Culture, EX, or CX?
This is where most strategies fall apart.
- Does this improve how employees do their work (EX)?
- Does it improve how customers experience the brand (CX)?
- Or does it reinforce or erode your culture?
Most AI conversations skip these questions entirely, which is why most AI initiatives don’t stick. Culture is the operating system; EX is how that system runs day-to-day; and CX is what customers feel when it’s working or not. If you deploy AI without knowing where it sits in that Thread, you’re introducing a variable into a system you don’t fully understand.
Hard truth: If your culture is unclear or inconsistent, AI will amplify the chaos. It doesn’t fix broken foundations; instead, it scales them – faster and in front of your customers.
3. What Decisions Will AI Actually Influence?
AI without decision impact is just expensive reporting.
- What decisions will change because of this?
- Who will act differently? and how?
- Are we automating decisions? augmenting them? or just informing them?
More hard truth: If no behavior changes, nothing changes.
4. Do We Have the Data? and Is It Worth Using?
Data is at the heart of designing and delivering a great experience. But not all data is usable, and not all usable data is valuable.
- Do we have enough reliable, contextually relevant, and clean data? Or are we working with whatever we’ve collected?
- Are we solving for signal? or have we confused volume with value?
- Are there ethical or trust implications in how we use this data? and have we had that conversation or just assumed it’s fine?
Watch out: Bad data + AI = faster bad decisions.
5. Are We Designing for Adoption or Just Implementation?
Most AI initiatives fail here.
- Will employees trust it? and how will we train them to work hand in hand with it?
- Does it fit naturally into workflows? or does it disrupt them?
- Have we designed for behavior change, not just system deployment?
Hard truth: If people have to “go out of their way” to use it, they won’t.
6. What Does Success Look Like?
As with any other project, initiative, or implementation, success metrics are critical. Without them, how will you know you’ve obtained your goal and solved the problems you set out to solve (see Question #1).
- What KPIs will move?
- How should the experience feel different for employees and customers?
- What does success look like? What will make us say, “This was worth it”?
Hard truth: If what success looks like is vague, outcomes will be too.
7. Where Could This Go Wrong?
This is the question leaders avoid, but it’s also the one that saves you.
- What risks are we introducing (e.g., bias, errors, over-reliance)?
- Where could trust break, with employees, customers, or both?
- What’s the fallback when the system (or the entire implementation) fails?
Conduct a pre-mortem and a post-mortem. A pre-mortem asks: “What could go wrong before we begin?” A post-mortem asks: “What went wrong and why?” You need both. The pre-mortem is where leaders have the most leverage, but they use it the least. It’s the moment to name the failure scenarios nobody wants to say out loud – but they need to.
Hard truth: If you don’t plan for failure, you’ll experience it publicly. Don’t overlook these questions.
8. Are We Building Capability or Dependency?
Short-term wins can create long-term constraints.
- Are we developing internal understanding? or are we outsourcing intelligence?
- Will this make us smarter over time? or will we just be more reliant on vendors?
- Can we evolve this capability as the business changes?
Hard truth: AI should compound your advantage, not rent it.
9. How Will This Scale Across the Organization?
Pilot success often dies in enterprise reality. Don’t skip the pilot.
- Can this work across teams, geographies, and use cases? or did we just design it for the team that was easiest to pilot?
- Does it integrate with existing systems and processes? or does it create a new layer of workarounds?
- Are we building something reusable? or a one-off that someone will have to maintain forever and explain to everyone?
Hard truth: If it doesn’t scale, it doesn’t matter.
10. What Are We Choosing Not to Do?
Strategy is as much about what you’ll be doing versus what you’re not going to do.
- What initiatives are we de-prioritizing (or avoiding) because of this? and was the trade-off deliberate or by default?
- Are we spreading resources too thin across AI initiatives so that no one feels like they’re falling behind?
- Is this genuinely the highest-leverage move right now? or does it just have the most internal momentum?
Hard truth: If everything is a priority, nothing is.
In Closing
AI isn’t a strategy. It’s a force multiplier. It reflects what’s happening – good, bad, or ugly – but it won’t roll up its sleeves and lead for you.
If your culture is unclear, your processes are broken, or your customer experience is inconsistent, AI will make those problems bigger, faster, and more visible. If you’re clear on what matters, but also disciplined about how you execute, AI can accelerate the outcomes you actually care about.
In my next article, I’ll go deeper on the one thing that determines whether any of this works and the one thing most AI conversations never actually address: culture. If the foundation is broken, no amount of sophisticated technology fixes it. It just makes the cracks harder to ignore.
AI is a catalyst for innovation. Leaders who embrace AI into their company culture avoid digital Darwinism. ~ Contentstack
Want this thinking applied inside your organization?
ABOUT ANNETTE
Annette Franz is an internationally recognized employee and customer experience thought leader, coach, speaker, and author who has helped organizations around the world build cultures that empower employees and create consistently better customer experiences.
She has authored multiple books that provide practical frameworks for leaders, including:
- Employee Understanding: A Three-Pillar Framework for Designing a Great Experience and Driving Business Success
- Built to Win: Designing a Customer-Centric Culture That Drives Value for Your Business
- Customer Understanding: Three Ways to Put the “Customer” in Customer Experience
Sign up for our newsletter for updates, insights, and other great content that you can use to up your EX and CX game.
Image courtesy of Pixabay.
