In a fast-moving business environment, speed is often the default. When new tools or technologies emerge, the immediate response is to train quickly and solve the problem at hand. Just-in-time learning meets that need. However, when it comes to building deep capabilities in data and AI, quick fixes are insufficient.
“We need both,” says Michael Harwick, Director of Learning Design at Data Society. “Just-in-time training works best when it is truly immediate. You’ve recognized a need, and within two to three months of that need arising, you’re able to get the people who need to do something differently in a room and say, ‘Hey, remember when we talked about making a chance? We’re going to do it, and we’ll do it together.’”
That kind of responsive training plays a vital role in behavior change. The risk, however, is that it becomes a cycle of reaction without reflection. “There are folks who go through training periods where it’s all just reactive,” Harwick explains. “It starts to feel like ping pong because they’re moving from problem to problem rather than thinking about the broader arc of what they’re trying to solve.
This cycle is a common challenge for organizations launching their first upskilling program. The focus shifts to solving today’s problems instead of building tomorrow’s capabilities. In domains like business intelligence development or artificial intelligence training, this approach quickly runs out of steam.
Solving the Wrong Problem
A single need triggers many training programs. A team may request help with a tool, platform, or new process. While that training often delivers short-term value, it may miss a larger opportunity.
Harwick sees this dynamic often. “Some of what I see is a desire to solve a problem that is potentially too rigidly scoped,” he says. “The training opportunity is ‘I need Team X to be able to do this one thing,’ with no willingness to deviate from the system or prescription in place.”
This mindset limits growth. The best upskilling programs use specific training needs as a doorway to broader capability building. A request for a Tableau workshop can spark a more strategic conversation about data fluency. A one-time session on Python might evolve into a complete data science course. Training is most effective when it encourages new thinking rather than limiting people to narrow use cases.
Asking the Right Questions
To make that shift, Harwick recommends a familiar but powerful tool: leaders must step back and ask why the training need exists in the first place. “That question has done a lot of excellent work on some of the bigger programs that we’ve worked on,” he says. “This little thing that we’re asking people to do, how is it going to remain relevant? How is it not going to get outmoded? How is it consistent with the broader vision of where you would like people to go?”
This kind of thinking is essential for teams implementing an artificial intelligence and machine learning course. If the goal is adoption, retention, and genuine change, training must be integrated into a broader learning journey. Asking better questions and relentlessly pursuing the “why” builds that bridge.
MUST READ: Beyond the Hype: Why AI Training Alone Isn’t Enough
Capability Is Cross-Functional
Another risk of reactive training is that it often serves only one team at a time, leading to isolated learning and fragmented approaches to data, tools, and strategy.
“Just-in-time training can be great for getting together people who are very similar in function who all need to know the same process,” Harwick says. “But in thinking about broader capability, the interaction of different teams, different groups, different cohorts of learners who may have dramatically different day-to-day remits is the exciting part.”
For example, a strong data science for managers program brings together technical and non-technical teams. Business leaders learn how to ask the right questions, understand analytics outputs, and engage in decision-making conversations without needing to write code themselves. Meanwhile, analysts develop a deeper understanding of how their work aligns with business priorities. The result is a shared language and more strategic collaboration.
“Through these tactical interventions, you can give people a common framework and a common language,” Harwick explains. “But if you’re not thinking about the holistic vision, and you wind up teaching people different languages to address the same thing, then you’re not solving the problem.”
The Power of Instructor-Led Training
Instructor-led training provides organizations with the opportunity to explore both immediate challenges and broader goals in real-time. Facilitators can adapt the experience to meet the needs of learners, address emerging questions, and respond to shifting priorities. Whether delivered in-person or as virtual instructor-led training, this model creates space for nuance.
“We’re not afraid to ask clients to look two levels up,” Harwick says. “Why is this problem happening? What’s going on here?”
This approach strengthens the learning experience, particularly in a business intelligence development context or when launching an artificial intelligence and machine learning course. Technical knowledge alone is not enough. Participants need the opportunity to apply it, reflect on it, and connect it to real use cases inside the organization.
From Tactical to Transformational
Organizations often feel pressure to move quickly. But long-term capability requires more than fast answers. It requires thoughtful, strategic investment in learning design.
“Both of these things have their place,” Harwick says. “But if you’re constantly reacting without thinking about the broader arc, you lose sight of what you’re trying to become.”
The best upskilling programs strike a balance between immediate action and long-term vision. They begin with today’s challenges, but they are built to grow with the business.
Want to design an upskilling program that builds both speed and strategy?
Data Society delivers instructor-led and virtual instructor-led training designed to solve near-term problems while building long-term capability. Whether you’re exploring a data science course, launching a business intelligence initiative, or developing a data science for managers program, we’re here to help you align learning with growth.
FAQs: Strategic Training for AI, Data, and Leadership
Cross-functional training creates shared understanding between business and technical teams. This enhances communication, accelerates adoption, and facilitates better decision-making.
