Learning Science Meets the Realities of Product Design

Turning learning science into product decisions

The hard part of applying learning science to a product isn’t stating the principles. It’s making them work for learners with limited time, competing goals and uneven motivation. That’s the challenge taking shape at elevator9, where learning theory is informing platform development. The distinction matters: a principle can sound persuasive in a presentation and still prove difficult to build into a routine people can maintain.

Many platforms are built without learning expertise in senior leadership. I once met a team whose guiding principle was to build what customers asked for. That sounds responsive, but customers’ requests don’t always address what they need to learn. People are often better at describing an immediate frustration than identifying the instructional design that would help them handle a future task. It may be an easy way to start building; it’s a weak basis for a lasting learning strategy.

There’s a product judgment hidden in that gap. Teams have to listen to learners without assuming that every requested feature will improve learning. They also have to avoid treating theory as a substitute for listening. The useful work is translating between what people say they need, what the evidence suggests, and what the product can reasonably support.

Spacing meets the complexity of real programs

Elevator9 is exploring how spacing and reactivation can help people retain and apply what they encounter in face-to-face events. The aim is to bridge the gap between attending a session and performing well later. A workshop can create a moment of attention, but the value of that moment depends partly on what learners remember and can use after they leave. Follow-up activities may help keep ideas available; designing them well requires more than sending a reminder.

The challenge is that much learning research is conducted with college students, over short periods and around a single learning goal. Real programs rarely fit those conditions. In a workplace program, learners may be balancing new material with existing responsibilities, and the intended outcome may be a change in how they act, not simply whether they can recall information. That makes it harder to carry a finding from a controlled study directly into a product feature.

Leadership development, one of the areas that prompted CEO David Grad to launch the initiative, may involve several events, each introducing new goals while revisiting earlier ones. A session might focus on recognizing emotions in oneself and others; a later meeting might turn to coaching. Designers then have to decide what to reinforce, when to introduce new material and how to keep prior learning active. Those choices affect the sequence of the program, the demands placed on learners and the signals a platform needs to track.

The sequencing problem is easy to underestimate. If a follow-up prompt arrives before a learner has had a chance to use an idea, it may feel detached from the work. If it arrives too late, the connection to the session may have faded. And when a program has several goals, repeating one concept can compete with time needed for another. There isn’t a universal schedule implied by the idea of spacing; the design has to account for what learners are trying to do and how the program unfolds.

The learner’s schedule adds another layer. Short activities can make follow-up more manageable, but people may still be busy or unmotivated. Giving learners a way to reschedule activities around their own availability could help, though it also creates design questions for both the learning experience and its interface. How much flexibility supports follow-through, and when might extra choices make the experience harder to manage? Those are product questions as much as instructional ones.

There isn’t a finished formula for these conditions. The work calls for repeated trials, evaluation and revision. That cycle is part of the learning: the team is discovering what the platform needs to do while building it. The numbers here are underwhelming only if you expect a clean recipe from research. In practice, testing how an idea behaves in a real program can reveal the constraints that a neat principle leaves out.

Building the product, and learning along the way

Elevator9 is moving from an initial concept toward a system shaped by both what people say they need and what learning expertise suggests they need. The work has also grown from an app-building effort involving a small group into a broader development project with a full team. New problems come with that change. More people and specialties mean more perspectives to bring into product decisions, but also more coordination around what the product is meant to accomplish. My own background is in learning design and cognition, so I’m listening to colleagues in other areas and contributing where I can.

That shift asks the team to treat development as an exchange between disciplines. Learning expertise can clarify what a feature should support, while engineering and product experience can expose assumptions about what’s practical to build and maintain. Neither perspective answers every question alone. If you’re working in this space, the implication is familiar: a good learning idea still has to survive contact with everyday use.

The same experimental approach is at work at LDA. There, we’re testing curricula and refining the offerings as we work to improve the learning and development field. Both efforts face established competitors, which makes progress demanding. It also makes practical evidence more valuable than confidence in a plan that hasn’t been tested. A claim about what learners need is more useful when the team can see how it holds up in an actual program.

Here’s the thing: iteration doesn’t guarantee a better product. It gives a team a way to find out where its assumptions fail and decide what to change. That requires a clear account of what each trial is meant to teach. Otherwise, revision can become motion without much learning attached to it (and this is the part most people overlook).

What the work could mean for learning platforms

The useful exchange runs both ways: theory can improve a solution to a real problem, and the attempt to apply theory can expose what the theory needs to account for. Getting close to the work means accepting that the design will change as new constraints emerge. For learning platforms, the test is whether the principles hold up across actual schedules, multiple goals and the demands of performance.

This is more significant than it looks. A platform built around learning science has to make choices about timing, repetition and learner control, then show that those choices fit the program it serves. The interface may make an activity easy to complete, but ease alone doesn’t establish that the activity supports retention or application. The harder question is whether learners can carry something useful from a session into the situations that follow.

That said, evidence from one program won’t automatically settle the design for another. Different goals and schedules can change what a useful follow-up looks like, so teams need to be precise about what they’ve tested and what they’ve learned. The next phase for work like this is less a dramatic reveal than a series of grounded decisions. Small tests. Better questions. And a clearer view of where learning theory meets the conditions people actually face.