Make It Stick: The Science of Successful Learning – A Summary for L&D Professionals
Published in 2014 by cognitive psychologist Henry Roediger, Mark McDaniel, and science writer Peter Brown, Make It Stick distills decades of learning science research into a book that is simultaneously accessible and unsettling. Unsettling because its central finding is that the study and practice methods most people trust — and that most training programs rely on — are among the least effective available.
For L&D professionals, this is not background reading. It's a direct challenge to the way most organizational learning is designed.
#What doesn't work (and why people keep doing it anyway)
#Re-reading and highlighting
The most common study method is re-reading, often combined with highlighting. Research consistently shows it produces minimal durable learning. Familiarity — the feeling of recognizing material — is mistaken for actual retrieval capability. Learners feel productive because the content feels familiar; they are actually just priming recognition, not building the ability to recall or apply.
This is the illusion of knowing: a subjective sense of mastery that isn't supported by actual performance. It's particularly dangerous in professional training because learners — and sometimes trainers — use that feeling of familiarity as evidence that learning has occurred.
#Massed practice
Massed practice means concentrating study or practice in a single session: a one-day compliance training, a three-hour workshop, a module completed in one sitting. It feels efficient and produces good short-term performance on immediate tests. But the research is clear that retention drops sharply without spacing. Material learned in a concentrated block is largely forgotten within days.
The problem for L&D is structural: organizational schedules favor massed delivery. One-time training events are easier to administer. Spaced learning requires ongoing commitment and system support that most organizations don't have infrastructure for.
#What actually works
#Retrieval practice
The most consistently supported finding in learning science is the testing effect, which Make It Stick calls retrieval practice. The act of recalling information strengthens memory more than re-reading or reviewing the same material. Low-stakes quizzes, practice problems, and self-testing are not just assessment tools — they are learning tools. Every time a learner retrieves information, the memory trace is strengthened and becomes more durable.
The implication for course design is direct: content review followed by quiz is a weak model. Practice that requires active retrieval — not recognition from options, but genuine recall — is the superior design choice wherever it's achievable.
#Spaced repetition
Spacing practice across time dramatically improves long-term retention. The optimal spacing depends on how long material needs to be retained, but even modest spacing (reviewing material days or weeks after initial learning rather than immediately) produces substantial improvements. For workplace learning, this means that a single course event is rarely sufficient for skills that need to be durable.
#Interleaving
Interleaving means mixing different types of problems or content rather than practicing one type exhaustively before moving to the next (known as blocked practice). Blocked practice feels easier and produces better results on immediate tests; interleaved practice feels harder but produces substantially better retention and transfer. The mechanism appears to be that interleaving requires learners to discriminate between problem types — a cognitive demand that strengthens learning.
#Elaborative interrogation
Connecting new material to existing knowledge — asking "why does this work?" rather than just "what is this?" — builds more durable memory traces. This is called elaborative interrogation. For course design, it suggests the value of prompting learners to explain, apply, or connect content rather than simply exposing them to it.
The gap between what learners believe works and what the research supports is a consistent finding across multiple studies cited in the book. Learners reliably rate massed practice and re-reading as more effective than retrieval practice and spacing — precisely because the latter feel harder and produce less immediate fluency. Desirable difficulty is a recurring concept in the book: the conditions that make learning feel effortful tend to be the conditions that produce durable retention.
#The illusion of knowing in organizational learning
The illusion of knowing is particularly consequential in compliance and safety training. Learners who complete a course on a regulatory requirement may feel confident they know the rules — especially if the course used a lot of recognition-based testing. But if they haven't been required to actively retrieve and apply that knowledge, their actual performance under real conditions may not reflect their confidence.
Make It Stick doesn't prescribe a curriculum design framework, but it makes the case for designing learning systems rather than learning events. Retrieval practice requires return exposures. Spaced repetition requires a schedule. Neither can be achieved with a single-shot course, however well designed.
#Implications for L&D design
The book's most direct implication is that the minimum viable learning intervention for durable knowledge is: initial exposure, spaced repetition, and retrieval practice. Not just once, but across time. For organizations that rely on annual compliance refreshers or one-time onboarding courses, this research is a challenge to the model itself.
Practically, this means building learning into workflows, scheduling follow-up activities, using brief knowledge checks over time rather than end-of-course quizzes, and treating learning as a process rather than an event.
Scibly's structure supports exactly this kind of extended learning design — assigning courses and follow-up activities, tracking completion over time, and making it straightforward to build spaced learning into organizational workflows rather than treating each training as a standalone event.