Brilliant: Learn by doing

Analysis by Appcrazy

Brilliant: Learn by doing is the kind of education app I reach for when I want to understand an idea rather than simply memorize a formula. It focuses on mathematics, data, and computer science through short interactive lessons, and that changes the experience from reading a chapter to making small decisions and seeing what follows. I found it especially useful for filling gaps in my understanding, testing whether I really grasp a concept, and turning a spare part of the day into a focused learning session.

The app is free to install, with optional in-app purchases ranging from $0.99 to $359.88 per item. It is made by Brilliant.org and is suitable for Everyone. Its Android listing shows an average rating of 4.5 from around 102 thousand ratings, with more than 5 million installs, so it has clearly found a large audience. The current version is 9.19.0, and it supports Android 7.1 and later. Those details matter less than the learning design, but they help set expectations: this is an established mobile learning tool, not a tiny experiment or a reference dictionary.

Where the learning experience usually gets stuck

My first adjustment was accepting that Brilliant is not designed to be passive. I could not get much from rushing through screens or tapping an answer just to move forward. The lessons regularly ask me to predict, calculate, compare, or identify a pattern. That makes the app more engaging than a conventional article, but it also means I have to slow down and think. Someone expecting a video course that can play in the background may find the interaction demanding rather than convenient.

The most common point of friction is the gap between recognizing an explanation and being able to use it. A lesson may make an idea look clear, then ask a question that changes one detail. That small change exposes whether I understood the reasoning or only followed the example. I consider this a strength, but it can feel uncomfortable when I am tired or returning to a subject after a long break.

Mathematics can be particularly unforgiving. If I make an arithmetic mistake, misunderstand a symbol, or skip a step mentally, I can end up treating the app as if it marked a correct idea wrong. My best habit is to write down the intermediate steps before entering an answer. That sounds obvious, yet it prevents a surprising amount of wasted time on probability, algebra, and data-related exercises.

Computer science lessons create a different kind of difficulty. The challenge is often conceptual rather than dependent on typing a long program. I need to track a process, understand an algorithm, or reason about how information changes. If I approach those activities as puzzles to guess, I may arrive at the next screen without knowing why. Replaying the explanation and describing the process in my own words works better than trying random answers.

Advertisements

Another place users can get stuck is choosing a reasonable starting point. The subject range is broad, and a learner who opens the app with only the vague goal of “getting better at math” may jump between attractive topics. I get more value when I choose one clear purpose, such as reviewing probability before a course, building comfort with logical reasoning, or exploring introductory computer science. A narrow starting goal makes the short lessons feel connected instead of scattered.

The bite-size format is convenient, but it can encourage false confidence. Completing several small activities is not the same as mastering a full subject. I treat each lesson as one step in a longer process and pause when I notice that I am selecting answers by pattern recognition alone. The app is strongest when I use it to develop intuition and then reinforce that intuition with deliberate practice elsewhere.

Take a Look at Our Blog

Setup checks before blaming the lesson

Before beginning seriously, I check that the app is running on a supported Android device. Brilliant lists Android 7.1 as the minimum operating system, so an older phone may create problems before the lessons even begin. I also make sure the device has enough battery and a stable connection when starting a session. This is not because every activity necessarily needs constant attention from me, but because interruptions during an interactive lesson can make it difficult to tell whether a problem came from my answer, the app, or the connection.

I also review the account and purchase path carefully. The app is free, but optional purchases are part of the experience, and the listed range is wide. I would not begin a subscription or purchase decision while distracted in the middle of a lesson. I first decide whether the free access gives me enough opportunity to judge the teaching style, then consider paid access only if I expect to use the service consistently. That approach is more sensible than paying simply because one difficult topic feels temporarily urgent.

Notifications and study timing deserve a little thought too. A short lesson can fit into a commute, a lunch break, or the quiet part of an evening, but I learn better when I reserve enough uninterrupted time to finish a reasoning chain. If I start while rushing between tasks, I tend to make careless mistakes and remember the frustration more than the concept. I prefer a simple routine: one topic, one or two lessons, and a brief note about what I still cannot explain.

For families, the Everyone content rating makes the app broadly approachable, but that does not mean every lesson will match every child’s level. A younger learner may need help with reading, notation, or the difference between a conceptual question and a calculation. I would sit with a child during the first sessions rather than assuming that a general age rating guarantees independent use. The same applies to adults who have been away from formal study for years; the material may be accessible, while the habit of working through abstract questions takes time to rebuild.

One useful preparation step is to keep a notebook or a notes app nearby. Brilliant’s interactive format makes it tempting to do everything mentally, yet written work reveals where the reasoning breaks. I use the notebook not as a second textbook but as a debugging tool. When I miss an answer, I record the assumption I made, not merely the correct result. That turns an incorrect response into information I can use later.

Recovering a session after a wrong answer or interruption

When I get an answer wrong, I try not to immediately repeat the same approach. I first identify the type of error: did I misunderstand the question, apply the wrong rule, calculate incorrectly, or overlook a condition? This small classification is one of the most practical ways to use the app. It stops a lesson from becoming a sequence of guesses and makes the interactive feedback more valuable.

If an interruption breaks my concentration, I restart from the last point I clearly understood rather than forcing myself to continue. A quick review of the preceding explanation is usually more productive than staring at the current question with no context. I then solve the problem on paper and return to the app. This is especially helpful in lessons where each answer builds on the previous one.

For a difficult topic, I use a two-pass workflow. On the first pass, I focus on the main idea and accept that some details may not stick. On the second pass, I slow down, explain each step aloud, and look for the assumption behind every answer. The first pass gives me a map; the second tests whether I can navigate it. This approach is more effective for me than repeating the same activity immediately after a mistake.

I also avoid treating progress as a race. The short lesson structure can make it satisfying to keep moving, but speed is not the goal in subjects where understanding depends on careful reasoning. If I cannot explain why an answer is correct without looking back, I consider that a signal to pause. Brilliant works best as an active study tool, not as a checklist of screens to clear.

A further recovery technique is to connect an abstract lesson to a real situation. After studying probability, I might think about why a small sample can be misleading. After working through data concepts, I might examine how a chart could create a false impression. After a computer science activity, I might describe the algorithm as a sequence of everyday instructions. These connections help me discover whether the idea has become usable knowledge rather than a temporary answer pattern.

It is also worth separating a content problem from a navigation problem. If I am confused by the concept, I revisit the reasoning. If I cannot find the lesson I intended to study or the session feels disorganized, I stop changing topics and return to the original learning goal. Constantly searching for a “better” lesson can become a form of avoidance. A simple written plan—subject, topic, and one question I want answered—keeps the workflow under control.

When the app is not the cause

Not every frustrating session indicates a flaw in Brilliant. Sometimes the real issue is fatigue. Interactive learning requires working memory, and I notice that my accuracy drops when I try to study late at night or immediately after a demanding task. In that situation, a shorter session or a full break is more useful than blaming the lesson design.

Screen size can matter as well. A compact phone is convenient for quick practice, but diagrams, notation, and multi-step questions may feel cramped. If I am struggling to inspect a visual or read a relationship between elements, I first change the device position, increase readability where the system allows it, or move to a larger screen. A display limitation can look like a teaching limitation when the real problem is simply that the information is difficult to see.

Connection instability is another source of confusion. If an activity loads slowly, behaves inconsistently, or resumes in an unexpected place, I check the network before repeating answers. I avoid submitting the same response several times while the screen appears unresponsive, because that only makes it harder to understand what happened. Closing and reopening the session calmly is preferable to turning a temporary connection issue into a mistaken conclusion about my progress.

My background also affects how the app feels. Someone comfortable with algebra may find an introductory mathematical idea pleasantly quick, while a learner who has forgotten basic notation may need more preparation. Conversely, a person with strong school mathematics but little experience with algorithms may find the computer science material more demanding. I would not judge the whole app from one subject area. Its broad coverage means the difficulty can vary noticeably depending on what I already know.

This is where Brilliant differs from common alternatives. A traditional textbook gives more space for definitions, worked examples, exercises, and reference material, making it better when I need a complete, searchable treatment of a subject. Video platforms can be easier when I want a teacher to explain a difficult idea from start to finish, especially if I learn well by listening. Practice websites may offer more repetition and exam-style questions. Brilliant’s advantage is the middle ground: it asks me to interact with ideas immediately and uses compact problems to expose weak understanding.

That middle ground is also its limitation. I would not rely on it alone for a formal exam syllabus, professional certification, or a programming project that requires extensive coding practice. It can improve intuition and reasoning, but it is not a substitute for the exact exercises, references, or tools demanded by a particular course. If my goal is to pass a narrowly defined test, I would pair it with the official curriculum. If my goal is to become comfortable thinking mathematically or computationally, it becomes much more attractive.

Who gets the most from Brilliant

I think the app suits curious adults, students who want a more active supplement to class, and people returning to technical subjects after a long absence. It is particularly good for the learner who says, “I can follow the explanation, but I want to know whether I can use it.” The interactive questions provide that check more directly than passive reading does.

A realistic everyday use case would be a commuter preparing for a data-focused course. Instead of trying to read a full chapter on a crowded train, I could complete a short lesson, pause at a question, and write down the point that confused me. Later, I could use that note to guide textbook study or ask a teacher a more precise question. The app does not replace the course; it helps me arrive with better questions and stronger intuition.

It can also work for an adult who wants a structured mental challenge without committing to a long class. The key is to choose consistency over intensity. A few thoughtful sessions are more valuable than opening the app once, racing through easy material, and abandoning it when the first unfamiliar concept appears. I would use the free entry point to test whether the lesson style fits my attention span and learning goals before spending money.

I would recommend skipping it, or at least treating it as a secondary tool, if you want a passive lecture library, a complete reference manual, or extensive hands-on programming practice. It may also frustrate learners who need every term defined before attempting a problem. The app often expects me to tolerate a little uncertainty and learn through the question itself. That is productive for some people and uncomfortable for others.

My overall impression is that Brilliant.org has made a thoughtful education app for active learners. The developer’s focus on short, interactive study makes the experience feel more like guided problem solving than ordinary mobile reading. The free price lowers the barrier to trying it, while the optional purchases mean I would pay attention to the purchase screen and decide based on sustained use rather than curiosity alone.

After using it, I would describe Brilliant: Learn by doing as a strong companion for building intuition in math, data, and computer science. Its best moments arrive when I slow down, write out my reasoning, and treat mistakes as clues. Its weaker moments appear when I need exhaustive coverage, long-form explanation, or a direct route to an exam answer. If you want to think through concepts instead of merely consume them, I believe it is worth trying; if you need a complete course replacement, I would choose a more specialized resource and keep Brilliant as a supplement.

The practical verdict is simple: start with one clear subject, use a notebook, study when you can concentrate, and judge the app by whether your explanations improve—not by how quickly you finish lessons. With that approach, this is an unusually effective way to turn small pockets of time into genuine practice, while its limitations remain easy to recognize and manage.