Relearning math as an adult goes faster when you stop restarting from chapter one and instead find the exact point where your understanding actually breaks down, then build forward from there in short, regular sessions. Most people who call themselves bad at math aren't missing the whole subject. They're missing one or two specific links in a chain, and everything stacked on top of those links has felt shaky ever since.
Find the exact gap before you pick a textbook
Open a book at the level you think you need and you'll usually land on material that's either insultingly easy or completely opaque, with almost nothing in between that tells you where the real problem sits. A faster approach: pick a problem two or three steps harder than what you can currently do, try it, and trace backward through each step until you hit the first one that doesn't make sense. That step, not the start of a textbook, is your actual starting point.
This takes twenty minutes and saves weeks. Someone who freezes on a word problem involving fractions might discover the fractions themselves are fine and the real gap is translating a sentence into an equation. Someone stuck on algebra might find the algebra is fine and the gap is in how negative numbers behave. Neither person needs to reread a whole textbook. They need the one missing piece.
Why "start from the beginning" wastes your time
Chapter one of most math courses covers material adults already half remember: counting, basic operations, simple word problems. Working through it feels productive because you're getting everything right, but getting easy problems right isn't the same as closing a gap. It just delays the point where you hit the actual wall, and by then you've spent real time on material you didn't need.
Study in short, frequent sessions instead of long ones
Math skill is built through repeated practice spread across days, not a single long session. A problem type you can do today but haven't touched in two weeks will feel harder than it should, not because you forgot the concept, but because the specific moves involved (which method to reach for, how to set up the first step) fade faster than the underlying idea does. Coming back to the same problem type every few days keeps those moves accessible instead of letting them go stale.
What "short" means in practice
Twenty to thirty minutes, four or five times a week, beats a single two-hour session on a Sunday. The short sessions cost less willpower to start, and the spacing between them is doing real work: each time you return to a problem type after a gap, you're practicing retrieval, not just repetition.
Work the problem before you reread the explanation
When a problem doesn't click, the instinct is to reread the explanation until it feels clear. That produces a feeling of understanding that often doesn't survive contact with a new problem. A more reliable habit: attempt the problem first, even if the attempt is wrong, then check the explanation only to see exactly where your attempt diverged from the correct approach. You're not testing whether you can recognize the right method, you're testing whether you can produce it, and those are different skills.
This is uncomfortable at first, since it means sitting with a problem you can't solve for longer than feels natural. That discomfort is a normal part of building the skill, not a sign you're doing it wrong.
Keep a running list of what's still shaky
After each session, write down the specific step that gave you trouble, not the topic, the step: "setting up the equation from the word problem," not "word problems." A vague note like "fractions are hard" doesn't tell you what to practice next time. A specific one does, and it turns your next session into ten minutes of finding a new problem to test that exact step instead of guessing where to start.
Turn the gap into a specific prompt, not a vague subject
Once you know the specific gap, describing it as a narrow prompt rather than a broad subject is what makes a generated course useful. A prompt like "setting up equations from word problems involving fractions" builds a far more targeted course, with drills and review planned into the sequence, than a prompt like "algebra." And if the plan generated from that prompt covers ground you already have solid, it's fine to rewrite the plan and narrow it further before the course itself gets built.
Expect uneven progress
A few weeks in, some problem types will click fast and others will still feel slow, and that's normal rather than a sign you're behind. Math skills don't rebuild at a uniform rate because the gaps aren't uniform: one might be a single missing rule, another might be a habit of setting problems up the wrong way that took years to form. Judge progress by whether your list of shaky steps is shrinking, not by whether every topic feels equally solid yet.