A friend of mine, Priya, bought a 47-hour "complete data analyst" course during a flash sale. Eight months later she had watched maybe nine hours of it and could not remember a single thing she had built. The course looked gorgeous. The instructor had a great voice. None of that helped her get a job.
I have done the same thing more times than I want to admit. I once paid for a course because the sales page had a slick animated curriculum and a countdown timer. The content turned out to be slideshows read aloud. What I have learned, slowly and with some wasted money, is that the polish of a course tells you almost nothing about whether it will move you forward.
So I built a checklist. I run every course through it before I commit a single weekend. Here is the exact sequence I use, in order, so you can stop buying shiny things and start finishing useful ones.
Write down the outcome before you look at any course
Open a notes app and finish this sentence: "After this course I will be able to ___." Be embarrassingly specific. Not "learn Python" but "write a script that pulls data from a spreadsheet and emails a weekly summary."
This one habit kills half of all bad purchases. A vague goal makes every flashy course look relevant, because anything can sound like it fits when you do not know what you actually need.
If you cannot name the outcome yet, that is a planning problem, not a course problem. I have written before about how learning a skill without a plan quietly wastes months, and choosing a course is where that mistake usually starts.
If your goal could be satisfied by ten different courses, your goal is too vague. Tighten it until only two or three courses could plausibly deliver it.
Read the curriculum like a skeptic, not a tourist
Now open the course outline. Ignore the headline promises and look at the actual module titles and lesson lengths.
I scan for two things. First, does the curriculum move from doing to harder doing, or does it just pile up topics? Second, how much of the runtime is "intro," "welcome," "setting up your environment," and "course wrap-up"? Those are filler minutes dressed up as content.
A good outline shows you building something by the second or third module. A weak one keeps you in theory for hours and saves the real work for a single rushed section near the end.
Watch for the "comprehensive" trap
Courses love the word "complete" or "everything you need." Breadth is not depth. A 50-hour course that touches 30 topics for 90 minutes each will leave you fluent in nothing. I would rather have a focused 12-hour course that makes me genuinely competent at one job-relevant thing.
Check whether you build things or just watch things
This is the single biggest predictor of whether a course works for me. Skill comes from doing, so I want projects, exercises, and assignments, not a parade of demonstrations.
Look for language like "build," "you will create," "hands-on project," or downloadable starter files. Be suspicious of courses that are mostly the instructor typing while you watch. Watching someone code or design is roughly as useful as watching someone exercise.
When I taught myself design, the lessons that stuck were the ones that forced me to make real artifacts. That experience is the backbone of my notes on learning graphic design without a formal program, and the pattern holds for almost any practical skill.
If I cannot find evidence that at least roughly 30 percent of my time will be spent building or practicing rather than watching, I keep looking. Passive video alone has never taught me a skill I could use on Monday.
Vet the instructor and the recency
Two quick checks here, and both take five minutes.
First, has the person actually done the work, or only taught it? I look for a portfolio, a real job history, or projects I can see. "Best-selling instructor" tells me they are good at selling courses, which is a different skill from doing the thing.
Second, when was the course last updated? In fast-moving fields, a tutorial from four years ago can teach you tools and workflows that are already obsolete. Check the last-updated date and skim recent reviews for comments like "outdated" or "the screens look different now."
A five-star average from 80,000 students sounds impressive until you read the recent one-star and three-star reviews. Those are where people describe broken exercises, dead support, and content that no longer matches the software. Always sort to the critical reviews first.
Confirm the credential actually counts (if you need one)
Sometimes you are learning for yourself, and a certificate is irrelevant. Other times you want something an employer recognizes, and that changes the math entirely.
Be honest about which situation you are in. A random platform "certificate of completion" carries almost no weight with hiring managers. A credential from a known issuer can carry real signal, especially for early-career roles. I have walked through which ones tend to land in my breakdown of Google certificates that hold up in a job search.
If the credential matters to you, verify it before buying, not after. Search the exact certificate name plus "jobs" or "resume" and see whether real people report it helping.
Compare your top two on the things that actually matter
By now you should have narrowed it to two or three candidates. Put them side by side. I literally make a small table so the shiny one cannot win on vibes alone.
| What to check | Good sign | Walk away |
|---|---|---|
| Outcome match | Maps directly to your written goal | Vaguely "covers" your goal |
| Practice ratio | Projects and exercises throughout | Mostly watch-along video |
| Recency | Updated in the last year or two | No date, or visibly old screens |
| Instructor | Has done the actual work | Only known for teaching |
| Real reviews | Recent reviews stay positive | Critical reviews flag broken content |
The course that wins on this table is usually not the one with the best landing page. That is the entire point of doing it on paper.
Buy small, set a deadline, and protect your refund window
Do not commit to a giant bundle or an annual subscription on day one. Buy the single course, or use a short free trial, and treat the first week as an audition.
Give yourself a concrete first milestone, something like "finish the first project within seven days." Most platforms have a 7 to 30 day refund window. If the course feels like filler once you are inside, get your money back before that window closes. There is no prize for sticking with a course that is wasting your time.
Name the exact outcome. Read the curriculum like a skeptic. Confirm you build, not just watch. Vet the instructor and recency. Verify the credential if you need one. Compare your top two on paper. Buy small, set a deadline, and keep your refund window open.
A few common mistakes I keep seeing
People buy on sale instead of on need. A 90 percent discount on a course you will never finish is not a deal, it is 100 percent wasted.
People also confuse the feeling of learning with actual learning. A smooth, entertaining video can leave you feeling productive while teaching you almost nothing. Friction, like a hard exercise you have to retry, is usually where the real progress hides.
And people hoard. Twelve half-finished courses is not a head start, it is a guilt pile. One finished course beats a library of abandoned ones every single time.
How long should a good online course actually be?
Length matters less than density. A focused course of around 10-20 hours that you finish and apply will teach you more than a 60-hour course you abandon. Judge it by how much you will build, not by total runtime.
Are free courses worth it, or should I pay?
Plenty of free courses are excellent, especially for testing whether a subject even interests you. The real cost of any course is your hours, not the price, so run a free one through this same checklist. Pay only when a course clearly offers structure, projects, or a credential you cannot get for free.
How do I know if a course is too advanced or too basic for me?
Read the prerequisites and skim a sample lesson if one is offered. If the first project sounds slightly uncomfortable but not impossible, the level is about right. If you already know everything in the outline, or none of the words make sense, pick a different course.
Choosing well is mostly about being honest with yourself before you reach for your card. Run the checklist, keep the receipt handy for that refund window, and protect your weekends. The right course is the one you actually finish and then put to work.
