How to stop repeat free trials on an online course platform
When a whole course fits inside a free trial, repeat trials cost you the sale. The arithmetic, the fixes that help and the basic checks that break.
The cancellation came in on day 6 of a 7-day trial, and the comment box was filled in: "Did SQL on my other email last month, Python on this one. Dashboards next. Best free education I've ever had, thank you!"
It's the nicest thing anyone has written in that box all quarter. It also tells you exactly what you didn't get paid. This is a scenario, and the costs are assumptions to replace with your own.
Say you sell a course library for $39 a month, with a 7-day free trial. The path in all your ads is data analyst: SQL at 10 hours, Python at 12 and Dashboards at 14, each with a graded capstone and a certificate. A subscriber fitting that around a job does about 3 hours a week, so 36 hours takes 12 weeks and you collect 3 × $39 = $117. Someone between jobs does 2 hours a day, which is 14 hours per trial, and every course on that path fits inside one. Your best-performing ad says "Learn SQL in 7 days." Your trial lasts 7 days.
That's why online course free trial abuse stings more than it does in other subscriptions. A project tool gets used again next Monday. A course ends, and once the certificate is on someone's profile there's nothing left to pay for.
One learner in this scenario, illustrative:
Those finished courses cost you real money, too. With assumed figures:
- Instructor royalty per completion: $8
- Capstone grading, 20 minutes at $30 an hour: $10
- Streaming and captions, an average 12 hours at $0.05: $0.60
- Cost per completed course: $18.60
- Three courses: 3 × $18.60 = $55.80
Now say 60 people a month do this. That's 180 of your 1,500 monthly trial starts. They wouldn't all have paid, so assume one in four would have if a second trial hadn't been there:
- 60 × $55.80 = $3,348 a month spent on them
- 15 people × $117 = $1,755 a month not collected
- $3,348 + $1,755 = $5,103 a month
- $5,103 × 12 = $61,236 a year
With 900 subscribers you bring in $35,100 a month, and $5,103 is what about 131 of them pay you.
The grading queue is where this reaches the people who pay. Your teaching assistant has 120 hours a month. Subscribers hand in 240 capstones, which is 80 hours. Those 60 trial learners hand in 180 more, another 60 hours. That's 140 hours of work for 120 hours of time, so 20 hours roll over every month, and 20 ÷ 120 × 30 adds 5 days to the wait.
By month 3, a subscriber who has paid you $117 and needs that final grade for a job application waits 15 extra days. On day 15 they cancel. The reason box is empty and nobody writes in. The learner who never paid sent you a thank-you note, and the one who did left without a word.
So now you've got a decision. Say the trial brings in 3 of every 4 new subscribers. Turn it off and you lose the thing that sells the library. Cut it to 3 days and honest learners never reach a capstone, which is the moment they decide you're worth $39. Your best marketing has become the reason people don't need to buy.
Why online course free trial abuse is so easy to repeat
The reward for a second account is a whole new course, and the effort is one more email address. Course progress lives on the account, so starting fresh loses nothing that matters to someone who's skipping ahead to the next subject.
These usually aren't scripts. It's a patient person on a laptop trying to change careers, so tools aimed at bots have nothing to notice. What you'd like to know is whether this learner has already had their free course. The usual first answer is one trial per email address, and "my other email" is what that rule looks like from the other side. It's a form of trial farming, and before you treat all of it as abuse, it's worth asking why people keep restarting in the first place.
Change what one trial can deliver
Start with the offer, because you can change it this week.
Find out whether a trial can finish a course. Divide course hours by trial days times daily study hours. For SQL, 10 ÷ (7 × 2) = 0.71. Anything at 1 or below can be finished in one trial.
Keep the ending for subscribers. Certificates, capstone grading and downloadable workbooks go on the paid plan. In the scenario, that removes 180 × $10 = $1,800 a month in grading, and your queue is back to 80 hours of work for 120 hours of time.
Open the same first lessons in every trial. Say the first 3 hours of up to 2 courses. No new account can reach the last module, and that "Learn SQL in 7 days" ad has to go.
Write down your rule and what a no says. One free trial per learner per year is easy to explain. Then decide what to say when you deny a trial and what to do when someone asks for a second one.
What this does and doesn't change
A smaller trial makes each repeat worth less, but every new account still gets its sample. You may also convert some honest learners less often, so measure before and after.
The basic checks, and how each one breaks
Each of these does catch something. Gmail ignores dots, and plus-tagged mail reaches the same inbox, so one mailbox can look like several addresses. That's just how email works.
| Basic check | What it catches | What it misses | Who it wrongly blocks | What it costs to keep working |
|---|---|---|---|---|
| Email verification | Typos and dead addresses | Any second real inbox | Learners whose code lands in a school spam folder | Sending and deliverability work |
| Unique emails | The same address twice | "My other email" | Almost nobody | Very little |
| Disposable-domain blocks | Known throwaway services | New domains and ordinary mailboxes | People using privacy forwarding | A list someone updates forever |
| Gmail dots and plus addressing | One Gmail inbox written several ways | A second real mailbox | Real learners, if the rule misfires | Upkeep as mail services change |
| CAPTCHA | Simple scripts | A person signing up by hand | Learners using screen readers | Fees and lost signups |
| Phone requirement | Casual second accounts | Anyone with another number | Learners without a reliable mobile number | A charge for every text |
| Card requirement | Casual second accounts | Anyone with another card | Students and people between jobs | Fewer trial starts |
| VPN blocks | Known VPN ranges | Everything off your list | Learners on an employer's VPN | Another list to maintain |
| Rate limits | Bursts from one account | 2 hours of video a day | Almost nobody | Tuning per endpoint |
Four more break in ways that are particular to courses.
A one-trial-per-IP limit does stop someone restarting from the same home connection, and it misses them the moment they use a phone network. Then it lands on your real audience. A college computer lab, a public library and a training room where 14 colleagues start the same course on Monday all share an address, and everyone after the first is told they've had their trial. Your week goes to exceptions and tickets.
A watch cap of 5 hours per trial protects the end of a 10-hour course. A fresh account brings a fresh 5 hours, so the path takes 8 trials instead of 3 (36 ÷ 5 = 7.2, rounded up). It's still $0, while your honest learner hits the cap in the middle of a lesson on a Sunday.
A 3-day trial is 6 hours at 2 hours a day, so no single trial can finish a course on this path anymore. The path just takes 6 trials instead of 3. Anyone who studies after work gets three evenings to judge a $39 library, and fewer of them ever reach a capstone.
Manual review catches what a person can spot, but one minute per trial start is 25 hours a month, and the screen shows a name and an email that tell you nothing. Whoever looks odd to a tired reviewer gets turned away, and nobody writes "my other email" until they're on the way out.
Add them all up and a 19-year-old on campus Wi-Fi faces an email code, a puzzle, a phone number and a card before lesson one. The person who wrote that thank-you note has another email. And none of these can say whether a trial claim looks like a repeat of one from last month.
Evaluate the claim when a trial starts
AbuseZero exists so you can stop maintaining that table. You get multi-layer AI risk analysis that examines 140,000+ signals, detects linked claims and flags abuse that single checks miss, and you also get a proprietary algorithm, with AI that checks the claim again. Email is covered too, because AbuseZero checks mailbox aliases, disposable domains and mail routing. Where an IP limit stops at one address and locks out a computer lab, AbuseZero sees beyond an IP address with VPN, hosting, network reputation, Tor and proxy intelligence. And since a new email address is what beat nearly every row above, it matters that AbuseZero recognizes repeat devices across new accounts, even when email addresses keep changing. Creating a fresh account isn't enough to erase previous activity.
You don't have to build your own abuse prevention system. Right before a trial opens, your server makes one API call and gets a clear allow or deny, under rules you control. Eligibility can be once ever or once every set number of days, so a new claim next year can go through if that's what you want. Observe mode shows you what would be denied before anyone is blocked, and the quickstart walks you through your first benefit.
Frequently asked questions
What is online course free trial abuse?
It's one person using extra accounts to collect more free trials than you meant to give, usually to finish a course or a whole path without paying.
Is every second trial abuse?
No. Someone might come back two years later for a different subject, or use a work account their employer set up. Decide how often a trial can be claimed and write it down.
Should I just remove the free trial?
Leave that until last, because it costs you the honest learners too. See whether to remove a free tier. AbuseZero evaluates each claim when a trial starts, so your trial length can stay a teaching decision.
Keep your trial for new learners
Start with Builder
Pay as you go. Protect your first trial today.
See pricing
Builder, Pro, Scale and Enterprise.
Talk to us about Enterprise
High volumes and custom terms.