How to stop multiple entries in a SaaS giveaway
One entrant with hundreds of accounts can take the prizes you meant for your audience. What SaaS giveaway multiple entries cost, and how to run a fairer draw.
Your anniversary giveaway closed with 7,400 entries, almost three times the 2,500 you'd have been happy with. You run an app that helps independent podcasters edit and publish their shows, and you've spent 14 months trying to get in front of exactly these people. On Tuesday you draw 30 winners. Your script emails each of them a $300 voucher for recording gear and puts a year of your Studio plan, $348 at list price, on their account. You announce it and go to bed pleased.
On Wednesday morning, winner number 17 writes to support: "Hey, three of my accounts won. Can you move all the Studio years onto my main one?"
It's only an example, but the sums work the same with your numbers. You'd planned to feature the winners in your newsletter anyway, so you ask all 30 for a link to their show. Eighteen send one. Twelve have no show and have never uploaded a second of audio. That's SaaS giveaway multiple entries, and the draw has already sized the problem for you, because winners are a random sample of entries:
Bad winners: 12 of 30 = 40%
Duplicate entries: 40% of 7,400 = about 2,960
Real entries: 7,400 - 2,960 = 4,440
Odds per entry: 30 / 7,400 = 1 in 247Thirty names is a small sample, so treat that as a rough figure. At 1 in 247, three wins takes something like 740 accounts, which would make winner number 17 about a tenth of your whole giveaway.
Now add up what those 12 prizes cost. Assume a Studio year costs you $96 to serve ($8 a month in transcription and storage), and the demo episode every new account gets costs $0.35 to render:
- Vouchers: 12 × $300 = $3,600
- Studio years: 12 × $96 = $1,152
- Demo renders: 2,960 × $0.35 = $1,036
- Total: $5,788
On $11,000 of monthly revenue, that's more than half a month. By lunchtime your voucher dashboard shows nine of the 12 already redeemed, so most of that cash isn't coming back.
Those renders also shared a queue with your paying customers. Say 640 of the extra accounts arrived in the last hour before entries closed at midnight, each demo takes 5 minutes, and you run 8 render workers. That's 3,200 minutes of rendering against the 480 your workers clear in an hour. The 2,720 minutes left over are 340 minutes of waiting (2,720 ÷ 8). A customer exports Thursday's episode at midnight for a 5 a.m. release, and it comes out at 5:40.
The 12 prizes hurt most, because they had owners. They belonged to podcasters among the 4,440 real entries you built this for. Without the extra accounts, an entry's odds were 1 in 148 (30 ÷ 4,440). One host told their listeners about your giveaway on air, and now they're in your comments asking why so few of the winners seem to have a podcast.
You've got three options and you don't like any of them. Say nothing and hope nobody compares notes. Pull the 12 Studio years and explain why, knowing one of those accounts could be a real host who hasn't launched yet. Or draw 12 new winners and pay for those prizes twice: 12 × ($300 + $96) = $4,752, which takes the damage to $10,540, nearly a full month's revenue.
Every duplicate entry draws on the same 30 prizes. With 4,440 real entries, real entrants keep 30 × 4,440 ÷ (4,440 + duplicates):
Why SaaS giveaways attract multiple entries
An extra account on a free trial gets someone more of your product. An extra account in a giveaway gets another ticket, and the prize is worth the same to someone who has never heard a podcast. Divide the pool by the entries to see what a ticket pays:
- $9,000 in vouchers ÷ 7,400 entries = $1.22 an entry
- If an extra account takes a minute: $1.22 × 60 = about $73 an hour
Nobody has to care about your product to find $73 an hour interesting. And most other abuse at least leaves you something to cut off, since an account burning free credits can be suspended halfway through. A prize goes out once, in one batch, to the accounts you picked, so the damage is done on the day you reward them.
The obvious fix is a rule: one entry per person. Write it, because a plain policy is what gives you the right to disqualify anyone. But your form can't see a person. It sees an email address and whatever else you ask for, so the rule only reaches as far as those basics do.
The basic fixes for multiple entries, and how each one breaks
Each of these catches something. Email is where the limits are plainest: Gmail ignores dots, and plus-tagged mail reaches the same inbox, so a unique-email rule can count one mailbox several times.
| Basic fix | What it catches | What it misses | Who it wrongly blocks | What it costs to keep working |
|---|---|---|---|---|
| Email verification | Typos and dead addresses | Any real second inbox | Hosts whose code lands in spam | Deliverability work |
| Unique emails | The same address twice | Every different address | Almost nobody | Very little |
| Disposable-domain blocks | Known throwaway services | New domains, 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 | Almost nobody | Upkeep as mail services change |
| CAPTCHA | Simple scripts | People entering by hand | Screen reader users | Fees and lost entries |
| Phone requirement | Casual extra accounts | Anyone with another number | Hosts without a reliable mobile number | A charge per text |
| Card requirement | Casual extra accounts | Anyone with another card | Entrants who won't pay to enter | Far fewer entries |
| One entry per IP | Repeats on one connection | Any other network | Co-hosts, households, meetups | Exceptions and angry replies |
| VPN blocks | Known VPN ranges | Everything off your list | Privacy-minded listeners | Another list to maintain |
| Rate limits on the form | Bursts from one source | Entries spread over three weeks | People entering in the final hour | Tuning |
| A shorter entry window | Slow, patient entry building | A script that needs one evening | Anyone who heard about it late | A smaller audience |
| Reviewing every entry | What a person can spot | Whatever arrives faster than you read | Whoever looks odd to a tired reviewer | About 123 hours at one minute each |
Take the disposable-domain block. In week one it stops a batch of throwaway addresses. By week two the entries use a domain that isn't on your list, and a host with a privacy forwarding address can't enter and says so in public. One entry per IP is worse for this audience, because co-hosts record on one connection and everyone at a podcasting meetup shares the same Wi-Fi. CAPTCHA has its own limits, and stacking all of these leaves you with a form that asks for a verified email, a puzzle, a phone number and a card. The people who leave are the hosts you ran the giveaway to meet.
How to run a giveaway that's harder to farm
None of those rules protects the prize, so protect the prize directly.
Write the rules first. One entry per person, what counts as a person (decide about co-hosts now), and that prizes are held until winners are confirmed. Giveaway rules vary by place, so have someone qualified read yours.
Pick prizes only your audience wants. A $300 voucher is worth $300 to anyone. A year of Studio is worth very little to someone with no show. Drop the vouchers and the cash value of an extra ticket falls from $1.22 to nothing, though real winners may be less excited.
Hold expensive work until the entry means something. Render the demo after a first upload, and those 2,960 renders never start.
Draw alternates in the same draw. If 40% of winners might not hold up, 30 prizes need 30 ÷ 0.6 = 50 names.
Confirm before anything goes out. Ten minutes a name across 50 names is about 8 hours, which is worth it for $19,440 of prizes at list price (30 × $648).
Decide what a no looks like. Some people you turn down will be real, so say why and give them a way to reply.
That's the best you can do by hand, and it has limits. A link proves a show exists, not that the person who pasted it runs it. Your review sees 50 names out of 7,400, so the other 2,900-odd duplicates stay in your email list and your conversion numbers, ready for the next giveaway. And it eats the week you'd hoped to spend welcoming new customers.
Deciding whether an entry should count at all
Every row in that table stops at the same place. It can judge one field on your form, and your form has no memory of the 700 entries before this one. What you need is an answer about the claim: does it look like a repeat, and should it go into the draw? You don't have to build your own abuse prevention system to get that.
AbuseZero is built for that moment. It puts 140,000+ signals through multi-layer AI risk analysis, which detects linked claims and flags abuse that single checks miss. The disposable list that went stale in week two? AbuseZero checks mailbox aliases, disposable domains and mail routing. The IP rule that caught two co-hosts? AbuseZero sees beyond an IP address with VPN, hosting, network reputation, Tor and proxy intelligence. It recognizes repeat devices across new accounts, even when email addresses keep changing, and it blocks spammy bots, browser automation and malicious scripts while allowing genuine AI agents and assistants. We're proudest of the proprietary algorithm, with AI that checks the claim again. Creating a fresh account isn't enough to erase previous activity.
You create a benefit for the giveaway, set its eligibility to once, and make one API call before an entry counts. You get a clear allow or deny, under rules you control. Observe mode shows what would be denied before anyone is blocked, and the quickstart gets your first benefit running. Keep the prize hold, too. It costs little, and it's your last look before anything leaves.
Frequently asked questions
How do I stop someone entering a giveaway twice?
Allow one entry per account, verify emails, hold every prize until the winner is confirmed, and draw alternates. To evaluate each entry as a claim before it reaches the draw, AbuseZero returns allow or deny.
Can I disqualify a winner after the draw?
If your published rules allow it and you have a reason you can stand behind, usually yes. Keep what you found.
Do co-hosts count as two people?
That's your call, so make it before launch and put it in the rules. Counting a show as one entrant is simple and easy to explain.
Give the prize to the audience you ran it for
Start with Builder
Pay as you go. Protect your first benefit today.
See pricing
Builder, Pro, Scale and Enterprise.
Talk to us about Enterprise
High volumes and custom terms.