How to stop duplicate accounts on a free product analytics plan
Hit a free event limit and a second account beats upgrading. What split traffic costs an analytics product, and which basic fixes break.
You were pleased with the hard event limit when you shipped it.
Your product analytics tool gives every free account 1 million events a month, and for a long time that number was only a polite email. Forty teams ignored the email and kept sending about 15 million events each. So you wrote the enforcement yourself. At 1,000,000 events the ingest endpoint stops accepting, and the dashboard offers your $249 plan, which covers 20 million. You merged it, watched it fire in production and worked out what it should bring in: 40 teams × $249 = $9,960 a month.
This is a scenario, and every cost in it is an assumption to replace with your own.
For a week it looked perfect. Then you noticed that none of the 40 had upgraded, and that free signups were climbing. By week six the pipeline was back where it started, and the over-limit report you'd been so pleased to see go quiet still said zero. Each of those teams now runs 15 free accounts at 1 million events apiece, so 40 × 15 = 600 accounts, all under the limit. That's the analytics free plan multiple accounts problem, and no report built on a per-account limit will ever show it.
On a Tuesday at 11:20 a.m., one of your 45 paying teams writes in: "Our launch dashboard is 23 minutes behind. We're moving ad budget off these numbers. What's going on?"
Your pipeline clears 1,500 events a second, which was plenty when paying traffic peaked at 1,000. The 600 accounts send 600 million events a month. Over 2,592,000 seconds that averages about 231.5 a second, and say it runs three times higher at peak, about 694.
- 1,000 + 694 = 1,694 events a second arriving during a three-hour peak
- 1,694 minus 1,500 = 194 a second that can't be processed
- 194 × 10,800 seconds = 2,095,200 events waiting when the peak ends
- 2,095,200 ÷ 1,500 = about 1,397 seconds, or 23 minutes behind
Every paying team shares that line, so every paying team's charts are late. You credit that customer's month and open the bill. Assume $4 per million events to ingest and process, $0.20 per million events per month to store, and the 12 months of history your free plan promises.
- Ingest: 600 million × $4 = $2,400 a month
- Storage: 600 × $0.20 = $120 more every month, because each month's events stay
- Year one: $28,800 ingest + $9,360 storage ($120 × 78, the sum of months 1 to 12) = $38,160
Your 45 paying teams bring in $11,205 a month, or $134,460 a year. About 28 cents of every dollar they pay now covers events from teams who decided not to pay. On $30,000 of monthly spend, that's nearly 1.3 months of runway. The $119,520 a year you'd expected from those upgrades isn't coming either.
And it doesn't stop when they do. Say all 600 accounts go quiet on the last day of month 12. You promised 12 months of history, so those events stay on your disks and expire one month at a time: $120 × (11 + 10 + ... + 1) = $120 × 66 = $7,920 to keep data nobody will ever query.
So you're choosing between bad options. You can pay for more pipeline so unpaid traffic keeps moving. You can remove the limit you were proud of. Or you can shrink the free plan that brought you most of those 45 teams, knowing that if three of them leave over a lagging dashboard, that's another $8,964 a year (3 × $249 × 12).
Why a free event limit creates more accounts
An event limit counts what happens inside one account, and a new account starts at zero. Once a team hits the wall, the choice is $249 a month or a few minutes and one more email address. The second account comes with its own project key, so the web app can report to one account and the mobile app to another. The team loses funnels that cross the two. Plenty of teams can live with that.
Analytics hides this better than most products do. A small real site and a slice of a bigger one can both look like 900,000 ordinary events. Those events come from your customers' visitors, so they say a lot about the visitors and very little about who opened the account.
The obvious fix is a lower limit, so try it on this scenario. Cut the free plan from 1 million events to 250,000, and 15 million events takes 60 accounts instead of 15.
- 40 teams × 60 accounts = 2,400 accounts
- 2,400 × 250,000 events = 600 million events
That's the same pipeline load and the same $2,400 a month, and every honest free user just lost 75% of their plan. See how to set free plan limits that don't invite extra accounts.
Basic ways to limit free analytics accounts, and how they break
Each of these catches something. Three work differently in analytics.
One signup per IP address catches someone signing up again from the same connection. It misses anyone who moves to another network, and it turns away real developers sharing an office or a coworking space, so you end up maintaining exceptions. You can't move it to the ingest side either, because one project key legitimately receives events from thousands of addresses.
Rate limits and usage caps stop a burst from one key and stop one account overrunning its plan. Keep them. They can't see 15 accounts that each stay under the line, and a tight limit drops events on a real customer's best day, which is data they never get back. Someone also has to keep tuning them. Here's what rate limits do and don't stop.
Shorter free history does cut storage. At 90 days, you'd hold 3 × 600 million events for $360 a month, against $1,440 by month 12. Ingest stays at $2,400, and honest users lose their year-on-year view.
Email is where most teams start, and it's weakest with this audience. Gmail ignores dots and plus-tagged mail reaches the same inbox, so one mailbox can pass as many addresses. Developers often own a domain as well, and a domain can accept mail at any address.
| Basic fix | Catches | Misses | Wrongly blocks | Upkeep |
|---|---|---|---|---|
| Email verification | Typos and dead addresses | Every second inbox that works | Developers whose code goes to spam | Deliverability work |
| Unique emails | One address used twice | Any other address | Almost nobody | Very little |
| Disposable-domain blocks | Throwaway services on your list | New domains, normal mailboxes | Privacy forwarding users | Endless list updates |
| Gmail dots and plus tags | One Gmail inbox written several ways | A second real mailbox | People who share one inbox | Upkeep as mail services change |
| CAPTCHA | Simple scripts | Signups done by hand | Screen reader users | Fees and lost signups |
| Phone requirement | Casual second accounts | A second number | Developers without a reliable mobile number | A charge per text |
| Card requirement | Casual second accounts | A second card | Developers without the company card | Fewer installs |
| VPN blocks | Ranges on your list | Everything else | Developers on company VPNs | Another list |
| Manual review | What a person can spot | Slices that look like small sites | Whoever looks odd that day | 10 hours for 600 accounts, at 1 minute each |
Free analytics wins on a snippet pasted in two minutes, so every step here is a reason to paste someone else's. Card requirements and VPN blocks carry costs of their own. Signup steps also do nothing about the 600 accounts you already have.
You'd really like to know whether the person behind account 14 also runs account 1. Your own tables can't tell you. Two accounts from one team can look unrelated, and two strangers at the same coworking desk can look identical.
What to change on your free analytics plan
Do these whatever else you decide.
Write down what one free plan means. One per company? One per client for agencies? Put it in plain words so support and your limits agree.
Give paying traffic its own lane. Use a separate ingest queue, and delay free events first. In this scenario, paying traffic peaks at 1,000 a second on a 1,500 pipeline, so all the lag moves onto the free accounts.
Count events across the whole account. A new project shouldn't reset the meter. Where the limit sits decides what a reset costs.
Stop paying for abandoned history. Stop keeping events for free accounts nobody has opened in 60 days, and tell people before you do.
Put a plan between $0 and $249. A team at 2 million events has no reason to pay for 20 million. Give them something smaller, and a second account stops being the only sensible move.
Those steps make a free account cheaper to carry and a paid plan easier to choose. They leave one gap. When a new account starts a free plan, nothing in your stack says whether that claim should go through.
Evaluate the claim when the free plan starts
You don't have to build your own abuse prevention system to close that gap. Instead of a domain list you update forever and VPN ranges you keep guessing at, AbuseZero checks mailbox aliases, disposable domains and mail routing, and it sees beyond an IP address with VPN, hosting, network reputation, Tor and proxy intelligence. Every row in that table is a single check. AbuseZero's multi-layer AI risk analysis examines 140,000+ signals, detects linked claims and flags abuse that single checks miss. AbuseZero recognizes repeat devices across new accounts, even when email addresses keep changing. There's a proprietary algorithm too, with AI that checks the claim again. A per-account event limit starts every new account at zero, but with AbuseZero, creating a fresh account isn't enough to erase previous activity.
You make one API call when someone starts a free plan and get a clear allow or deny, under eligibility rules you control. Observe mode shows you what would be denied before a single agency or side project is blocked, and the quickstart walks you through your first benefit.
Frequently asked questions
Is it abuse if one company has several free analytics accounts?
Not always. An agency with one account per client, or a developer with a work account and a side project, may be using your product exactly as you'd want. Your written policy decides.
Will cutting the free event limit stop people opening more accounts?
No. It lowers what each account costs you and what each honest user gets, and it raises the number of accounts someone needs. Freemium plans run into the same arithmetic.
Does an abuse prevention API replace my event limits?
No, keep them. AbuseZero's evaluation covers the moment your free plan is claimed. Your app still enforces limits and retention after that.
Keep your free plan for real projects
Start with Builder
Pay as you go. Protect your free plan today.
See pricing
Builder, Pro, Scale and Enterprise.
Talk to us about Enterprise
High volumes and custom terms.