Voter fraud via Hearken is rare. We don’t think it’s a problem if a really enthusiastic person votes more than once. (If they’re determined, they can vote on different devices, or open an incognito window to vote multiple times.) We don’t consider that fraud. It’s just super-engagement!

With that being said, true voter fraud can happen on your Hearken embed and we consider that problematic. For example, someone could write a script that would allow them to vote hundreds of times and significantly alter the results of your vote.

Most of you probably don’t have to worry about this kind of interference. But just in case, we have a system that will alert you when there is potentially fraudulent behavior on your live voting round.

The fraud detector operates by tracking the IP addresses of your voters. Every few minutes, the system will scan for fraud.

If the system detects potential fraud, the Engagement Management System (EMS) will send an email to the person who created the voting round, as well as to the Hearken support email address. You can also tell us if there is anyone else who should receive that email by going to your Platform Settings page and checking the boxes for those people under " Notify Users on Potentially Fraudulent Votes."

You will be alerted as soon as the number of fraudulent votes exceeds a threshold (right now, that threshold is 20 votes in 24 hours). If more bad votes for the same question come in from the same source, they'll be rolled in to the same fraud event but you won't be alerted again.

Once you get that email notification from our system, you’ll be able to review your voting activity in the EMS.

The voting round in question will have a small banner above it that reads "We've detected possible fraudulent activity in this voting round." and a button labeled "Investigate."

Once you click "Investigate," you will see a list of any and all potentially fraudulent activity during that voting round. You'll be able to see the date the votes came in, how many suspected fraudulent votes came in, and which question the votes went to. Once you consider that information (like, did all of the votes go to a single question, or were there votes going to multiple questions? Was there suspicious voting activity on more than one day?), you can decide whether you want to delete the votes (by clicking the red x) or mark them as OK (clicking the green checkmark).

To see what votes have been deleted or marked OK in past voting rounds, you can go to your Voting Fraud Activity History page, which you can find a link to either from your voting fraud investigation page, or from your Platform Settings page.

If your organization is experiencing voter fraud attempts every day during a voting round, or every time you have a voting round, you can consider telling the system to automatically delete any votes above the fraudulent threshold. You can do so by navigating to your Platform Settings page, and then check the box labelled " Auto Delete Fraudulent Votes."

You can review what votes were deleted automatically by visiting the Voting Fraud Activity History page.

You may be wondering why we think it's ok for one IP address to have multiple votes per day. The answer is that one IP address doesn't necessarily equal one person. For example, it's possible for one whole office building to have a single IP address. We don't want to create a system that would keep everyone in that office building from voting!