You just kicked off this great campaign and got everyone excited about how much exposure and traffic it is going to generate.

Your campaign asks customers, prospects and friends to come in and share and vote for their favourite video, image or other content that helps promote your brand or cause. 

This is going to be great for your company and at the same time elevate your position as an innovative thinker.

All seems to be going great until you notice that the votes you are receiving are really starting to spike. While at first you have a feeling of elation that quickly turns to panic as you see some unusual behaviour with the voting. You then realise your campaign is being targeted by fraudsters trying to elevate the position of specific entries for either personal gain or just for purely malicious intentions.

So what do you do, or more importantly how do I stop or at least minimise this happening in the first place?

Well, first things first. Unfortunately, you are never going to be able to eliminate the risk of fraudsters submitting rubbish votes. However, there are steps you can take to stop them from ruining your campaign.

Here are some tips which we have found to help deter or reduce fraudulent votes from happening.

Make it clear that fraud votes will be removed

Always make it clear from the outset that you will be monitoring votes coming in and that votes that are suspected to be fraudulent may at your sole discretion be removed and that disqualification of the entry can result.

Don’t set public votes as the sole winning selection criteria

This is a big one. If you are providing a big prize to your winner(s) you will always want to have a final say in who should win the prize. For this reason we always recommend that public voting should be only one of the winner(s) selection criteria and not the sole one. Other engagement metrics can be used such as views, plays, shares, etc. Also consider setting up internal judging panels which will also play a part in selecting the winner(s).

By making this obvious on your contest site will discourage fraudsters from putting in dummy votes if it is not going to have a significant impact on the result.

Run multiple rounds

The selection of winners typically follows two patterns:

There are pros and cons to each option. 

The first option is great when you want maximum engagement. This is the best option when your Contest is focused on brand or social awareness as it promotes sharing and promoting of all entries. It also still provides a good safeguard through the judging panel. You can also include in your terms and conditions that wildcard entries can also be considered in the judging rounds.

The second option is more suited to contests where content reuse is the primary concern. This allows you to filter down to the best content and then use the popular vote to determine the best one. 

Use email voting gateway

By using an email gateway for voting means that voters will need to enter their email address to cast a vote. ContestPad validates the email by sending the voter a validation token key for them to validate the email address and only after doing this is the vote counted. While this is no guarantee that the voter doesn’t have multiple email addresses it does limit it somewhat. Note that if the voter is trying to use temporary email services or you suspect a fraudulent domain you can block these (more later on this). Also make sure that you configure the vote validation email template to include copy that requests the recipient to not select the link if the vote was not cast by them. You can also include an email address to report any potential fraudulent emails. 

Use social voting gateway

Another option for validating the identity of voters is to use the social voting gateway. This requires the voter to use their social network account to vote in the contest. This puts the reliance on validating voter identity on the social platform. Facebook is one of the better platforms for this as they go through greater lengths to validate the identity of account holders than say Twitter. Having said this it is important to note that Facebook still has up to 5% to 6% fake accounts.

Use IP Address validation

Another option is to leverage the IP address of the person’s device to validate their identity. This is the voting method that has the lowest barrier to voting. While this doesn’t really identify a user’s identity it does assume that each device typically belongs to one person. This can come unstuck if a person has multiple devices (unfortunately many click farms do thisL). Also, if voting is being conducted behind a corporate or institutional firewall sometimes the end users IP address is hidden. This means that the IP address for every device coming from this intranet will look the same. This will result in possibly being locked out from legitimate voting. 

Monitor voting

As voting commences and even after voting is complete, there is no substitute to monitoring your vote data. You can export out all votes in a CSV format so that you can even do more detailed analysis using Excel or some other analysis tool. Look for patterns such as:

Block or remove suspect votes

Once you have found suspect votes you should consider invalidating these votes and more importantly block them from coming in again. If you identify IP addresses, individual email addresses or even whole email domains (e.g. that belong to a suspect corporation or temporary email service provider) then you can block these in the global settings of ContestPad. ContestPad already has an internal blacklist of email domains and blocks email votes coming from these domains (these mainly include temporary email service domains). However, there are new domains popping up all the time so if you find some we haven’t blacklisted let us know so we can continue to update our internal list.

To show or not to show vote metrics

Another thing to consider is whether to show to the public how many votes each entry has received. Obviously, this can incentivise honest voters or entrants to get more votes as they see their vote counts climb. Unfortunately, it can also incentivise some people to do the wrong thing as they directly see the results of their fraudulent behaviour. Also, once votes are displayed it can make for an awkward discussion if votes are removed as the entrant may claim no knowledge of vote fraud. An option is to state the votes counts displayed are indicative and are subject to change upon internal validation.

Protect yourself legally

The final line of protection is ensuring that the legal terms and conditions for your contest (and which all participants agree to when entering the contest) covers your rights related to some of the following items:

Also remember to ensure that the above and any other conditions are legally enforceable in the jurisdiction of your contest (and make sure you state what your jurisdiction is). Make sure you enrol the help of legal professionals to take care of this for you.

While there is no way to 100% guarantee that voters will not do the wrong thing, if you follow many of the steps above you minimise the risk that potential fraud issues bring your campaign into question.