Issues Regarding Public Voting in Design Competitions

Like many designers, I have friends who regularly take part in design awards and competitions, and one thing that makes me really sad is to see people begging others for votes – to win a design award. It happens to me many times that I receive an email that says, please vote for my design; it has been "shortlisted" and is open for "public voting".

Experience: Recently, I received an email from a friend who was "shortlisted" in the Some Decoration Magazine's Design Awards, he is a great designer and he does great furniture designs, but still he needs me to vote for him. For him, he says, my vote would help him win the grand award (which is true), but in reality, the "viral" scheme exists so that the award program webpage can get more hits.

I did vote. I did see this Decoration Magazine Design Awards' webpage, I created an impression for the this Decoration Magazine, and I am now receiving their newsletters because I had to register to vote. But, guess what, my respect for this Decoration Magazine's Design Awards just decreased to zero, from a point of view of a designer, because I believe a good design award should not send their participants to beg for votes.

Because if you send participants to beg for votes, and assign winning status based on such votes, then I can naturally question the award and the winner: did the winner design win because it was good design, or because the winner had more friends than the other designers? Is it an award about popularity or design quality? If the awards are given based on the number of votes, then I would question the authority and credibility of this award.

Experiment: In the past we have created and run custom design competitions and contests for companies that wanted to maximize visibility, thus we decided to incorporate public voting, for the very purpose of having "more visibility" i.e. new visitors for the webpage who would "check" what the buzz is about. Our initial aim was to have more people know about the competition, and to do so, we "shortlisted" some entries and asked participants to invite their friends to vote for these "shortlisted works".

It was very successful from a marketing point of view, within a single day after we go for public voting, we had tens of thousands people pouring to the website. If you don't know what it means economically, it translates to thousands of euros worth of CPC advertising. However, it was a failure for us, because our aim was to let more people "know about the competition", our aim was to "create interaction", and "added visibility for other entries" not some dummy "hits".

A common thing we noticed first, was that people did not voted for others' works, let alone check them, they rather voted for who sent them the invitation to vote in the first place. This was important for us, because we wanted the voting to be fair, which turned out to be a popularity contest, it was one of the primary goals of the organizer to run a fair competition and the competition did just became unfair.

We did an analysis on the votes. Given a data sample size of 22.300 entries as a basis for consideration and interactions saved in the analytics database we have discovered that 92% of the time people vote for their friends' designs only, they did not even check others works (92% is calculated by considering the initial 22.300 people, out of which 20.500 i.e. 92% dropping / leaving the website, after voting for their friends, without interacting any further).

Now, to argue in favor of the afore mentioned Famous Design Awards, you would say, "having public voting is good, as it allows and creates opportunity for others' designs to be checked", and my response is this: With great statistical significance, we can say that it is "scientifically proven" that invited voters from one designer, do not check other designers works for most of the time, unless, you improve the page.

Therefore, we changed the voting page, in which we put all the designs together, and from what I see, this worked better, as at least, those "begged - invited" visitors, can now see also what others also did. This helped us to decrease the bias for beg-voting, but still, scientifically speaking, more than 85% of the votes would still go to the person who invited their friends to vote. If you let people vote multiple times, this value will decrease further, going to around 75%.

Therefore, we have several conclusions, first of all, design awards that ask the participants to "beg for votes" should not be considered awards, they are "sweepstakes" i.e. advertising for the organizer. Second, public voting, decreases the prestige of an award or competition, thus should be avoided where possible. Third, all public voting does it to create "new hits" on your website, but these "hits" are useless as the new visitors do not interact, they just click "vote" and leave (unless, they are forced to register newsletters etc, which is a terrible practice").

Finally, it is sad but true that "public voting" can create significant viral traffic, increasing the website traffic by hundreds of folds, for a brief amount of time, and is good, if your aim is to advertise more. If run in a carefully and planned manner by incorporating the following principles, the system will work better: 1. All designs in one page (and vote for the best slogan should be incorporated), 2. Require minimum number of votes from each visitor, 3. limit vote of a single user technically to avoid cheating, 4. Publicly advertise the voting page, so it is not just friends of friends who are voting. 5. Hide the names of people, so you do not know which design belongs to who. (in any ways some of the designers will cheat by sending a photo of their designs..). 6. Do not compare apples and bananas (explained below).

Supporting evidence for the six requirements stated above, do not come from design issues, as designers rarely discuss it, so we get these insights from non-design fields, such as the public science; i.e. voting theories and economics. The conclusions above are based on the insights below, I just preferred to arrange this article in this way to make it easier to read, and to sparkle questions about why that would be the case, here is why:

1. All designs should be in one page together with vote-for the best slogan: because this can be considered "voter education", which has been proven in political science studies to decrease the effects of "vote buying" (or vote begging in our case) which results in a more competitive, fair competition where better designs get more votes. [Vincente, 296]. This is further supported by our field experience where we see a more "normal standard distribution" when the voters are able to see alternative designs to vote for.

2. Require a minimum number of votes for each visitor: Simple reasoning, because if you let the voters just pick one, they will just pick the design of their friends, and you cannot catch their "real preferences", to catch the real preferences, you must force the votes to vote at least two designs, where the second design could be considered either the "best" in the competition or the "second best" if his friends' design is good. It is true that some people will choose some random designs, but they would vote the best they like. By doing so, we would be utilizing partial Condorcet voting and Borda Count methodologies which are ways to choose candidates in sports events. [Johnson, 15]

3. Limit vote of a single user to avoid cheating – field experience and observation. If you let people vote as many times as they want, the designer could easily vote for himself hundreds of times, and some could even hire clickers to click as many times as required. On the other hand, scientifically-speaking, consider the political elections, why do you think that people are marked or registered (be it digitally or manually) after voting, that is because they will cheat in case there is not a mechanism to prevent it. For internet based voting, there are many papers written on the subject, all to prevent that votes are cast fairly, especially when a big prize is involved, security will be essential. [Parakh, 8].

4. Publicly advertise the page to make sure that a wide audience votes, this is of course to decrease any biases that are created by vote-beggers' friends voting on their friends' designs; a person, who comes up to the page without a participants' invitation to vote for a particular design will vote sincerely. The base concept is based on "voter turnout", i.e. number of stakeholders that cast their votes would increase fairness of the voting scheme. [Birch, 3]

5. Hide the names of people, so you do now know which design belongs to who: This is especially important to reduce the tendency of designers' friends to vote for them, and also to decrease voting biases towards other famous designers who are involved. Some competitions even list CV's or number of awards these designers won, and this create a bias on voters that should be decreased to increase the quality of voting process. [Taylor & Yildirim, 4]

6. Do not compare bananas and apples, this is perhaps the worst thing, as you see design competitions where you could compare (vote for) produced works or concepts, furniture or vehicles, graphics and interiors etc, this simply does not work, as bananas and apples are different. If public voting would be used, this should only be in a niche competition, or the voting must be category-specific.

Democratic considerations and future work: One could and perhaps should ask the question, shall we allow "random" population to vote for designs? From one point of view, as economic viability of a project, the random i.e. general population's preferences could be considered sound; as they are the customers who would buy the voted designs, but we should understand that the votes of the general population would be based on mostly several elements: 1. The presentation of the work, 2. Aesthetic appeal of the work, 3. The conceived value of the work (which should be removed by not mentioning the designers name). Thus, opening an award into public voting, turns a design competition into a "beauty pageant" that mainly focuses on the physical beauty of the participating works, while jury voting ensures that in-depth design thinking such as engineering, sustainability or ergonomics are considered and accounted for.

• Vicente P.C., Wantchekon L., "Clientelism and vote buying: lessons from field experiments in African elections" Oxford Review of Economic Policy, Vol 25, Issue 2, Pages 292-305, 2009.
• Taylor, C.R., Yildirim, H., "Public Information and Electoral Bias", Department of Economics, Duke University, Working Paper, Pages 1-23, 2009.
• Johnson, P.E., "Voting Systems", Department of Mathematics, Kansas University, Partial Book & Essays, Pages 1-71, 2005.
• Birch, S. "Perceptions of Electoral Fairness and Voter Turnout", Department of Government, University of Essex, Working Paper 3 on Electoral Malpractice in New and Semi-Democracies", pages, 1-44, 2010.
• Parakh, A., Kak, S., "Internet Voting Protocol Based on Improved Implicit Security", Cryptologia Journal, Vol 34, Issue 3, 2010, Pages 258-268, doi: 10.1080/01611194.2010.485421.