To accomplish our goal of getting a target audience to take this survey, we first determined our buyer persona.

Based on the pictures in the survey, we want to target young and middle-aged women. We settled on the age range of 18 to 50. The target audience includes those with an interest in fashion. We approximate that at least one million women fall within this demographic (aged 18 to 50 and interested in fashion). According to the SurveyMonkey sample size calculator, this would indicate that we need 1066 people (this number only increases to 1068 if our population was as large as 1 billion). We also know that political polls in the US, which are meant to summarize the attitudes of 200 million Americans, generally only have about 1000 people in the sample [1]. To have a confidence level of 95% with a margin of error of +/- 3% (which is standard for polling), we would only need approximately 1000 people in the sample.

The best places to target those interested in fashion are places where women consume fashion: Instagram and Pinterest. Instagram and Pinterest is a ubiquitous marketing platform where people go to look at the fashion/lifestyle of others, among other things [2]. If our brand was big enough and had many followers/accumulated many ‘likes’ on our images, we would post an image where a model was wearing one of the outfits and leave the survey in the caption.

Given that we want to attract those who actually care about fashion, we would incentivize people to take the survey by offering them any outfit of their choice from the company (while retail cost probably exceeds $100, manufacturing costs likely do not). In the caption, we would post a message explaining the incentives for taking the survey and why it is important (e.g. “Make your voice heard in shaping this year’s fashion trends! Take this quick survey for a chance to win any outfit of your choice”)

This post from Everlane is a good example of what we envision. The brand has over 300k followers and regularly gets 2,500 to 7,000 likes per post [3], so we think we could engage 1000 survey takers within a week (if we make this a week long campaign, with a new outfit post every day or every other day). Duplicate posts could be shared via any other social media channels this brand already has (e.g. email newsletters, Facebook, Twitter, etc). We considered potentially reaching out to social media influencers, but upon discovering the costs (which can go as high as six figures), this seemed impractical for the given constraints.

If we were to implement this survey at Harvard, we would email all Harvard upperclassmen houses and first year dorms listservs. In this email, we would incentivize engagement with the survey by emphasizing the opportunity to win a free ticket to IDENTITIES/ELEGANZA shows (budget: 10 IDENTITIES/ELEGANZA tickets @ $10 each). Since students who attend these events have some appreciation for fashion, this would engage a relevant audience for a survey that is determining desirability in fashion.

Survey modifications:

  • Important logistical element if we want to increase motivation: Landing page should ask for email address to prevent duplicate entries and contact info (for prize collection)
  • Decreases ability: Some questions are inconsistent (always put “Like” on left and “Dislike” on right)
  • Decreases motivation and ability: Asks many duplicate questions, which may be intentional, but makes the survey longer and more boring
  • Decreases ability: Instructions aren’t always clear (select one outfit? select all that apply? click? drag?)