Social Recruiting For Research: How To Reach Unreachable Audiences

ARTICLE
BY AMY NGUYEN, MINJUNG KOO

As researchers, we rely on the accuracy of the data we collect to provide our clients with the insights they need and, while there are a number of factors that can impact the accuracy of data, how you sample a population is often the most important factor. Sometimes though, finding the respondents you need to reach for your particular study can be easier said than done.

At Environics Research, we’re often challenged by our clients to find niche audiences and, over the years, we’ve employed a variety of methods to reach them. We’ve used tools like Environics Analytics’ PRIZM™ segmentation, for example, to pinpoint where hard-to-reach respondents live, and built up partnerships with companies that hold specialized connections to niche audiences to help us when we need them. Ultimately, the approach we take to sampling a population is dependent on the accessibility of that population via traditional methods.

Recently, as part of our annual Canadian Newcomers study, we wanted to top up our national sample of Canadian newcomers with a more robust sample of Filipino Canadians living in the Prairies. There were a number of reasons why this activity proved challenging, but the main issue was that recent immigrants to Canada (those who have immigrated in the past 10 years) are not well-represented in online panels—particularly those who are not living in major urban centres such as the Toronto and Vancouver. Ultimately, this meant we needed to find a more innovative approach to recruitment.

We knew that this newcomer audience was staying connected to their families and friends back home, so why not connect with them the same way they connect with their networks—through social media. By utilizing Facebook’s pay per click advertising to reach our target, we hoped to reach what our partner panels were calling an “unreachable audience.”

Qualifying our audience with our newly formed recruitment strategy relied on three basic pillars: social targeting, the advertisement content and the survey itself.

 

Social Targeting

Facebook’s pay per click advertising provides a wide range of targeting possibilities, spanning from age and gender, to interests and page likes. At this point, it was important for us to determine, based on the filters offered by Facebook, what variables would most effectively target our audience. A solid understanding of your ideal respondent at this stage can greatly increase the effectiveness of your targeting—but, as we found out, the content of that ad is of equal importance in terms of qualifying your respondents. Our targeting demographics and advertisement content worked in tandem to qualify our respondents.

 

The Advertisement Content

Qualifying respondents with your advertisement means writing ad copy and choosing an ad image that speaks directly to your target and only your target. In order to ensure the effectiveness of our ad, we conducted some simple A/B testing. Every headline and image was tested against an alternate headline or image to ensure we were using the most effective combinations. This process was repeated in phases over the course of our recruitment campaign and, as a result, the effectiveness of our ad was optimized as we moved further and further along.

although the sample size was relatively small, our early analysis actually showed that respondents recruited on social were more in line with some of the key demographic indicators of the target population when compared to respondents recruited through traditional means

Using this strategy, we were able to limit the amount of unqualified clicks on our ad and, in turn, lower our overall recruitment costs.

 

The Survey

Targeting helped to pinpoint our audience, and the ad content grabbed their attention. It was, however, the qualifying questions on the survey that enabled us to cross-verify that respondents were who they said they were. Respondents had to qualify across a series of questions—questions that were not identified within the ad content that otherwise would have primed them. Logic was built into the survey to ensure that respondents were screened out dynamically if they did not qualify. On the back end, respondent IPs were checked to ensure that they had not repeated the study, in addition to the contact information they provided to receive their honorarium. As this study focused on one particular ethnic group, we were also able to further authenticate our respondents by identifying surnames that were common to that particular group.

The campaign was a big success, and we were we able to recruit a sample over and above our initial goal. Social recruiting yielded n=45 valid, completed surveys at a fraction of the cost of some more traditional recruiting methods. On top of that, although the sample size was relatively small, our early analysis actually showed that respondents recruited on social were more in line with some of the key demographic indicators of the target population when compared to respondents recruited through traditional means. For example, the Filipino respondents recruited on social were significantly more likely to speak Tagalog (the official national language of Philippines) at home, while the respondents recruited through traditional means were more likely to say they speak English. This, along with a variety of other indicators, could suggest that the respondents recruited on social are actually closer to what we expected from the average Filipino ­newcomer. There are also products like Environics Analytics’ OriginsCanada software that can help us analyze this data for greater validation.

While this type of social recruitment is by no means a replacement for traditional recruitment techniques, it can be an excellent option to consider for niche respondent groups. It’s just one more example of the rise of digital communications and interactions changing the research industry along with our clients’ needs. If we, as researchers, expect to keep up and truly provide our clients with the insights needed for data-driven decision-making, we must continue to innovate and take advantage of new platforms and technologies wherever possible.

Find out how Environics can help your organization

Related insights

Toronto

366 Adelaide Street West
Suite 101, Toronto, ON
Canada M5V 1R9
416 920 9010

Ottawa

116 Albert St
Suite 300, Ottawa, ON
Canada K1P 5G3
613 230 5089

Calgary

421 7th Ave SW
Suite 3000, Calgary, AB
Canada T2P 4K9
403 613 5735

Share This