A few months ago, we carried out market research based on social media for one of the brands we work for. I counted on the expertise of Aitor in this case.
How to Carry out Market Research based on Social Media
The way the research was carried out and communicated in social media was as follows:
- Website: create a slide with the call to action that leads to the news.
- Pinterest/Instagram: an image with the call to action, both as text and as a visual. For instance, a question mark.
- Google +: developing the questionnaire as a post.
- YouTube: using speech bubbles (like in a comic strip) with the link at the end of the channel's most viral videos.
- Facebook: in the form of "questions" in Facebook.
- Twitter: inviting the community through surveys and screening the information obtained.
- Monitoring and active listening: detecting and analysing mentions and filtering them according to sentiment.
All of this was later taken to a spreadsheet for further analysis.
Performance and Results
We worked on the basis of three variables that sought to obtain information on brand performance in three of its main activities.
The surprise, perhaps, was that the results obtained in 2 of the 3 activities were quite far off from our perception and what we expected.
The company's management didn't understand what was going on. According to them, the results should have been clear. The thing is, they were forgetting two essential factors: firstly, their own bias and perception of the world; secondly, the community on the other side, a group of people with their own widespread tastes, habits and, sometimes irrational, behaviour. Believe it or not, this information refers back to the opinion of the brand's community, its fans, the community that buys, shares and helps expand the brand, that observes what the brand does on a daily basis.
Limitations and Opinion
Considering that a reasonable error margin in market research is +/- 5%, a reasonable error margin in social media, and more specifically Facebook, may be approximately +/- 35%, harbouring a guess. In this case, the study's reliability is completely lost by not being able to segment our sample universe homogenously.
In any case, the information has been collected from the answers provided by the community over a 3-month period. Trusting any information collected in the Social Web is a delicate matter; however, this doesn't mean that these figures aren't significant or that we can't work with them. If you ask me, I believe in the truthfulness of the public. But that's just my opinion.
The people in your community love what you do; at least the crazy fans. They'll bend over backwards for the brand every time you do something. When you're asked to collect primary information from a social media source you're exposed to the risk of reliability and heterogeneity, making it difficult to assess this information.
You can base your research on several social platforms, websites and online marketing. In our case, the basis has always been Facebook. However, we used other sites to bounce things back and forth to other platforms. I don't really think that all of our public is there, but most of it is. If you need to check what people on the street think, that's not our job. However, it's important to do so.
Don't Ignore What's Important
We spare no efforts to get as large a sample as possible to obtain a variety of results. Perhaps you don't like the information we've obtained, but it's what your potential public thinks. Are you going to ignore it?
Conclusion: can you trust surveys 100%? The answer is no, but they can serve as a guide, as a way to get feedback and improve. That's what it's all about, isn't it??