Finding Goldilocks Influencers: How Follower Count Drives Social Media Engagement

Researcher: Dr. Alexander Edeling, KU Leuven

How many social media followers do you have? In Alexander Edeling's latest publication, the authors examine the relationship between follower counts and engagement with sponsored content. Read the full article in SAGE's Journal of Marketingopen in new window and see below what he has to say about this work.

Tell us about your research background and your field of study.

I’m an associate professor at Katholieke Universiteit Leuven in Belgium and a marketing researcher. I started off in the field of what we call the marketing-finance interface, which is the relationship between marketing on one hand and finance on the other hand, so how do marketing decisions actually drive investor decisions for example.

Recently, I’ve started to work on social media marketing and influencer marketing and that also brought me to the field of eye tracking for the first time, so that’s also how I got in touch with Labvanced for the first time.

What motivated you to look into this area?

So generally, with the study we are doing, we are looking at the question how the followership or the number of followers of an influencer affects the behavior of consumers on social media such as Instagram and we thought it would be nice not only to do that experimentally with a survey but also looking at eye movements and at the behavior of the eyes using eye tracking and because, as we all know, there was this big ‘corona crisis’ and all the eye tracking labs were more or less closed and you couldn’t really reach any participants so we were looking for providers of remote eye tracking. We actually got the hint that there is this German more or less startup that is focusing on that, so we got in touch and that’s how we ended up using your technology.

Can you please describe the research design and how you set up the experiment?

So basically, we did it kind of simple, I would say. We divided our subjects into 2 groups. The one group, which is the experimental group that actually got the task to think about whether they would engage with an influencer. So they had to look at several influencer profiles with this motivation in mind, while the control group was told more or less just look at the profile how you normally would look at influencer profiles. So that you are kind of, it was that the study design so that we designed an area of interest which was the follower count for number of followers and we compare then basically data regarding this area of interest between the two groups. So we created social media profiles based on real influencers but we somehow manipulated them for our demands and that’s how we designed that.

Tell us about your conclusion and the implications of that.

So, the main finding is yes, there is this difference between the two groups. It’s quite interesting to see that actually, based on eye movements you can find significant differences between subjects based on what their experimental stimulus is, their motivation is, so that was kind of interesting for us really and also contributed to our story because we tried to map the mechanism in our study and that was an important part of it.

What are your next steps with this research?

So basically, we are working on other projects in this field of influencer marketing, we’re looking at something where we try to investigate the number of previous brand relationships of influencers and how often they did campaigns with certain brands and how that affects consumer behavior with the expectation that the more commercial these influencers are, the problematic it is for them to appear to be genuine and trustworthy and so that’s what we are working on right now. There might be an eye tracking case in there, we are not sure yet but there as well you could do something with eye tracking.

In what ways is conducting research online different from in-lab?

I think the big advantage is really to get the data really quickly, so you don’t have to set up a lab and have to make sure you get respondents walking into the lab. I mentioned COVID, here it was almost impossible but still after COVID what we see at campus is there is just not the same level of attendance that we had 3 years ago. So that’s a big advantage. What’s maybe kind of a limitation is that you are not really able to observe how the people behave when they are doing these experiments, especially with respect to eye tracking. In an eye tracking obviously you can make sure that the person behaves in a way that is optimal for the experiment and that’s more difficult with online data collection. Nevertheless, there are obviously various, let’s say metrics, that you can use such as calibration error to make sure to clean the data set in a way that only the “reliable” observations remain.

Do you see online research as the future of your field?

Yeah definitely. I would say already more than 80% of all experimental studies, like lab studies, in our field are actually online.

How did you choose Labvanced for your research?

So basically I think it was really word of mouth. So one of our co-authors actually got the hint that there is this company that is active in eye tracking, remote eye tracking, and so we checked out the website and got in contact with Caspar and had a look at the technique and especially the design and found that it’s not that easy of course but it’s doable. With some kind of investment, it’s understandable and very well explained via tutorials and so on. We also checked out other providers and realized that the technology is not as advanced yet so that’s why we chose Labvanced.

What stands out to you about Labvanced?

So I would say the document, the kind of documentation that we have there with all the sheets and the tutorials and that. So you basically, with some investment, you can get all the necessary information you need. Online, there is also good chat function and support function which is great, and then just the great amount of opportunities you have in terms of designing frames and developing surveys and experiments. I think that’s something that I haven’t seen with kind of standard market research tools that you are that flexible, such a high number of opportunities how you design your stimuli. That’s very unique. I don’t know the whole field, but yeah.

What would you recommend to students hoping to begin research in your field?

I would recommend to just try out things and do simple stuff first, before you go into more advanced studies. So maybe kind of start simple and do as many survey experiments as possible because with every new survey you learn something new.

Do you have a message to share with other Labvanced users?

Yeah, I would tell them to try it out, it’s definitely not easy but it’s very rewarding once you got the gist of it. It’s kind of a little bit like magic after some time. It takes some time to get to this point but once you are there, then it’s really very nice and very satisfying to a certain extent, yeah.