FEATURE22 May 2012

Open to all?

Peter Dann and Adam Staton of The Nursery probe consumer attitudes to privacy in social media – and ask whether researchers are welcome.

Res_4007447_social_apps

There’s an intellectual battle going on in research – and social media is at the heart of it. Conflicts are often disputes over boundaries, and this one is no different. In this instance, we have researchers fighting over what is and isn’t public and what can and can’t be used when it comes to researching people within social media environments.

Social media research is not like other forms of active market research. We do not recruit respondents or observe them, we simply source and analyse snippets of people’s conversation on social media about a particular brand or topic. It is the modern equivalent of reading local papers, parish magazines and public noticeboards, except that in the 21st century everyone has the ability to publish their thoughts, no matter how banal or irrelevant. How can it be wrong to read that stuff? It is, after all, publicly available.

But is it meant to be? For some the argument goes: people don’t know that these forums are public, and they should be protected from their own unintended publication. Both the Market Research Society and Esomar, two of the industry’s standards-setting bodies, say broadly that social media listening exercises run contrary to the principle of informed consent which lies at the heart of industry codes of conduct. One commentator likened it to peering into someone’s living room and overhearing the conversation they were having inside. Surely that observation is misguided and inappropriate. As professional and ethical market researchers we take people’s privacy seriously; this makes us sound like snoopers.

Of course, the important question to ask is whether social media users see market researchers in this way. Do they feel they need protecting from us?

Hawkers and stalkers
We conducted our own survey of 423 social media users, a mix of those who were active Twitter users and those who weren’t. The majority were recruited through Research Now’s mobile panel, topped up with 17 respondents re-recruited from a previous Twitter survey.

What was very clear from the research was that people who comment about brands on public social media platforms want their voices to be heard: 9 out of 10 Twitter users in our sample said this was the case at least some of the time.

Equally, most people are already so familiar with brands being present in social media that they are perfectly comfortable with researchers being there too. Indeed, of our sample 56% already knew that companies conduct market research in social media and 42% of those who weren’t previously aware weren’t really surprised to hear it. We asked if people were bothered by researchers being active in social media and 71% said they weren’t.

What’s the bother?
We were interested to see what it was that concerned the other 29%, so we re-contacted them to ask them to explain. Nearly three-quarters responded, of whom the majority ( 76%) were worried about their private data being compromised, while 59% were nervous that they would be sold to. Both of these concerns fall squarely under existing codes of conduct, which would apply to any research in any media – social media included. There is an obligation on all of us as researchers to preserve our respondents’ anonymity and to refrain from selling our clients’ wares or letting their details be used for that purpose.

More tellingly from a researcher’s point of view is that respondents were very clear that different social media channels have varying levels of privacy, as the chart below shows. Broadly, very little is seen to be private – people have woken up to the fact that they have to be as careful in social media as anywhere else on the internet. Indeed, theoretically private spaces such as password-protected groups are still seen as being open by a fair few of our respondents.

But we saw some confusion about the status of personal Facebook pages. There are as many who consider it a public space as there are who see it as private. Of course, Facebook itself is constantly shifting its position. It starts out as a social networking site where you keep in touch with friends, then you start befriending people who aren’t friends – and liking brands that aren’t even people. Facebook itself might change its privacy rules; then it allows advertisers more space on a page. No wonder there is so much confusion.

Let’s be clear
This confusion is absolutely no reason to avoid social media as a research resource – although it is good reason to ensure that our social media research is as respectful and professional as any other research we conduct. When we work in social media research we can’t lurk in private forums. We don’t try and crack open private Facebook pages. We anonymise verbatims in social media just as we do in traditional research. And we trust our clients not to Google a verbatim to track down its author, just as we trust them not to follow respondents home from a viewing facility and mug them on their doorsteps.

We think that if 9 out of 10 people want to be heard by the brands they’re talking about on social media, then we as researchers should facilitate that – just as we do in traditional research. And to the 10% of people who tweet about companies and don’t want them to hear, we say: “Wake up and read the not very small print on Twitter’s own ‘About us’ page: “Businesses use Twitter to quickly share information with people interested in their products and services, gather real-time market intelligence and feedback, and build relationships with customers, partners and influencers.”

Recently, we wanted to use one of our Twitter surveys as a case study and asked the authors of all our verbatims if they minded us using them. Not only did all but one say yes, they were incredulous that we seemed to be missing the point of Twitter. “If you post anything on the web it’s out there for everyone to see,” said one. “If you feel you want it to be private then don’t post it in the first place.”

If researchers don’t embrace this area ethically and professionally, we will leave a huge and valuable space open to exploitation by social media practitioners less guided by ethics than we are. But the principal of informed consent shouldn’t apply here any more than it does in any published media.

Peter Dann won the Best Presenter Award at MRS Conference 2012

6 Comments

12 years ago

Nearly all Market Research is either contract driven by a client who wants the research done or it is done on spec with the idea that the data can be sold to clients to whom it is relevant. Why does this article approach this topic from the perspective that "research" is purely academic? The only researchers who are not driven by economic considerations are academics who are seeking to aquire something other then actual money...Ethical considerations regarding "research" are mostly the purvey of academics who are not seeking some financial consideration for their efforts...Or is this a cynical view of why research occurs?

Like Report

12 years ago

A very helpful addition to the debate about codes of practice. I wonder if we should introduce a new kind of status: someone who is not a respondent but a correspondent or witness. The difference being that their anonymity is not protected - this removes them from the sphere of market research entirely but their data is usable for developing marketing insights. If people want an audience then let us give it to them without the encumbrances of codes designed to protect them from all powerful corporates who online have none of the power they were reputed to have.

Like Report

12 years ago

I agree with the author and I do believe that people understand that social media is a place of public conversation. (with the exception of a few politicians who seem to continue to forget). From a researcher's point of view social media provides a unique opportunity to hear what people are saying about a brand or topic without the need to kick off a conversation with "I'm just about to ask for some opinions" or "here's money in exchange for views". Where social media can provide phenomenal research insight is when you look over a (for example) 5 year period and look for trends (e.g. When do people start talking about starting Christmas shopping?, What time of day are people talking about EastEnders? Are more men or women talking about the Jubilee?) Or where trying to identify common associations with words or phrases (e.g. Wouldn't it be useful to know that your phone brand was frequently used in the same sentence as "poor battery" but 20% of your competitors are singing the praises of the screen?) There is more of an argument that engagement is more intrusive. If I tweet that I've just sat down in McDonalds, do I really want the UK's social team to tweet me thanks and an invitation to make sure I try their new burger? There are billions of conversations going on in the social stratosphere - why not find out what they're saying. Paul Bennison

Like Report

12 years ago

After more than 40 years in the buiness of marketing research, I have no problems with many of the privacy concerns that are discussed here. But I do have a concern with the assumption that members of an online research panel represent the general public. "What was very clear from the research was that people who comment about brands on public social media platforms want their voices to be heard: 9 out of 10 Twitter users in our sample said this was the case at least some of the time." ResearchNow is an excellent panel source. That is not my point. But assuming that members of the panel are exhibiting attitudes that are held by the general public misses the point that these folks are on the panel BECAUSE they are interested in having their opinions heard by brands. Too often people take results from surveys of online audiences and assume an entire population feels the same way. Remember whose information you are quoting and avoid the tendency to build sweeping proofs of your argument on results from a limited population.

Like Report

12 years ago

Dwight - Your point of view that "Ethical considerations regarding "research" are mostly the purvey of academics who are not seeking some financial consideration" scares me. Ethics are a massive part of how I do my work and I am in this business to make money. If listening researchers don't care about the ethics of what we do, we will be put out of business by government and the public. Paul: Please don't make the mistake of assuming that we can treat all people the same way. Despite the fact that the majority of people are open to hearing from brands in social media, a large percentage are NOT. They find it to be an intrusion, to be creepy, and completely inappropriate. We cannot treat everyone as open to it simply because it's convenient for us. Researchers must remember that the internet consists not of internet savvy individuals like ourselves who continuously seek to educate ourselves on this topic but also people who have NEVER read the TOS for any social network, who don't know that a comment on a website will be read by people all over the world, and who don't know any hot-keys. As researchers, we are bound to do no harm. Ignoring the wishes of 25% of the population because it's not convenient sounds like doing harm to me. Harm to research contributors, and harm to the research business.

Like Report

12 years ago

Sue - This is my argument with any panel (focus group, telephone, online or otherwise). You are always going to get people who want their views heard. The mere fact they've raised their hands is testament to this fact. Annie - As I said in my post, "There is more of an argument that engagement is more intrusive". And I think that (as described in my McDonalds example) this is almost certainly the case. When getting involved with social engagement, it is clearly a mistake to assume that everyone is open to this kind of conversation. On the other hand, there are many companies completely ignoring social channels at the moment and not responding to messages directed directly at them. This is like installing a phone line, and letting it ring out. The continued argument that people don't realise everything posted is public is a difficult argument to maintain. Sure, 2 years ago these channels were alien to a lot of people, but there will come a point it's beyond reasonable to assume that people 'get it'. As more and more national news stories highlight again and again something that's been posted by accident or that just makes the author appear stupid, the public nature of these channels will eventually be understood by all. When I speak about using social media for research, I'm really talking about using what has been said and gaining insights from that,rather than taking a group from this channel and using it to fire through specific questions. Of course, this doesn't work when researching a future product or a planned advertising campaign. But can be incredibly effective at discovering thoughts and opinions on a current or past product My argument is that there is a huge amount of data out there packed full of thoughts, opinions and trends if we take the time to look into it we can gain a lot of valuable information that can help us inform future decisions. Utilising the data in this way is surely do no more harm than counting individuals at a shopping centre to gain traffic numbers as we're not specifically tracking individuals, but trends. I mean, if I were to say that there had been in the region of 1 million mentions of the next Bond film, Skyfall, within social media channels over the last 2 weeks. 37% of which mentioned Daniel Craig of which 87% was positive - surely this would be of interest to Columbia Pictures? Especially if Daniel's contract was just about to be signed for another 2 movies. Of course these stats have been constructed from tweets, blogs etc. and you could argue that a % were posted with the impression that they were being posted privately. But I don't think anyone could disagree that using data of this scale and the information, that can be extracted from it could be really useful to a company or organisation. Paul Bennison

Like Report