This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more here

OPINION4 May 2011

Who are we meant to be ‘spying’ on?

Opinion

Online listening, scraping and mining might sound invasive, says Rosie Campbell, but is privacy really at risk when individuals get to choose how ‘public’ they want to be and what persona they want to project? The real danger for research, she says, is the increasing faux-ness of the online universe.

At its most extreme there is a danger that ‘listening’, ‘scraping’ and ‘mining’ online have a distinct stalker-esque character about them, especially where we blur the edges of reality to the point of employing ‘para-humans’ or bots which are merely an amalgam of characteristics, behaviour patterns and viewpoints to effectively spy on people in social media situations.

But I don’t think privacy violations are likely to be our industry’s undoing, and here’s why. Any heavy user of social media almost certainly chooses how ‘public’ they want to be and indeed – here’s the rub – who they want to be.

We are all a slightly different version of ourselves on Facebook, Twitter or Linkedin. Indeed some of us consciously choose to project different personae in different contexts. We choose not to be the same character all the time and we certainly aren’t always the character who meets friends face to face.

The extent to which we bend and twist and fictionalise ourselves in what we choose to say or post, the stance and tone and amount we vernacularise our language all act to create a slightly (or considerably) alternative persona. And the evidence is that we are increasingly inclined towards flexibility with the truth and a degree of personality refurbishment in our online self-manifestations.

What we are looking at is an increasingly unreal research process, where we are surveying or interviewing (or, more likely, accruing ethnographic material by stealth) among an increasingly unreal sample of state-changed people. Who gets valid or useful findings here? What are the meanings of people’s online views? Who are they trying to impress when they post a clever follow-up comment on YouTube? Why are people apparently so much more politically active on Twitter than they could ever be in the real world?

These layers of disguise are tough enough to crack, and that’s before our analytics tools have been put to work to try to make sense of all the sarcasm or local-reference wordplay employed in so much of our online postings, opinions and comments.

This all adds to the increasing faux-ness of the online (especially social media) universe where we make friends with brands, we imagine we have loyal communities of people who like things, we fake co-creation processes from amalgamated voices and viewpoints and consider we have democratised the research process.

So whose privacy should we be concerned about? With a plethora of cooked facts, feelings, retorts and even entirely concocted identities, maybe it is the sanity rather than the morality of the research exercise we need to question.

And if our increasing incursions into consumers’ presumed private worlds causes a meltdown in goodwill towards research, if a regulatory whistle is blown about unsolicited or unpermitted access (our industry’s version of the phone-tapping scandal) maybe we’ll be saving ourselves from another fate – the accusation of being off with the fairies.

Rosie Campbell is a director of Campbell Keegan. She has been a qualitative research consultant for over 25 years, working for major FMCG companies, service industries and government departments. She focuses on communication and language.

7 Comments

6 years ago

Very interesting article and I can believe the idea the industry is spying on personas, but those online personas are part of a real person. The Facebook version of you is still you. Few people are truly ‘acting’ online. If the online persona slags off a brand online, that criticism is still valid isn’t it? Maybe people say things online they wouldn't otherwise, but that doesn't make those views untrue / invalid.

Like Report

6 years ago

Rich - agree with your comment that an online persona is (may be?) part of the real person, but my concern is that it is just a fragment of the person and therefore we have to be careful what we make of it (as ever in research). In qual research (I believe) we are always looking for patterns, connections, trying as best we can to understand the parts in relation to the whole. Individuals often believe two quite contradictory things. For instance last night I boldly claimed I didn't care what happened to my body after I died... but I wanted to be buried. (Sorry, perhaps not the best example!) I hold both these positions equally strongly. I think it is very common for people to hold conflicting views without attempting to reconcile them. What's interesting, from a marketing/social/cultural perspective is not so much the positions themselves but how these positions inter-relate/ evolve/are expressed in different contexts and sometimes how they could be shifted. But we can't understand this unless we attempt to explore the whole person. Without this further analysis the criticism of a brand, in isolation, is just data. We don't know what it means. I am quite curious that within the research industry we are talking a great deal about the benefits/necessity of co-creation (which I would see as including co-creation between different aspects of ourselves within our own heads, as well as the more common definition) Sometimes, talk of co-creation even takes on an evangelical tone. However, simultaneously we seem to be metaphorically chopping people up into discrete personae depending on the medium in which they are communicating or the nature of their communication. It seems to me that - crudely - the more chopping up we do, the more we de-humanised people in our research and analysis, the less likely we are to discover 'the truth' about them. But of course What is the truth? is for another discussion.

Like Report

6 years ago

Hi Rosie, Great article. You are certainly right that specific online properties elicit alternative personae on the part of the content generator - what I'm writing and the way I'm writing it here are very, very different to something I might write on Facebook. Yet we shouldn't doubt what I would call the "contextual authenticity" of the content: Whether ultimately useful to the researcher or not, what you get from me on Facebook, LinkedIn, Twitter et al is always authentic in context, and this to me is a clear differentiator over more 'invasive' methodologies like F2F, CATI, or even online surveys which all suffer from observational bias. As for "a regulatory whistle [being] blown about unsolicited or unpermitted access" - I suspect that the lobbying power of social media businesses will ensure that this never becomes a serious issue. I'm going to take a punt here and say that as analytical and semantic deconstruction capabilities grow, Social Media Research will in time replace much of the online access panel research we are doing now. Regards, Matt

Like Report

6 years ago

If anyone watched Catfish on Channel 4 recently they'll know that some people who might seem extremely convincing online (even when you've spoken to them on the phone!) don't actually exist, or are multiple 'versions' of one individual. This lady created 15-20+ different personas on Facebook, loosely based on existing people, and effectively generated a new world for these people to inhabit to make the lead 'characters' seem convincing. They all had photos, profiles, other friends, wall posts, status updates etc.etc. and the whole thing was extremely elaborate and convincing. Embarrassingly so for the guy who fell for the whole thing. Creating 20 online characters might be extreme, but it happens, and probably more than we might expect. In this case each of these virtual people represented fragments of what the woman in question wanted to be or used to be, but none of them were actually her. The film was quite a moving story of a lonely woman living out some of her fantasies online, although this certainly isn't what you start off thinking (I'd recommend watching it as it's a great film, although I've spoilt the ending - sorry!). I agree with the article too though, we need to be aware that we are only seeing what people want to share and project through social media which may be heavily edited or sometimes distorted. And we also need to be mindful that although 15 personas on Facebook might not be the norm, many people box off certain aspects of their life and may well have a personal and a work account (e.g. on Twittter) where they behave differently and project very different sides of themselves. Any research involves us interpreting what people say - whether that is online or not - and not always taking it at face value. If we are using social media to do this - assuming we can, following the release of forthcoming MRS/ESOMAR guidelines - we just need to be even more mindful of this.

Like Report

6 years ago

I really wanted to say what I really thought about this article but I didn't want to go public.

Like Report

6 years ago

The issue to privacy online has to do more with personal details including addresses, phones, email ids, bank details etc. It is very easy for any app or software to "self-authorise" itself to plough through the computere or track emails to identify details you didnt want to reveal. There's also the question of generalising profiles (e.g. any gentleman who's a fan of say Tiramisu is automatically classified as an homosexual) - its incorrect and not fair to the individual. You might gain other personal details from their conversations, which simply are not the mandate of insights/ research And as far as feedback on brands / companies are concerned, it is good enough to simply collect voluntary information, and be happy with that. Asking for more, only would allow abuse to set into the process...

Like Report

6 years ago

Maybe we are carrying this intrusion too far, and getting involved into people's lives more than we should (real or virtual, genuine or put on). Especially when all we need for our analysis are primarily profile data first and then maybe some other interesting connections, but not really probing deep into what relationships they have, what fantasies people indulge in etc. the point is that is subject to abuse, and like the alpha and beta error, it is better to be cautious. At least as far the MR industry goes, the respondents rights and privacy comes first. If you dont believe this - you should see what is perpetrated in the name of research amongst various online communities - from multilevel marketing ponzi schemes, to fake surveys - its high time MRS and ESOMAR took an active role in promoting market research and the guidelines we adhere to protect our most valuable inputs - the respondents!!

Like Report