Instagram: a case study on one of the PR lessons from 2012.

It all started with what could and should have been a fairly reasonable change to their privacy policy…if you believe their explanation.
On December 18 Instagram announced via its blog that it was proposing changes to its privacy policy and Terms of Service which would take effect on January 16 2013. Briefly, it said that unless users deleted their account by then they had by default agreed to allow the company to share information about them with Facebook, its parent company, as well as other affiliates and advertisers.
So far, so fairly usual in the online world. The problem was what they wanted to share and on what basis.
The terms proposed Instagram being able to send on users’ username, likeness, photos (along with any associated metadata), and/or actions they had taken for use, “in connection with paid or sponsored content or promotions, without any compensation to you”. Basically, passing on their personal photos and dats for possible use in conncetion with ads without being given a penny.
The backlash wasn’t long coming and was vehement. Here’s how the BBC website reported it. Before long there were threats of mass account closures from regular users and, importantly, power users followed by many others, especially celebrities for whom control of their image is a major financial and PR consideration. So basically it was an online PR firestorm of their own creation.
Within 24 hours Instagram had seen the furore and realised it needed to try to reassure its users or it would have a lot fewer of them to monetise. So co-founder Kevin Systrom posted this fresh blog post saying they aren’t planning to sell users’ photos, anyone who thought that had unfortunately misread the legal language, but it was Instagram’s fault that it wasn’t clear from the start.
Importantly, he also said they were listening to the feedback and were going to modify specific parts of the terms to make it more clear what would happen with users’ photos. He went on to detail their intentions and the reason for them.
Did this extinguish the firestorm? Sadly not. Why? By now users’ mistrust of the company was so high that the reaction of many key players, including power user National Geographic, was to say ‘fine, but we want to see the detailed changes before we trust you again’. Some, including National Geographic, suspended posting their accounts as a sign of their continued unhappiness.
So Instagram went away again to have another think. Sooner after it announced that it was reverting to the original terms of service and if in future it planned to change them it would consult with users about the proposed changes before putting them in place.
So everybody’s happy now? No, not really. Why? Because their trust in the brand has been seriously dented. And, as we all know, trust, like reputation, is easily lost and rebuilt slowly and with difficulty.
Instagram, of course, wasn’t the only brand to suffer reputational damage in 2012. Barclays, RBS, HSBC, Starbucks, Amazon, Apple and Google were among the high-profile companies affected by events entirely of their own creation. Self-inflicted and entirely avoidable injuries.
Instagram’s was one of the worst because they failed to take sufficiently into account how personal the service they provide to their users actually is — its place on what marketers call the ‘my continuum’ (i.e. a hairdresser provides such a personal service that you refer to them as “my hairdresser” while, say, an online retailer is more distant and impersonal, so not referred to as “my”).
Users trust Instagram with their personal photos, some of them very personal — like shots of their newborn baby or wedding — so in threatening to hand on these photos to third parties it was inadvertently making a threat to very personal possessions of its users.
It also failed to take into account quite how competitive its market had become and how low the barriers to exit are for users. Though clearly the dominant player, it quickly found how loosely tied users felt to it (compared to,say, Facebook versus the threat of Google+) and how quickly they were prepared and able to take their custom to rivals. 
I suspect this misjudgment may have been by Facebook managers, who are used to the cosy notion that without the ability to easily download and move the data and friends Facebook users have spent years adding to the site they feel compelled to stay simply by the thought of the massive ‘transaction costs’ (hassle) of shifting it all to a rival).
It was a similar sudden appreciation of how competitive its market was and how low the barriers to exit (switching) were that forced Starbucks in the UK to change its policy on paying UK Corporation Tax. 
A Parliamentary Committee named and shamed it among several international companies which generated large revenues in the UK but didn’t pay much, if any, Corporation Tax because of their use of indefensible transfer charges to export profits to jurisdictions with lower rates of tax on them.
The root cause which links these and all the other PR disasters cited above is an imbalance in their perception of the correct balance between meeting the needs of shareholders (who they have a corporate fiduciary legal duty to) and their other key stakeholders, in particular their customers, who perceive that they are owed a moral duty extending beyond any legal-contractual one.
Balancing their often-competing needs is not always easy, but it’s easily seen that if you work exclusively in the interests of shareholders (i.e. by maximizing profits) you can easily be working against their long-term interests as customers may defect to rivals, reducing sales and ultimately profits. Equally, if you only do what customers want, your profits may be few and far between and shareholders will take their capital elsewhere.
So how to find a balance?
Part of the answer, I suggest, is in discovering what your Reputational Elasticity of Demand (RED) is and how far that allows you to consider your shareholders versus your customers.
So what is RED and how is it made up and measured? That and more will be in my next blog post.

Twitter lesson for the money-saving expert: like wishes, choose your hashtags with care

By Alan S. Morrison

This morning Mr MoneySavingExpert, Martin Lewis, launched a crusade (check out its manifesto in his Daily Telegraph article, it’s more than a campaign) to have people resist the peer pressure to buy Christmas presents for non-family members so as to avoid unnecessary spending and waste.
To promote the article and the campaign, Telegraph editor Tony Gallagher tweeted a headline and link at 9.11am. Martin Lewis retweeted it almost immediately to his 178,239 followers. Seven minutes later he was tweeting to ask people not to automatically slate it, but read the details first.
Nineteen minutes later he tweeted his gratitude that people were supporting him. But, crucially, he included the hashtag #banXmaspresents
After retweeting a few tweets of support from his followers, he put out an appeal at 10.12am for retweets with the hashtag to get it trending.
By 11:24am he admitted that he was starting to get abuse on Twitter for his call and asked the antis to read the article before coming to judgement.
Three minutes later he was, all credit to him, retweeting a comment that, I believe, gets to the heart of the lack of wholehearted support — the hashtag.
People just reading that and the call for support were, wrongly, writing him off as a Scrooge and telling him where to stick his humbug.
After a few more retweets of support, by 12.03pm he felt it necessary to explain further while repeating the hashtag again.
Which he wouldn’t have needed to do if he had chosen a better hashtag in the first place. I actually tweeted him a suggestion - #KeepXmastoFamily I haven’t yet had a reply.
So what’s the lesson here? Simply, if you’re going to create a custom hashtag, do it with care. While it’s perfectly possible he created this one for maximum controversy (to gain most free media coverage of his campaign), I’d say it has at least partially backfired on him by alienating potential supporters who only had time to read the tweet and not follow the link to the full explanation.
The thing with hashtags is that, like newspaper headlines, they should help communicate the core of what you’re saying, or at least the feeling. They’re often best used for humorous, often sarcastic, comment on the rest of the message and can be very effective as rallying calls for support on a heartfelt issue. But they must be quickly and easily understood properly, which this one, IMHO, isn’t.
If he’d chosen a better one the amount of abusive reaction from misunderstanding would have been reduced to those who purely disagreed with what he’s actually saying, not them plus people defending gift-giving. Who knows what damage has been done to the brand of him and his company as a result.
For a big company like his, with long-established credentials as championing the public, this will likely be a blip on the road. But for smaller firms who are less well-known the potential reputational damage of such an error could be catastrophic.
One careless tweet can be misremembered for a long time and, as the old saying goes, it can take years to build a good reputation and five minutes to ruin it.
Martin Lewis may well be the money-saving expert, but he still has a few things to learn about managing his reputation online.

Here’s why you need to think before you tweet & retweet

This morning’s news that actor and comedian Alan Davies is among 10,000 Twitter users facing legal action over the false Lord McAlpine allegations is the latest, but biggest, example of why you have to think before you tweet…and retweet.
One of the problems of Twitter is that users very quickly forget that they’re not only sharing their comments, and, crucially, those of anyone they retweet, with their pals but also everyone else on Twitter. So you’re pretty much publishing it to millions of people. Just like the conventional mainstream media.
The difference is they have long understood the consequences of getting their facts wrong. Apart from specialist lawyers, no-one knows the laws of libel better than journalists. It’s a key part of their training because the power to publish to millions (or even just thousands) of readers is something that has to be treated carefully.
So they understand that saying you simply repeated something someone else said isn’t a defence in law — you’re responsible for publishing it again. Aside from the Reynolds defence, you have to be able to prove anything you publish, or face the consequences.
So the old advice that you shouldn’t tweet anything that you wouldn’t say to the person’s face is reinforced by this latest example of what happens if it turns out to be untrue.
Think not naming the person will protect you? Not so. Look at Newsnight. They didn’t name Lord McAlpine, but they said enough for him to be identified by enough people for his name to start circulating.
Again, the concept of what’s known as ‘jigsaw identification’ is already well-understood by the conventional media. They already have to watch for it with cases with child victims or accused under 18 as well as rape victims — ensuring that individually and collectively they don’t give out enough details for the person to be identified by someone who might know them.
Similarly, if you keep it too obscure you could be sued by several people who could argue people might mistakenly think it was them — 10 policemen successfully sued a paper in England because it ran a story about ‘a policeman’ from a particular station.
Once upon a time to be a publisher you needed a printing press and all sorts of other expensive gubbins and so realised that you had a lot on the line if you got your facts wrong in print.
Twitter may be free and easy, but the consequences of saying or repeating something you personally have no proof for are just the same. So think before you tweet or retweet.

MIT may be inventing future technology, but it’s up to us to decide how we use it

Once again I’m indebted to the print edition of Wired UK for news of a couple of things which will be a real boon to people in the world of marketing.
The November issue (sorry, no link as its content isn’t online to avoid cannibalizing sales of the print edition) is entirely devoted to the work of the MIT Media Lab, which the cover headline rather grandly describes as “inventing the future”. Hyperbole aside, it’s come up with some cool stuff over the years — from e-ink (think Kindle) and Lego Mindstorms to the XO Laptop.
Among the latest things its team of world-leading researchers have developed are two pieces of technology of particular interest to marketers.
The first, Video Response codes, are basically like the QR codes we’ve seen in print for a number of years, but are instead embedded invisibly in video displayed on screens.
The idea is that you use your smartphone to take a photo of the screen and the software grabs the VR code, decodes it and links you to text, images, coupons, whatever, to be downloaded to your phone. Presumably, as the codes are imperceptible to anyone watching the screen, the video will have to have some kind of ‘snap this for cool stuff’ message included so you know there’s a VR code there to be captured.
Straight off the top of my head I can imagine this being used to create interactive cinema and TV adverts (QR codes have already been tried on TV, but there’s been a mixed reaction to them from consumers for various reasons). Or embedded in YouTube videos to link to other Web-based content.
As with so many technologies, their use is limited only by your imagination.
The second technology is Affdex. Originally developed in the Media Lab, it’s been spun off into its own company, Affectiva.
Put simply, the software tells you what the emotional state is of the person your camera is looking at by measuring key parameters on their face.
The original algorithm was built on autism research to help sufferers read the emotions of other people, It was first used to create “Mindreader” spectacles which told the wearer, via a tiny LCD screen, whether the person in front of them was happy, sad or whatever.
Rosalind Picard and her team took it further by training it recognize the emotions of more people from around the world by showing it videos. The goal — so a computer could recognize their emotional states.
It’s now so successful that marketing research giant Millward Brown has partnered with Affectiva to use it to get accurate real-time feedback from consumers and research subjects on TV adverts. The result is more accurate data where subjects may feel socially constrained from saying how they feel — not everyone’s comfortable telling someone their ad sucks.
But, as with almost any technology, the other possible uses are huge. Imagine having it for when you Skype a key journalist about whether they’re going to use a story you’ve pitched (think lie detector!) or are following up a business pitch with a potential customer (are they really as interested as they say?)
Unsurprisingly, pollsters are already excited about Affdex’s potential for more accurately judging voters’ reactions to candidates’ TV ads and debates. And the US National Security Agency has already expressed interest in using it for obvious reasons, although Affectiva aren’t keen to licence it for military use.
As much as we may legitimately want to know what others really feel in certain work circumstances, we also sometimes want to keep our own emotions private.
The article quotes other possible uses including allowing Facebook or YouTube to read your emotion (via your laptop webcam) as you watch a video (e.g. of a friend’s baby) and post your reaction online automatically. But what if you’re secretly jealous of your friend’s joyful event but don’t want to upset her? What if you don’t really think your friend’s new outfit is lovely?
Do you really want Facebook and YouTube, or your government if you live in a repressive country, to know how you truly feel about certain articles or videos online?
Clearly this technology has great potential for ethical uses, but given its claimed ability to get behind social barriers in the communication of feelings it has to be used within a strict set of opt-in guidelines and licenced carefully.
MIT may be inventing technologies of the future, but it’s up to us to decide how they’re used.