Social media doesn’t drive sales? A response


Every so often someone comes out with a study that tries to make the point that investing in social media is a waste of time as it won’t help you sell your stuff.

The latest incarnation is from Forrester (via Business Insider ) analyst Sucharita Mulpuru, who says that less than 1% of online transactions could be traced to a social media post.

I’ll respond in two ways:

That’s one study, here are five that show the opposite
First of all the Forrester research is wildly counter-intuitive. The simple fact of the matter is we do pay attention to consumer opinions online.

Exhibit A – Tripadvisor and Amazon. If social media had no affect on sales why do hoteliers and authors get so anxious about what’s written about them to the extent of (in a few extreme cases) trying to stuff them with fake reviews ? A review site like Tripadvisor is popular precisely because consumers consult it *before* a sale.

1- In fact, according to Nielsen , which conducted a global study of 28,000 consumers in 56 countries, online consumer recommendations are the second most powerful source of ‘brand advertising’ ahead of even editorial endorsements.

Number one was word of mouth endorsement from friends and family which of course sometimes also happen online.

That’s one (fairly persuasive) study that shows the opposite of Sucharita Mulpuru’s report. Here are four more:

2 – 15% of US social media purchasers have bought alcohol due to content they read on a social media site . However the same study by ROI research found that this can increase significantly if that content is about more than just promotions, ad messages and discounts.

For example, Constellation Brands, which has 1.3 million Facebook fans, says 36% have purchased more because of social media.

3 – Constellation Brands raises a key point, it’s about creating a community of advocates who do ultimately buy. According to Bain and co consumers active on social media spend 20-40% more with companies they engage with online than other customers.

4 – And those advocates in turn will persuade other consumers to buy. A comblu study says that super-fans and advocates are 50% more likely to create content that results in purchases

5 – Finally, that positive content and sentiment created by fans is important as social media can certainly work in reverse and harm sales. According to the Convergys Group, a single bad tweet or Facebook comment can cost you 30 customers.

Social media isn’t direct mail
Those statistics I outline above all make a single point – people can create positive sentiment around your brand. That sentiment spreads thanks to online tools. And that sentiment then creates the right mood music for people to buy your products.

However, social media by and large is not a direct response tool, and in that sense the Forrester report is correct. Social media is about everyday human interactions, and human interactions don’t work like that.

I wouldn’t post a Facebook update or send out a tweet along the lines of “hey everyone, click on this link for the new Singer pilsner. It’s refreshing and low calorie too!”

People don’t do it because its crass, you wouldn’t talk like that in real life and so its a sure- fire way to lose friends online. Yet the impetus is sometimes for social media to operate in this fashion, and so sadly some brands talk like that online. No wonder it doesn’t always work as intended.

As a result, I think Sucharita Mulpuru’s study has done us all a favour.

No, social media isn’t comparable to direct mail. And no, I won’t click on a link a friend has sent me to buy a new TV or lawnmower.

However, next time I am browsing in store or online, the chances are when having to choose, I’ll remember him / her raving about it, and I’ll get out my credit card.

Image – Michael Holden


  1. Really appreciate the kind comment Nancy. I wonder if the problem really is the label ‘social media’, which I think often causes misunderstanding. I agree with you in that what we are really talking about is taking opinions and sentiment that would have existed anyway, and amplifying it many times over online. That’s quite a simple proposition but a very valuable one too.

  2. hear hear!!
    Great article Dirk.
    I think to say the statement that Social Media does not impact sales is pure ignorance. Users understand the experience & impact of social media because they see it and feel it.
    The sooner people realise it is a marketing and communication tool that is part of a business activity the sooner expectations will be realistic.
    People connect and recommend products in real life, social media just enables this behaviour beyond that. :-))

  3. Hi Yasin, Asif thanks for the comments and the kind words!

    Indeed, we have a job to do in being a bit more up front in admitting that there isn’t, as you say a direct line between a tweet and someone spending their disposable income (unless you run a specific sales / offers twitter feed), but the recommendations and positive reputation that results from good social media content is extremely valuable.

  4. Great post and to be honest I don’t think it could have been better put.

    The amount of times I have to tell people not to treat or see social media as direct mail or a virtual salesman!

    Found your article via a Tweet

  5. Hi Dirk,

    Great post, the title of the website will always invite responses to posts like this.

    As someone who straddles Market Research and Social Media, my point is this:

    The premise/frame for the research was flawed. It assumes direct causality between SocMed and sales. I know of so many marketing departments who struggle to justify with metric based business cases of their worth to an organisation, because their is an inherent misconception that marketing leads directly to revenue. But it cannot be meassured how much of it can be attributed to generating trust in a product or service. (sentiment analysis would help sometimes)

    Its the same with Social media specifically. It doesn’t have a direct line from a tweet to someone investing their disposable income, if it did, Forrester wouldn’t have made themselves look foolish and out of touch with how businesses really work.

    I wont labour the point, but you got it right, there is a correlation and its not measurable purely by quantitative metrics.

    Keep up sounding the foghorn, some of us are listening!

  6. Hi Richard,

    I absolutely don’t mind follow-up comments! In fact, I think these sorts of discussions are very useful as they make us think about value and how social media should be framed when talking about it as a marketing tool.

    I guess I’ll make two follow-on points

    1 – Of course, social media can be used as a sales tool if that is the desired result. As I am sure you know, there are a number of Twitter feeds that exist purely to sell off excess stock. (US Airline) JetBlue is one example with their @jetbluecheeps handle, which currently has 300k+ people watching for deals

    2 – However, I’ll use a personal example of how social media chatter can have real comercial value without it being direct response.

    I’ve literally just returned from a conference in Abu Dhabi and flew Etihad. On the flight back, the TV screens were broken. Etihad had anticipated this by providing personal Archos video devices for passengers – in economy.

    When we arrived, staff were waiting who handed each of us a letter apologising and offering us 5000 free miles. A lot of airlines would have just shrugged their shoulders, said sorry and that would have been the end of it.

    I will certainly fly with them again, but so far all we’re talking about is good customer service.

    The social media bit comes in by virtue of me tweeting about my good experience, by the fact that I am talking about it on this blog, and because I intend to leave a positive review on (a sort of tripadvisor for air travel).

    Will someone have bought an Etihad ticket there and then as a result of my tweet? Highly unlikely, unless that someone happened to be in Expedia booking a holiday at the very moment my tweet went out.

    However, while I’m certainly not deluded enough to imagine myself as some kind of big league influencer, I do think my various posts will have some financial value for Etihad as (because of the day job) I happen to interact with a lot of people who are frequent travellers.

    The question is how much value? It could be that in a few weeks time, someone remembers what I said and chooses to fly with them. The benefit to Etihad? If one person books, £500+.

    It’s more likely though that my recommendation will just be one piece of what you might call the recommendation puzzle.

    Whoever is booking their flight may not even remember where they heard it or who said it and it will be mixed in with other opinions s/he heard. But how is that different from (say) TV advertising or most other marketing disciplines? I’d argue its equally difficult to isolate the exact value of that one TV ad impression.

    The bottom line though is recommendations clearly matter. And that’s what social media ultimately is all about. It is a platform(s) for brands to harness and amplify positive sentiment. Or, in some cases, to catch and address negative sentiment. That’s by and large not about direct sales. But I’d say its potentially more powerful than that.

  7. Hi Dirk,

    Thanks for the response. I’m really passionate about the topic of data, so don’t treat this response as anything like ‘trying to get the last word in’

    I don’t think the concerns about the studies have been addressed. Take Bain & Co, this study doesn’t prove that people who engage more purchase more, you could easily say the people that purchase more engage more. The two aren’t the same. Surely those that purchase from a brand more often are more likely to engage more (not vice-versa). e.g. their purchasing habits have remained consistent regardless of their level of engagement. If we have a study that proves otherwise, I’d love to see it.

    Constellation Brands might be right. Engagement rates might go up. Yet there is surprisingly little evidence that links engagement to sales. In fact, there are many brands that are engaging heavily right now and seeing sales declining.

    The Nielsen study doesn’t measure buying habits. It measures things which ‘could’ lead to sales. I don’t doubt that negative publicity causes sales to decline and positive promotion causes increases in sales. Yet to what degree isn’t shown here. I’d also guess that this is HEAVILY sector-specific. I consulted tripadvisor social media before booking my honeymoon, but not before buying a home. It’s also very dependent upon the demographic sector too.

    What we’re lacking at the moment is proof that social media does direct drive sales. We need some independently undertaken research that measures the before/after affects of customers being involved in social media – not collected from information provided by the people in charge of running these programmes – of course they’re not going to say their work doesn’t make an impact. Once we start getting them, instead of clutching at a variety of different data points which ‘could’ lead to sales, then we’ll have something.

    At the moment there is plenty of anecdotal examples in both directions.

  8. Hi Richard,

    I appreciate you stopping by and taking the time to comment in such depth. To go through each of your points:

    I’ll let Constellation Brands speak for themselves, but I would hope the point they are making is self-evident, namely that engagement rates go up if the content is right for the audience in question.

    Yes, Bain & co say people who engage with brands spend more. Hence the benefit of trying to get them to engage with you

    Comblu: Yes people who are already passionate about a brand are more likely to refer others. Great! All the more reason to give advocates who love your products a chance to shout about them via platforms where they can reach a lot of people

    The Convergys study: Now I’ll admit that *I* probably can’t lose a brand 30 customers with a social media post!

    So I accept that Convergys puts it a bit simplistically. But negative chatter on social media can indeed lose you customers.

    I have done a lot of aviation / airlines work over the past two years, including working at the sharp end of interacting with passengers myself, and this is now a generally accepted point in the sector.

    Will it lose you 30? Well that depends really. Who is saying it and what were the circumstances? One lost bag, probably not. An injury due to staff negligence (and I’m speaking hypothetically of course), then we could be talking about more than 30

    I think my core point still holds true: most of us have bought something because we heard good things about a brand. And most of us haven’t when we heard bad things

    Social media lets those good and bad things get amplified many times over. To go back to what I think is the most powerful study I cited, the Nielsen one, 28,000 consumers in 56 countries said online consumer recommendations were the second most influential form of brand advertising – ahead of editorial endorsement. Remember those are recommendations from, in a lot of cases, complete strangers

    So what’s the ROI of having a lot of people sing your praises? I would think that’s something any brand would consider pretty valuable indeed


  9. All the data you cite is pretty flawed too.

    Take the 36% – that’s self-reported. That’s not a study. You can’t even find the real data that supports that. Considering Constellation Brands’ revenue has been declining for the past few years, that’s a pretty interesting statement to make. And if it was true, which it might possible be, then it would account for about about 0.3% of total sales. Hardly a major impact. If it worked so well, why haven’t they scale it out across all their brands?

    The Bain and co. report doesn’t account for the fact that the people that participate in social media probably purchase more of the brand anyhow – that’s why they became fans. They try to cross-reference this with referrals, but it doesn’t do the job.

    The ComBlu report also mixed cause and affect. The people that ‘fan’ a page are also likely to be those that are most passionate about the product/service and thus more likely to refer others (and people lie like crazy on self-reported habits). They don’t show that becoming a fan increases advocacy.

    As for the convergys group study, their logic is insane. One bad tweet ‘can’ (note, unproven) cost you customers. They claimed this by looking at the average number of people that receive the message and asking them if they would continue buying. 2/3rds said they ‘might’ not. That’s not very scientific. It doesn’t report actually habits.

    None of these measure the investment, only the returns. Most don’t account for shifting customers from one channel to another. Not the impact that offering discounted products doesn’t increase the ROI.

  10. Thank you for the comment Kirk and that’s certainly interesting about Forrester and their Facebook Factor report, which I’ll definitely check out

  11. Thanks for pushing back against the darkness, Dirk. On the lighter side, one of the very best pieces of recent research I’ve seen employs logistics regression modeling to firmly link social media and purchase behavior. It comes from…(drum roll, please)…Forrester. “The Facebook Factor” was written by analyst Gina Sverdlov, a research and insights specialist. As she wrote, “Brand engagement is a driver of loyalty and purchase for companies, and Facebook is a great channel for advocates to share brand experiences with others. Facebook therefore should be a leading platform when launching a brand advocacy program. Companies should create social content that will keep fans engaged with the brand.”

  12. Thank you Karen and Phil for commenting and Karen for those very useful stats! Yes the type of content and how you engage is key, and I’d also emphasise the importance of integration, rather than looking at social media efforts in isolation.

  13. Totally agree. When I read this, I did a double take at how counter-intuitive it is. To throw in some more stats from Future Foundation:

    37% of the UK population agree that they’re always telling friends/family about new products and services they’ve discovered. Another third are ambivalent about the statement which could reasonably be interpreted to mean that they’re sometimes telling people. So that’s a whole lot of viral going on. Can it really be that this just translates to 1% efficacy?

    1 in 3 of us say that we’re more influenced now by experts online than 6 months ago. As the power of social media grows, so does the number of very niche experts across so many fields. Why would this still not translate into selling stuff?

    And of course, as you say, 1 in 2 of us are keen to share with family and friends. We pass on info about holidays, social lives, places we’ve been, movies we’ve watched, jokes we’ve told etc. Why not purchases we’ve made and recommendations to others?

    65% of active social networkers keep up with/ follow brands on Facebook. And interestingly, those we term Facebook fans of a brand are much more likely than others to agree that they are always sharing details of new products and services with others. This really only converts to 1% purchasing?

    End of rant. But the point, as you so clearly make, is that this report takes a step back to the days of purely transactional relationships between brands and consumers. Social media is much richer than that – it allows dialogue, interface, communication, recommendation. None of that stuff appears on the Accounts Dept’s bottom line but perhaps it should.

  14. Good post Dirk and I think the argument that particularly drives it home is that the point that social media isn’t direct mail. Any marketing discipline doesn’t work if you don’t do it properly and social media’s no different.

Comments are closed.