Putting a value on a Facebook fan

by Seb Robert in News Google+

The IAB kindly invited us to present at their Great British Social Media Festival last week. Given the theme, there couldn’t have been a more perfect brand to be co-presenting with than Bulmers – the only British cider brand to have a Royal Warrant from the queen.

As anyone who’s ever attended this type of social media event knows, the age-old question of the what’s the value of social media tends to come up, quite a lot.

As you may have seen in Marketing Week and NMA, we had some research findings in this area to share.

The work came about because Bulmers wanted to find an answer to how relevant it was for them to invest and play in the social space. TNS pulled together a short 25 question online interview which we placed in a Facebook tab and shared on a status update with Bulmers fans. TNS went about gathering data from a control sample of general cider drinkers.

TNS gathered their results and worked with our client and fellow presenter, Doug Cook – Brand Manager at Bulmers to come up with a model to realistically reflect the incremental value of a fan of Bulmers on Facebook versus a normal UK cider drinker.

When we got our answer, it was good. Using claimed consumption data from fans versus the control sample, we were able to clearly determine that fans of the brand were more favourable towards the brand vs competitors and more importantly, consumed more of the brand than a non-fan. Using sales data Bulmers calculated their Facebook fans were worth £3.82 more a week, or £198.64 per year – around 38% more than the control sample.

As we pointed out in our presentation, there are some caveats to this, including the likelihood that Bulmers fans who interact the most with the page would have a clear bias towards the brand and would have been the most likely of all our fans to see the status update calling for people to complete the questionnaire. Nonetheless, these results are extraordinarily positive.

Since then, we have increased the average daily ‘People talking about this’ score and over doubled our reach without any paid support so we are really keen to see whether we see an upward trend in fan value in the future; either by shifting that £3.82 per fan number upwards, or whether we are increasing the number of fans who are worth this amount through deeper engagement. We are also working with TNS to roll this research out across our client base, as well as using it as an opportunity to generate more insight into who our fans are so we can engage more effectively in the future.

Back to the festival itself, it was a fantastic day with some great presenters – special mention goes to Russell Goldsmith at Makettiers4dc on ‘Driving offline purchase through social with video couponing’, Lizzie Murray on ‘How the Met Police make the most of social’ and Dave Coplin from Microsoft who gave us ‘a luvvies guide’ on what it means to be a social brand. Hope to see you at the next one!

If you liked this post, why not subscribe to We Are Social by or ?

  • Steve Cater


  • Pingback: Bulmers finds their Facebook fans are worth £3.82 per week - Simply Zesty()

  • Jonny Stark

    I know you’re not going to go into too much detail on here (IP and all that), but the methodology, as described, sounds overly simplistic and incredibly open to factors that would render the results almost useless.
    How did TNS select the control sample, what was the size of the sample, what was the size of the FB sample, what was the statistical significance etc. etc.
    You point out the inherent bias of self selection of the FB sample (and then dismiss it) but this is, as any student of statistics – even entry level – will tell you is quite a big thing to dismiss.
    Appreciate that sounds somewhat negative, and I will say that the attempt at exploring this area and trying to set up some benchmarks for Bulmers is admirable.
    Look forward to hearing how you refine and improve the methodology in future posts.

  • Jonny Stark

    Just to append to my original comment (I cunningly didn’t log in via Disqus so can’t edit the original), I am aware there is some info on slide 7 of the preso re sampling and methodology. I’d just like to know and understand more!

  • http://gearboxmagazine.com/ Brian Driggs

    This is an interesting way of going about things. Self-confessed fans are likely to spend an average of £3.82 more per week or 38% more than the control group. 

    Just to be clear, this does not mean the value of a Facebook fan for this brand is £3.82. The value of a Facebook fan for this brand is the same as it is for any other brand: average net profit per fan after the costs of building/staffing the social media channel are subtracted. 

    1. Ensure a way of tracking fan purchases (coupon code, club card, etc.).
    2. Calculate sum total of sales to Facebook fans via Facebook.
    3. Subtract cost of operating the Facebook channel. 

    For the sake of easy numbers, if 100 Facebook fans each spend £3.82 over the course of a single week as a direct result of a Facebook campaign, that’s £382.00 total revenue generated that week. Assume that’s net profit. If the total cost of the channel operations during that time was £130.00 (one staff member paid £13/hr for 10 hours’ labor), the ROI is £382 – £130 = £252, or £2.52 per fan (aka: 193%). 

    And, if you see your Facebook fans spend an average of 38% more than those who know have taken the bait on your conventional marketing efforts, I’d say it’s probably a good idea to start reallocating your budget.

  • Pingback: TWICs XXXIV()

  • Pingback: Heineken – Q: How Much are our Facebook Fans Worth? A: 80 Extra Bottles a Year Each ($25.8M) | Social Commerce Today()

  • Pingback: Bulmers: Bir Facebook Fanının Defter Değeri | Dimağ Mercan()

  • Jemma Watkins

     I know you’ve added a couple of caveats but surely this is just demonstrating the correlation between going out of one’s way to like the Bulmer’s Facebook page (and answer a questionnaire which I assume was not incentivised), and drinking a lot of Bulmers, which is less “extraordinarily positive” and more “as expected”?

  • hans

    But as we all know: correlation is not causation. =)

  • mony