Monday, July 30, 2012

No Correlation Between Facebook Fans and Stock Market Performance [Social Media Study]

Social media professionals and their bosses take some things for granted. For example, having more Facebook fans is always better than having fewer. Hard to argue with that; the more Facebook fans you have, the stronger business results will be, because fans are happy customers. Plus, every fan acquired is another person who can be targeted for marketing with low-cost Facebook posts and apps.

It makes perfect sense. It also happens to be wrong, at least as evidenced by an analysis of 40 top brands with some of the largest Facebook fan counts. For these brands, there is a negative correlation between fan count and customer satisfaction and no relationship whatsoever between fan count and stock performance.

Do I think these results truly mean that Facebook fans have no (or even negative) value? Of course not, but there are vital lessons to be learned from this analysis: We put too much stock in the importance of our brands' Facebook fan counts; how you get fans is more important than merely having fans; and most companies are still doing social wrong.

Analysis: Correlating Facebook Fans, Customer Satisfaction and Stock Performance

I examined a set of 40 companies, chosen because they are the top 200 corporate brands on Facebook (based on fan count) that are both tracked by the American Customer Satisfaction Index (ACSI) and publicly traded on the NYSE, NASDAQ, Frankfurt, Tokyo or Taiwan Stock Exchanges. By focusing on three attributes--Facebook fan counts, customer satisfaction ratings and 12-month stock performance--it is easy to calculate the statistical correlations between these variables.

Out of the top 200 corporate brands, 160 were eliminated for one of three reasons:
  1. The ACSI does not track the brand's customer satisfaction rating. This precluded major brands with large Facebook fan pages, such as Disney, Red Bull and Converse. 
  2. The brand is part of a privately held company and thus lacks market performance data. Brands such as Subway and Levis were omitted for this reason.
  3. The brand is only one of many owned by the parent company, and as a result, individual stock performance data is not available. For example, Taco Bell, Pizza Hut and KFC all have large fan pages but are collectively owned by YUM! Brands, so no stock market data is available for each of the individual brands. Also, despite having large fan counts, Bing and Xbox were not included since they are relatively small portions of Microsoft's business operations, but Microsoft itself is included in the list since it has a large fan count and is publicly traded.

Every brand lusts for the sort of enormous Facebook fan counts that these brands have accumulated, but as it turns out, these large large fan bases have no statistical association with either customer satisfaction or business results as valued by the market. Looking at the correlation coefficient (where 1 is perfect positive correlation, -1 is perfect negative correlation and zero is no correlation at all), here is what statistical analysis demonstrates:

  • Modest negative correlation (-0.3) between Facebook fans and customer satisfaction:  ACSI Customer Satisfaction Scores for these brands ranged from Facebook's dismal 61 to Amazon's strong 86. Overall, large Facebook fan counts are not correlative with customer satisfaction; in fact, there is a negative correlation between the size of a brand's Facebook fan base and its ACSI score, at least among these top 40 brands. My suspicion is that this is related to the manner in which some large brands amassed fans, many having lured fans with giveaways and freebies rather than superior products or customer service. (I'd also speculate that a larger study might find the same trend, since people tend to be more satisfied with smaller, locally owned companies versus large international conglomerates.)
  • No correlation (-0.1) between Facebook fans and stock performance: The data did not reveal any correlation between the size of a brand's fan count and its 12-month stock performance. Among these top fan pages you will find brands like Walmart, Hershey and eBay, with share prices up more than 25% in the past twelve months, but you will also find companies with stocks that have dropped more than 35%, such as Research in Motion, Nokia, Facebook and Best Buy. On average, the companies in this list saw share prices increase just 2.9% in the past 12 months, lagging the major market indices (such as NASDAQ and the DJIA), which are up 6.6% to 6.9%. There is no evidence that investors care about large Facebook fan bases or that having large numbers of fans enhances business results in a way that investors notice or value.
  • Weak positive correlation (+0.2) between customer satisfaction and stock performance: Not surprisingly, real affinity mattered more than the fake variety represented by Facebook fan numbers. For these 40 brands, there is a slim but positive correlation between the companies' ACSI ratings and their stock performance in the past year. Satisfied customers tend to result in satisfied shareholders.
The companies included in this analysis and the data collected can be found at the end of this blog post. 

Recommendations: Social Media that Matters

Based on this evaluation, some conclusions may be drawn that can assist your brand's social media efforts:

  • Stop Using Fan Counts as a Measure of Social Media Success: Working to increase fan counts without consideration for brand affinity can lead to activities that hurt rather than help. If you strive to collect Facebook fans that have true affinity with your brand, your fan growth may be slower, but you will be rewarded with authentic fans with higher EdgeRank; these are fans that see and engage with your content and are more likely to make their friends aware of your products and services. But if you resort to trickery such as sweepstakes and game giveaways, you end up collecting fans with little affinity and low EdgeRank. These faux fans are unlikely to see your content or speak on your brand's behalf, which means you will have invested your company's budget without having delivered business results.
  • RIMM Market Cap ChartDo Not Blindly Trust Published Studies (Except This One):  There are many "studies" that are unleashed on the blogosphere for the purpose of PR and not enlightenment. Use judgment as you read blogs and press releases, and do not mistake assumptions and opinion for facts and data. (This is why I shared the actual data: so that you can consider and verify the analysis for yourself. Plus, I have nothing to sell you!) For example, a while back, a social media management provider got a lot of press by announcing Blackberry's Facebook fans were worth $240 million to the company--$83.98 apiece. Social media bloggers and presenters breathlessly repeated this figure without stopping to consider if this was plausible. To put that figure into perspective, Sprint has 56.4 million subscribers and a market cap of $12.9 billion, which means investors value Sprint's actual paying customers at $228.72 apiece. Does it make sense that a real, paying, contracted subscriber is worth $228.72 while an individual who clicked a "like" button is worth $83.98? Of course not, as evidenced by Research in Motion's results in the two years since this crazy fan valuation figure was announced: Revenue has dropped 39% and the firm lost $20 billion in market valuation. RIM's huge Facebook fan base did not prevent it from arriving on death's doorstep.
  • Focus on Advocates:  You do have some customers who may be worth $83.98 to your organization; in fact, they may be worth far more than that. These are your advocates--people who spend more, recommend your brand and influence the purchase decisions of others. BzzAgent, in a study I find very credible, found that brand advocates create twice as much brand content, are 70% more likely to seen as a good source of information by others and are 50% more likely to influence a purchase decision. To be clear, advocates and influencers are different, as noted by Cara Fuggetta on the Zuberance blog. If you wish to pursue a PR-like influencer strategy, do so, but do not confuse this with strategies designed to meet the very different needs, wants and outcomes of your value-driving advocates.
  • Scale to Create Satisfied Customers, Not "Fans": Which came first, the chicken or the egg? That philosophical dilemma has no answer, but I do know which comes first, a happy customer or a Facebook "fan." As evidenced by the correlation between customer satisfaction and market performance, brands are much better off increasing the number of satisfied customers rather than the number of their Facebook fans. There are ways to use social media to increase customer satisfaction; this is not done with sweepstakes and games but with in-channel responsiveness and functionality that satisfies customers' needs. In the past three months, I reached out to three national brands using Twitter--Dell, Sprint and Yahoo--and did not receive a response from any of them. Doing social at scale is not about the biggest marketing budget or the most fans--it is about having the right resources to meet customer expectations in the social media channel. 

Some folks claim that we have entered a stage of maturity in the social media era, but from where I am sitting, we are still barely taking baby steps. So many brands are desperately pursuing ROI and increasing their social media marketing budgets, but few are deploying the sorts of social business strategies that create happy customers and engaged advocates. There is too much focus on the click of a "follow" or "like" button and not enough on how social changes the brand/customer relationship.

Social media managers love large fan bases because they look good to the bosses, but it is how you get fans that matters, not that you merely have fans. To drive business value, invest in creating happy customers who want to be your Facebook fans, not in Facebook fans you hope will become your happy customers. 

Data: Top Publicly-Traded Facebook Brands with ACSI Ratings

Brand Facebook fans
(In millions)
Most Recent ACSI Score
(2011 or 2012)
Percent share change in past year
(as of 7/27/12)
Facebook (FB) 70.1 61.0 -37.6
Coca-Cola (KO) 45.9 85.0 16.3
Starbucks (SBUX) 31.1 76.0 18.7
McDonald's (MCD) 21.4 73.0 2.8
Walmart (WMT) 18.2 70.0 40.6
Target (TGT) 16.3 80.0 22.0
Blackberry (RIMM) 11.6 69.0 -71.0
BMW (BMW) 10.7 83.0 -16.8
Google (GOOG) 10.2 82.0 3.9
Nike (NKE) 9.9 80.0 7.2
Yahoo (THOO) 9.8 78.0 19.3
Adidas (ADS) 8.6 80.0 11.2
Nokia (NOK) 8.5 75.0 -63.1
Pepsi (PEP) 8.5 85.0 13.1
Amazon (AMZN) 7.9 86.0 6.0
Kohl's (KSS) 7.6 81.0 -10.7
Domino's (DPZ) 7.1 77.0 27.5
Dunkin' Donuts (DNKN) 6.8 79.0 6.1
Best Buy (BBY) 6.1 77.0 -36.4
Macy's (M) 6.1 77.0 26.0
Burger King (BKW) 5.4 75.0 1.8
Hershey (HSY) 5.3 84.0 27.0
eBay (EBAY) 3.4 81.0 38.1
Netflix (NFLX) 3.4 74.0 77.9
Southwest (LUV) 3.0 77.0 -10.6
Verizon (VZ) 2.9 70.0 27.2
JCPENNEY (JCP) 2.7 82.0 -25.2
GAP (GPS) 2.6 77.0 55.1
Chili's (EAT) 2.6 76.0 35.2
Hanes (HBI) 2.5 82.0 -2.6
Wendy's (WEN) 2.5 78.0 -0.4
Papa John's (PZZA) 2.5 83.0 64.3
AT&T (T) 2.3 70.0 26.9
HTC (2498) 2.3 75.0 -66.2
Honda (HMC) 2.2 85.0 -18.6
Walgreens (WAG) 2.1 75.0 -8.5
HP (HPQ) 2.0 78.0 -47.2
Microsoft (MSFT) 1.9 75.0 8.6
Dell (DELL) 1.9 77.0 -26.0
Sears (SHLD) 1.8 76.0 -25.6
Average 2.9
DJIA 6.8
S&P 500 6.6
Correlation coefficients:
Facebook fans and ACSI Scoree -0.3
Facebook fans and Stock changenge -0.1
ACSI Score and Stock change 0.2



Unknown said...

The first thing I noticed looking at this data set is that the fan count is dominated by Facebook who has 30 million more fans than the next nearest brand.

Running the analysis without this massive gorilla switches the Fans/ASCI correlation positive (0.1) and the Fans/Stock correlation positive (also 0.1)

I was thinking that Fans/Stock correlation would be even more positive once Facebook's dismal stock performance was removed, but this is not the case.

Augie Ray said...


Thanks for taking the time to comment and consider the analysis. Ironically, I did the EXACT same thing you did as I was composing the blog post. I noted, as did you, that Facebook has the largest fan count, the lowest ACSI rating, and among the worst stock performance, so I yanked it out and found the same thing as you. Yes, it changes the correlations a little, particularly making the fan:satisfaction rating more of a random correlation than a negative one. But, in the end, removing Facebook doesn't have any real impact on the other correlations.

Thanks for the thought and consideration on this!

Terry said...

nice post augie

My question is whether Facebook pages with fans in the millions are social media at all. There can be no concept of dialog so why do fans join at all? Is it to get deals, updates or to make complaints. What do brands want - is it just message delivery -mass marketing? If so, we cannot expect any correlation between satisfaction and fan count because this is not the common bond.

Augie Ray said...

Excellent comment, Terry. You're right--mass media communications done in a social channel is hardly a good definition of social media marketing. Still, the stupid social media "arms race" continues, and too many social media pros are being complimented or promoted based on their fan counts.

While looking for a job, I had a conversation with a fairly well known social media mover and shaker with her own social media company. I was candid in my thoughts of one of the programs they had run, which added TWO MILLION fans to a brand's Facebook page in ONE DAY. I made the same arguments I always do on my blog--about the value of true engagement and real fans versus just giving crap away--and she fundamentally could not see my point. To her, more fans was good under any circumstance--period.

To me, this attitude demonstrates there is much we need yet to learn and act upon before social media is done right. And in the meantime, I think man brands are harming themselves, not helping themselves with their poor social media strategies.

Thanks for the dialog!

Tom Buchheim said...

Great insight, as always, Augie. I'm using your data to make my case against the arms race with anyone who will listen (which hopefully includes my bosses).

Thanks for the ammo. Hope the job search is going well, sir.

Terry said...

As we move to the next phase, we will be able to appreciate multiple types of strategy. Each will have different objectives and required tactics. The "more the better" does have a role but for a very limited segment of the market. Other strategies will focus on community building, closed group partner communications, and customer communications. Right now we are in that naive stage where we value just one strategic vision, but your blog helps us all to become more mature.

Augie Ray said...

Thanks Tom.

Terry, I don't think "the more the better" is almost ever a good strategy unless it has a qualification. The more the better--of our actual customers. The more the better--of our advocates. The more the better--of people who want to know more about our brand. Any brand can collect Twitter followers who won't see their tweets or Facebook followers who won't see their posts.

But, I'm right with you--the quicker social can become mature in companies and used for multiple, value-focused strategies, the better!

Fard Johnmar said...


I like this analysis, but one of the things that I often think about when analyses of this nature are conducted is whether doing a bi-variate analysis like this really takes into account the full business value of social channels -- especially from an intelligence gathering, product development and customer insights perspective. First, I agree with you that the get more fans arms race is misguided, which is the primary purpose of your post. But, as we move toward leveraging digital channels to enhance business operations and influence other important metrics shouldn't the analysis begin to focus on the holistic benefits of these technologies rather than the impact of one-to-many communications (such as Facebook fans)?

I'd also love to get a sense of whether Facebook fans are driving profits at these companies, which obviously has a lot more influence on stock price (revenue), versus audiences who are not being activated or encouraged to do anything that provides short or long-term economic benefits to these firms.)


Augie Ray said...


I very much agree with you--I would have liked to compared Facebook (and Twitter) presence to more than just stock market performance (and do so for far more companies.) Unfortunately, it's just me blogging, and that analysis took me 8 hours as it is. I hope it might inspire someone in the business of research and analysis to get a little more serious about using statistical analysis to tie social media efforts to REAL business results. I am growing tired of seeing "case studies" that are measured in the number of retweets, new fans or Pinterest pins are created.

Thanks for the comment!

Fard Johnmar said...


Thanks for your response. Yeah, I feel there's a real need to elevate the overall analysis of social (I like to use the word digital) analysis in terms of the impact of this content on business drivers beyond sentiment, volume and engagement metrics.

We're doing some work in this area (in the field of digital health) and will have more to share in the coming months. Thanks again for the dialogue.