Who Actually Joins a Brand Community
Companies buy communities on the theory that belonging produces loyalty. The single randomised test of that theory found no effect on bids or revenue earned and a negative effect on listings and amount spent, and the positive findings all come from designs in which the customer chose to join.
Answer to the question
Who actually joins a brand community?
Disproportionately the heaviest buyers a brand already has. Two brands' Facebook fan bases came out skewed towards the heaviest buyers while their real buyer bases followed the ordinary spread, and the one randomised test of a brand community on commercial behaviour, run for a year at eBay Germany, found no gain in spending and a loss on listings.
Key takeaways
- The one randomised test of a brand community on commercial behaviour, run for a year at eBay Germany, found no effect on bids placed or revenue earned and a negative effect on listings and amount spent.
- Two brands' Facebook fan bases were skewed towards the heaviest buyers, while the brands' real buyer bases followed the ordinary spread of heavy and light.
- The field's best positive result is observational, and its authors say in print that self-selection is a concern they addressed by ruling out alternative explanations.
- The trade's 20 to 40 per cent spending multiple traces to a 2011 consultancy brief citing its own survey of more than three thousand consumers, with no published method.
The proposition sold to companies is that a brand community builds belonging, and belonging builds loyalty. The evidence offered for it is thinner than the claim. The founding papers measured no purchases, the figure the trade quotes rests on an unpublished survey, and the one randomised trial of the idea is a study we could reach only in abstract.
That last limit applies throughout, so state it once. Every peer-reviewed study named here was read at the level of its publisher-deposited abstract, because the full texts are closed, and four separate routes to the eBay Germany experiment failed. What survives that constraint is still a thesis: belonging is real, and the narrower commercial claim, that a firm can build a brand community and create spending it did not already have, rests on studies in which the spenders chose to join.
What did the founding papers measure?
The field's founding paper is a definition and an ethnography. Albert Muniz Jr and Thomas O'Guinn, in the Journal of Consumer Research in 2001, described a brand community as "a specialized, non-geographically bound community, based on a structured set of social relations among admirers of a brand", working from ethnographic and computer-mediated data on three of them, centred on the Ford Bronco, the Macintosh and the Saab. They found the markers of community: shared consciousness, shared rituals and traditions, a sense of moral responsibility. There is no outcome variable in it, no comparison group and no purchase measure, and the paper claims none.
The second founding paper creates the problem. James McAlexander, John Schouten and Harold Koenig, in the Journal of Marketing in 2002, placed the customer at the centre of a fabric of relationships, and their stated fourth contribution is "a new and richer conceptualization of customer loyalty as integration in a brand community". That is a definition rather than a measurement, and it is presented as one. Where loyalty is integration in the community, membership and loyalty are the same quantity, and no sales figure can bear on whether either produced the other.
Hope Jensen Schau and colleagues gave the field its best account of what members do, in 2009: twelve common practices across nine communities, in four thematic aggregates, through which consumers realise value beyond what the firm anticipates. The value there is value realised by consumers in the practice, and the paper reports no revenue. Nine communities is a great deal of ethnography, and no quantity of ethnography is a control group.
Where does the trade multiple come from?
The figure a manager is likeliest to have met is a range: customers who engage with companies over social media spend 20 to 40 per cent more money with those companies than other customers. It comes from a Bain and Company brief of September 2011, "Putting social media to work", which attaches its evidence in a single sentence.
> "Our recent survey of more than 3,000 consumers helped to identify what makes social media effective."
The trace stops there. A survey of more than three thousand consumers, reported in a client-facing brief, with no sample frame, no categories, no questionnaire and no comparison design published anywhere. There is no academic paper behind it and the brief does not claim one. It sets people who engage beside people who do not, and that comparison is the subject of everything below.
The same document is unexpectedly candid elsewhere, warning that sentiment analytics was then "in its infancy, with serious limitations". It also reports, with no method attached, that the average Facebook user will like no more than seven companies or brands, a claim about how small a brand community can be. A document sceptical about a neighbouring vendor's measurement and silent about its own headline method is a good specimen, and the asymmetry is the thing to notice rather than the number itself.
One further multiple circulates attached to a peer-reviewed study, and two attempts to confirm it at source failed, so it appears nowhere here. Other circulating figures we have not traced.
How strong is the positive evidence?
This is not a literature with nothing in it. Puneet Manchanda, Grant Packard and Adithya Pattabhiramaiah, in Marketing Science in 2015, studied a multichannel entertainment products retailer that launched an online customer community and found "a significant increase in customer expenditures attributable to customers joining the firm's community". They called the benefit social dollars. Posters generated more of it than lurkers, members with more ties more than members with fewer, and the informational channel came out null, which they read as a stronger social than informational source of economic benefit.
Their own caveat is the sentence the trade never quotes: "While self-selection is a concern with field data, we rule out multiple alternative explanations." Ruling out alternative explanations is not the same thing as assigning people to a community. The magnitude usually attached to this paper is another matter, since no percentage appears in either abstract and two passes failed to recover one, so none is printed here.
Two transaction-data studies sit alongside it. Rishika Rishika and colleagues, in 2013, combined customers' social media participation with individual transaction records and used propensity score matching with difference in differences to address self-selection, finding higher visit frequency and higher profitability. The moderator is the real finding: the effect was greatest for customers who already had a strong relationship with the firm and bought premium products. On its own terms that describes an amplifier on existing patronage.
The moderator is the real finding: the effect was greatest for customers who already had a strong relationship with the firm and bought premium products.
Kwok Yee Goh and colleagues joined an apparel retailer's fan page to individual transaction data and reported, in the 2012 working version we read, that engagement in social media brand communities leads to an increase in purchase expenditures, with content from users mattering more than the retailer's own. Their design compares people who joined with people who joined later, and their reported elasticities for one construct run from 0.006 in directed communication to 3.140 in undirected, a spread wide enough that the large figure should never travel alone.
The strongest genuinely randomised result nearby belongs to Ravi Bapna and Alexander Umyarov, in Management Science in 2015. Using a social feature to induce adoption of a paid service among randomly selected users of a large online social network, they found that peer influence causes more than a 60 per cent increase in the odds of buying the service when it comes from a friend who has adopted it. The treatment there is a friend's adoption, not a membership, and the outcome is a subscription. It is the strongest card the pro-community side holds, because the mechanism works at the level of a single tie.
Who actually joins a brand community?
The measurement that bears most directly on who is inside one is small. Karen Nelson-Field, Erica Riebe and Byron Sharp, at the Ehrenberg-Bass Institute, compared the buying concentration of two brands' Facebook fan bases against those brands' actual buyer bases. The buyer bases followed the ordinary negative binomial spread of heavy and light buyers. The fan bases were "skewed in an opposite pattern, skewed toward the heaviest buyers". Two brands is two brands, both in fast-moving consumer goods, so what follows is the best available hypothesis and not a settled count.
On that reading the people inside a brand community are not a cross-section of a brand's buyers. They are disproportionately the people who were already buying the most, which is the likeliest reason a comparison of joiners against non-joiners finds what it finds, and the reason the Bain range describes who joins.
There is an older account of where that loyalty came from. Andrew Ehrenberg, Gerald Goodhardt and Patrick Barwise restated double jeopardy in 1990: a small brand has far fewer buyers than a large one in any given period, and its buyers also tend to buy it less often. [Brand Loyalty Is Mostly Brand Size](https://thewhyweproject.com/articles/brand-loyalty-is-mostly-brand-size) argues that case out, and asks why loyalty differs between brands. The question here is narrower: whether a particular managerial object causes the loyalty of the people inside it.
Mark Uncles and colleagues supply the line that bears hardest. Loyalty in competitive repeat-purchase markets, on their reading, "is shaped more by the passive acceptance of brands than by strongly-held attitudes about them", from which the demand-enhancing potential of loyalty programmes is more limited than might be hoped. A brand community is the purest available bet on strongly held attitudes, and the adjacent promise made by [loyalty apps](https://thewhyweproject.com/articles/why-loyalty-apps-use-random-rewards) belongs to the same family. [Does Brand Purpose Move Buyers](https://thewhyweproject.com/articles/does-brand-purpose-move-buyers) met a structural cousin of this problem, where the largest synthesis measured what people said.
What happens when joining is assigned?
In 2010 Rene Algesheimer, Sharad Borle, Utpal Dholakia and Siddharth Singh published, in Marketing Science, the one randomised test of the commercial proposition we have been able to find. Their abstract puts the question more cleanly than any critic has, asking whether participants are the firm's fans to begin with and self-select into the community, or whether participation leads to the behaviour. The data come from a year-long study at eBay Germany, where a simple email invitation significantly raised customer participation in the firm's own community. The invitation is the instrument: it moved people in, and then the firm watched what they bought and sold.
> "Community participation did not translate into increased behaviors, as would be commonly expected. Although there is no impact of participation on the number of bids placed or the revenue earned, there is a negative impact of participation on the number of listings and the amount spent."
The authors read that as selectivity rather than damage. Participants became more selective and efficient sellers, and more conservative in their spending on the items they bid for. A community that makes buyers better informed can reduce what they spend without making them worse customers in any sense they would recognise. One finding runs the other way: the invitation worked on ordinary customers as well as on enthusiasts, so a community programme can be aimed wider than a fan base.
No sample size, model or coefficient can be given. It is one firm, one market, 2010, and a marketplace instead of a shop. The missing study is that design run at a firm that sells products.
We went looking for a second such test on two passes, through the index that catalogues this literature and through the citation graph of the 2010 paper. What came back was field experiments about getting people to join communities and post in them, plus one further field experiment inside a brand community, published in 2017 by Yi He and colleagues, whose outcomes are brand identification and member role behaviours and which measures no spending. No keyword web search ran, so this is a report about our searching.
Why does plural membership protect the member?
The most unsettling result here comes from the field's own friendly witnesses. Scott Thompson and Rajiv Sinha followed the participation, membership duration and adoption behaviour of 7,506 members across four brand communities and two product categories, and found that heavier participation and longer membership raised the likelihood of adopting the preferred brand's new products while lowering the likelihood of adopting an opposing brand's. That oppositional loyalty held only where the rival's product was not first to market.
Then the reversal. Where memberships overlap, higher participation in one brand community may raise the likelihood of adopting a rival's products, a result the authors themselves describe as surprising and disconcerting, because managers usually do not know which other memberships their members hold.
Michael Morris, in Tribal, reaches the same fact from the human side, and his framing is why the commercial question is worth asking at all: tribe-level learning, more than individual brainpower, is on his account the secret of how our species adapted. His structural definition of a cult is a community that severely restricts the prestige signals its members receive, so that leaders acquire extreme mystique and control, while the protection ordinary life affords is that people belong to several communities with differing heroes, which "empowers us to question any one of these heroes".
He cites no study in the book that measures a commercial outcome of a brand community, so he is here for framing and vocabulary. Plural membership is what stops any one community owning a person, and it is also what stops a firm owning a customer. A brand that genuinely became someone's only tribe would have built a pathology and booked it as an asset.
That leaves the member with a more useful reading than the manager gets. If the heaviest buyers are the people who join, then joining is a signal of what you already do rather than a cause of it, and the badge you were handed for belonging was issued against spending you had already done. The programme did not make you loyal. It found you, correctly, and charged your loyalty to its own account. You owe a brand nothing for a habit you already had.
Common questions
Who actually joins a brand community?
Disproportionately the people who were already buying the most. Comparing two brands' Facebook fan bases with their real buyer bases, Karen Nelson-Field and colleagues found the buyer bases spread across heavy and light buyers while the fan bases skewed towards the heaviest. That is two brands in fast-moving consumer goods, read at abstract level, which is a narrow base for the point it carries here.
Does joining a community make customers spend more?
On the strongest available design, no. The year-long field experiment at eBay Germany, reported by Rene Algesheimer and colleagues in 2010, found that community participation did not translate into increased behaviours, with no effect on bids placed or revenue earned and a negative effect on listings and on amount spent. Observational studies do find gains, and in all of those the customer chose to join.
Is the 20 to 40 per cent spending figure sound?
Treat it as a description of who engages. The range comes from a 2011 Bain and Company brief citing its own survey of more than three thousand consumers, with no sample frame, questionnaire or comparison design published. It sets people who engage with companies on social media beside people who do not, which is the comparison that the fan-base skew finding contaminates.
What did the founding brand community papers establish?
That these communities exist and have a describable structure. Albert Muniz Jr and Thomas O'Guinn defined one and described three, finding shared consciousness, shared rituals and traditions and a sense of moral responsibility, with no outcome variable anywhere in the design.
Sources
Muniz, A. M., Jr., & O'Guinn, T. C. (2001). 'Brand Community.' Journal of Consumer Research, 27(4), 412-432. doi:10.1086/319618
McAlexander, J. H., Schouten, J. W., & Koenig, H. F. (2002). 'Building Brand Community.' Journal of Marketing, 66(1), 38-54. doi:10.1509/jmkg.66.1.38.18451
Schau, H. J., Muniz, A. M., Jr., & Arnould, E. J. (2009). 'How Brand Community Practices Create Value.' Journal of Marketing, 73(5), 30-51. doi:10.1509/jmkg.73.5.30
Algesheimer, R., Borle, S., Dholakia, U. M., & Singh, S. S. (2010). 'The Impact of Customer Community Participation on Customer Behaviors: An Empirical Investigation.' Marketing Science, 29(4), 756-769. doi:10.1287/mksc.1090.0555
Manchanda, P., Packard, G., & Pattabhiramaiah, A. (2015). 'Social Dollars: The Economic Impact of Customer Participation in a Firm-Sponsored Online Customer Community.' Marketing Science, 34(3), 367-387. doi:10.1287/mksc.2014.0890
Rishika, R., Kumar, A., Janakiraman, R., & Bezawada, R. (2013). 'The Effect of Customers' Social Media Participation on Customer Visit Frequency and Profitability: An Empirical Investigation.' Information Systems Research, 24(1), 108-127. doi:10.1287/isre.1120.0460
Goh, K. Y., Heng, C.-S., & Lin, Z. (2012). 'Social Media Brand Community and Consumer Behavior: Quantifying the Relative Impact of User- and Marketer-Generated Content.' SSRN Electronic Journal. doi:10.2139/ssrn.2048614
Bapna, R., & Umyarov, A. (2015). 'Do Your Online Friends Make You Pay? A Randomized Field Experiment on Peer Influence in Online Social Networks.' Management Science, 61(8), 1902-1920. doi:10.1287/mnsc.2014.2081
Thompson, S. A., & Sinha, R. K. (2008). 'Brand Communities and New Product Adoption: The Influence and Limits of Oppositional Loyalty.' Journal of Marketing, 72(6), 65-80. doi:10.1509/jmkg.72.6.065
Nelson-Field, K., Riebe, E., & Sharp, B. (2012). 'What's Not to "Like?"' Journal of Advertising Research, 52(2), 262-269. doi:10.2501/jar-52-2-262-269
Ehrenberg, A. S. C., Goodhardt, G. J., & Barwise, T. P. (1990). 'Double Jeopardy Revisited.' Journal of Marketing, 54(3), 82-91. doi:10.1177/002224299005400307
Uncles, M. D., Dowling, G. R., & Hammond, K. (2003). 'Customer loyalty and customer loyalty programs.' Journal of Consumer Marketing, 20(4), 294-316. doi:10.1108/07363760310483676
He, Y., Chen, Q., Lee, R. P., Wang, Y., & Pohlmann, A. (2017). 'Consumers' Role Performance and Brand Identification: Evidence from a Survey and a Longitudinal Field Experiment.' Journal of Interactive Marketing, 38(1), 1-11. doi:10.1016/j.intmar.2016.11.001
Bain & Company (2011). 'Putting social media to work.' Brief by C. Barry, R. Markey, E. Almquist & C. Brahm, 12 September 2011, bain.com
Morris, M. (2025). Tribal: How the Cultural Instincts That Divide Us Can Help Bring Us Together. Third printing, LCC GN490 .M67 2025