17 roles that emerge in online customer protests (with Amazonfail examples)

by marycw on April 18, 2009

amazonfail graphics from gawker

Online “riots” or protests are not random, despite appearing to be spontaneous.  They have patterns.  One pattern is a set of emergent roles that individuals assume.   People take on one or more roles based on the individual’s interest and skills/ability to perform those role(s) effectively.

Below are some of the roles that I’ve seen in  online customer protests. I’m using examples from AmazonFail, since it was recent, but other protests show similar patterns.

1) Aggrieved –  people  injured or offended by the company.  In most large-scale online protests,  the Aggrieved are a class or group, not a single person. In AmazonFail, there were two groups: the impacted authors, and those who saw the situation as an attack against GLBT topics/people.

2) Spark – someone who gets the initial audience.   A Spark has to describe the issue in a compelling way and have “enough” readers for the word to start to spread. The main Spark for Amazonfail was Mark Probst’s post.   Sparking doesn’t always “take” (as this example shows.)

3) Town Crier – spreads the news  — the “have you heard this?” folks.  This role creates a full-scale protest.  Twitter is an ideal tool for Town Criers. From April 11-14, there were thousands of #amazonfail tweets.

4) Roll Caller – posts/tweets about the news — not to pass along the information (since at this point, there’s already multiple posts on the topic)  but to publicly demonstrate their support.  People may feel social pressure to participate in Roll Call because *not* participating can look like they *don’t* support it.

5) Activist – creates  online protest locations / petitions.   Facebook Pages/Groups and petition sites are typical protest tools.  Another example:  a blogger that orchestrated a Google bomb around Amazonfail.

6)  Joiner – people who sign up on the protest lists/pages.  Like Roll Calling, Joining is an easy way for people to feel like they’re contributing to a worthy cause.

7) Investigative Reporter –  researches the situation and shares  results.  Here’s an example where a blogger did a master list of books impacted by AmazonFail.  Here a blogger looked at the books’ metadata tags.

8 ) Theorist –  people who speculate about underlying causes.  These speculations can be complex and often involve references to past events that are suspected of being linked to the current problem. One of the most popular theories on AmazonFail was this post about how AmazonFail might be the result of an online group of troublemakers called Bantown.  Another example of a Theorist post here.

9)  Paranoid – subcategory of Theorist. These posts show a strong suspicion of authority (incl. companies and governments).  Tend to see everything as an intentional plot.  This post includes quotes that tend towards the Paranoid view (“…Amazon is a bully….it’s all about sales…they want to be Walmart…”).

However:  what looks like paranoia at first can start looking plausible, esp when there’s an incriminating data trail (example here, where a post that initially sounds paranoid starts sounding justifiable when you read the history).

10) Tinhat – subcategory of Theorists, but crazier than Paranoid.  See the comments on this thread for tinhattery type comments (“…some unnamed company ((Apple)) is attacking Amazon over the online music market…”)

11)  Contrarian – argues that the protest is off track — misinformed, overblown, etc.  Here’s one example.

12) Troll – subcategory of Contrarian who writes an inflammatory opinion to exploit the emotionl/attention of the protest crowd.  Blatant trollery is usually quickly spotted and warned against.  More subtle trollery, like the anonymous coder who claimed to have engineered AmazonFail, take time to debunk (via Investigative Reporters).

13) Expert – subcategory of Theorists, the Expert actually knows something about the industry or company being discussed.   The Expert provides background, context and explanations based on their specialist knowledge.  Here’s a post by a person familiar with ecommerce architecture.

14) Insider – related to  Theorists and Experts — people who claim inside knowledge (as in the stereotypical “I have this friend….”).   Here’s a post and another from persons who cite conversations with Amazon employees.

15) Historian – people who collect and organize the history and key links related to the protest in the days/weeks/months as the situation unfolds. Here’s an example and another .

16) Celebrity – a “name” who adds the weight of his/her reputation to the debate.   Example: this post referenced Neil Gaiman weighing in on Amazonfail, as indeed he did. Clay Shirky commented in the aftermath here.

17) Professional – the mainstream news outlets — usually those organization’s blogs.  Amazonfail was mentioned in various mainstream media outlets, including UK Guardian, LA Times, Washington Post, Seattle Post-Intelligencer, ZDNet , Globe & MailGawker, etc.

If anyone’s come across similar typographies for emergent roles in online protests — pls comment here or email me, I’d love to hear about it.

Useful articles on Amazonfail:

The New York Times has an interesting article on algorithms — how they’re both unthinking automata and also human artifacts that reflect human mental models and biases.

Church of the Customer blog gives a nice list of the many ways that Activists participated in Amazonfail.

Clay Shirky’s after-the-fireworks essay is here, in which he examines the mob effect of online protests.

B.L.Ochman’s “Lessons from Amazon, Domino’s Debacles” is here.

Other references re: online customer protests — books on related topics:

I can’t find any books or articles that are specifically about online customer protests — please anybody who knows of one, post it in the comments.

Here Comes Everybody book cover from Powells
Here Comes Everybody: The Power of Organizing Without Organizations by Clay Shirky (2009 reprint). Awesome book — not about online protests per se, but an overview of how social networks change the power dynamic between individuals and traditional organizations.

Throwing sheep book cover from Powells
Throwing Sheep in the Boardroom: How Online Social Networking Will Transform Your Life, Work and World by Matthew Frazer and Soumitra Dutta (2008). Similar to Shirky’s book, by a couple of guys associated with INSEAD.

Tribes book cover from powells
Tribes: We Need You to Lead Us by Seth Godin (2008).  How self-0rganizing groups form and grassroots leaders emerge.

Cyberactivism: Online Activism in Theory & Practice, edited by Martha McCaughey and Michael Ayers (2003) — a bit dated but still good.

Other links re:  online protests:

List of WordPress blogs with “amazonfail” as a tag.

For one example of another large-scale customer protest with emergent roles, see Strikethrough 2007. Summary:  LiveJournal (popular blogging platform) reacted to a complaint/threat about pedophilia on LJ (complainer was later discovered to be a small – perhaps single person –  protest group). LJ hastily deleted a number of journals for “improper content/TOS violations,” in the process impacting a number of innocent groups and causing an uproar among its customers.


David Harmon 04.19.09 at 5:31 pm

It seems worth nothing that your 17 “roles” fall into a five or six spectrums and clusters. I’d put them in this rough sequence:

Aggrieved, Spark, and Town Crier start the process, which draws in Roll Call, Joiner, and Celebrity. As word spreads, several other groups join:

Activist and Investigator pave the way for Expert, Theorist, and Historian. Insider and Professional tend to trail this group.

Along the way, the hubbub pulls in the “chasers”: Contrarian, Paranoid, Tinhat, and Troll.

Obviously, one can argue about the order and grouping, especially as individuals from each group are trickling in throughout the process. This is just my own take on how the roles fit together and enable each other. It’s also worth noting that individuals often have multiple roles, which can interact — for example, Gaiman’s measured response would have fit the Roll Call pattern, but his net-active Celebrity probably gave it the effect of a Town Crier.

Joel Polowin 04.20.09 at 1:41 am

Just wanted to point out that you might want to check your HTML. At the moment, the counter for section 8, with the right parenthesis, has been converted into a smiley graphic For Your Convenience.

Mary Walker 04.20.09 at 4:21 am

Aha — thanks for the heads up — I'll check it. muchas gracias.

Mary Walker 04.20.09 at 4:25 am

Good observations — yup, mine was very much a first cut at a model / typography. As you say, there's a higher order of abstractions — and there's a timeline for how the protest develops, that ties into the roles. And the roles interrelate, as you observe — a single participant/post can serve the purpose of multiple roles.

I do hope somebody's doing serious research on this — other than people like me, speculating in our spare time — it's a rich topic, and there's plenty of data out there.

thanks for commenting!

Steven Devijver 04.20.09 at 10:36 am

It just occurred to me that it could be very interesting to re-do this exercise for Susan Boyle's success. My first impression would be that you would end up with slightly less role and some different names for some roles but all in all very much the same.

Thanks for this great blog post.

Mary Walker 04.20.09 at 2:09 pm

Thanks for reading and commenting! Yes, that's my feeling too — that while there are differences in each online crowd action — there are underlying common patterns.

There's been a lot of research done over the decades on the patterns of in-person crowd actions — protests, riots, etc — I hope some PhD student who's studying online behavior will do some more in-depth study of online crowd actions.

U?ur 03.09.10 at 3:49 pm

Hi, Instant communication yields instant opinion. We all have to be aware of that. All the time. Thanks

Comments on this entry are closed.