4 Factors That Encourage Hashtag Spamming of Twitter Chats


posted by on Marketing, Music, Social Media


I was physically present (IRL, In Real Life) recently at an event where the Twitter hashtag stream was completely co-opted by twitter spambots. I’ve live-tweeted from a half-dozen tech and cultural events since the beginning of this year, when I first immersed myself in Twitter. I’m very curious about how social media interactions work – and when and why they can go very off-track. When I live-tweet, I try to observe the hashtag stream in real time, usually using Tweetchat.com or setting up a Hootsuite stream. I’ve followed a handful of events remotely via the Twitter hashtag as well, including a recent conference in Boston called Rethink Music (#rethinkmusic). In addition, I participate regularly in a weekly Twitter chat called #ggchat, one of thousands happening all the time in the Twittersphere. Following Twitter hashtag streams has become an integral part of my participation, and that of many others, in this virtual global sociological communications experiment called Twitter.

Maybe because I’m relatively new to Twitter, I’ve never seen a Twitter stream completely taken over by spambots. I found it fascinating and dismaying at the same time. This article in The Atlantic Wire by Rebecca Greenfield gives a good overview of some of the different ways in which Twitter hashtag streams can get co-opted or become annoying. The stream I was on recently was taken over by the Types 1 and 2 spammers which Rebecca mentions: Porn Bots and Jokesters. I didn’t click on any of the links; I could tell the Porn Bots by their Twitter avatars of scantily clad women and the fact they had few tweets, no followers and were following no one. The other category of spammers I saw which Rebecca doesn’t mention I’ll call Job Bots – these are the same as Porn Bots, except the links they promote are to scammy Craig’s List ads, you know: “Easy job! Earn $500 a week using your computer…”

The live event for this hashtag was an interview and performance by avante-garde cellist, Zoë Keating. Zoë has a pretty substantial social media presence and has done all her own marketing online to promote her music. Her interview was part of an online series done by Chase Jarvis, a Seattle-based, internationally known photographer, social media maven and ardent supporter of the DIY online artist culture.  As a result of both Zoë and Chase’s strong social media presence, I suspect there were a large number of social-media-savvy people following the event via the Twitter hashtag, both in the US and outside. I don’t know the statistics for this event, but I asked the Chase Jarvis production team a few months ago how many people follow their live video streams, and they said 10,000 or more. That’s a lot of people participating, actively or passively, in the hashtag stream. Chase’s production team was also running a giveaway promotion with a sponsor, Hewlett Packard, for the best tweet. You had to include the event hashtag, #cjlive, the sponsor’s handle @hpprint, and a link to the streaming event, as well as a quote from Zoë, in order to enter the contest. The prize was a signed photo of Zoë taken by Chase just that morning during pre-production – a big deal for an ardent fan of either artist. The more tweets, the more chances to win. To top that off, Zoë’s performance was amazing to watch – so I think there were a lot of these kinds of tweets: “OMG. You gotta watch this… Amazing…”

I believe these four factors conspired to attract the spambots’ attention (likely just an algorithm in their automated keyword search software):

  1. savvy international social media following
  2. large number of participants
  3. real-time contest (combined with #2 to generate a high number of TPMs, or Tweets Per Minute)
  4. great live performance that sparked keywords like “amazing” or “gotta watch”

Once the bots appeared, the Twitter stream became useless to the production team to accomplish their  goal of promoting the sponsor. More importantly, it became useless to anyone wanting to actually participate in the event remotely. Meanwhile, I had spent precious minutes wasted trying to report and block the spammers. I eventually just gave up watching the Twitter stream and focused on the live performance – probably what I should have done in the first place. (Beware, one of the pitfalls of live-tweeting: it can wreck your actual experience of the event.) 45 minutes after the event ended, I checked the hashtag stream again. The bots were gone, and a slew of (greatly delayed) tweets by real people about the event suddenly appeared on the hashtag stream.

In marketing, online or offline, it’s always an arms race between consumers and advertisers. It is only a matter of time in any marketing cycle before the black hat programmers and advertisers find ways to clutter eyeballs with useless and annoying promotions. It is antithemical to the whole idea of social media as an open, frictionless, and transparent platform for communication and participation to have bots inject spam of any kind into the process, but my old-school marketing brain tells me it was inevitable. I’m sure Twitter will eventually get better at blocking the Twitter spambots, as have email providers, but in the meanwhile, this experience has certainly made me think twice about my participation in both live-tweeting events and Twitter chats.

If you’ve had any similar experiences with spam on Twitter chats, please comment below.

Here’s a helpful article on blocking spambots:


Tags: , , ,


    • Solveig

Trackback e pingback

No trackback or pingback available for this article

Leave a Reply