The two points of contact theory
Here’s a hypothesis: It only takes two points of contact for people to validate a reference.
If I tell you Flogging Molly is the best live band ever, you’d nod your head politely but pay me no mind. But if on the same day, or same week, another aquiantance of yours, one completely indepenent from me, mentions Flogging molly, you will instantly validate both sources. I think we assume that random, unsolicited references that correlate with each other must be true. And we only need two of them to place our confidence in recommendations.
So I think the word of mouth effect is really about two points of contact. The first time we hear about something it’s not far from noise, but the second time, especially if the sources are diverse enough, we’re ready to take action. Should we hear a third mention of whatever it is, we’d probably say “Oh. Flogging molly. Yes, I’ve heard lots of good things about them.” Even though we’ve only had two previous points of contact, possibly from sources we’ve never relied on before.
I’m looking to see if anyone has ever studied how people make recommendations and how often they’re based on small amounts of second hand data from third party sources. I’m familiar with The madness of crowds but this two points of contact idea is about more pedestrian things.
Neal Stephenson’s Crytonomicon has been a very compelling study for me on this topic. In many ways it is about our search for order in the midst of chaos.
To quote —
Oh yeah — and of late, Episode 19 of the first season of Lost is called ‘Numbers’ and it is about much of the same thing — I just about cried when I watched Hurley struggle to verify the pattern he was seeing — nice cinematic comedic drama approach to discussing the same question.
And we probably shouldn’t forget that this discussion about pattern recognition has been around for thousands of years — as evidenced in this piece of antient literature —
“For since the creation of the world God’s invisible qualities — his eternal power and divine nature — have been clearly seen, being understood from what has been made, so that men are without excuse.”
Scott,
I am a huge fan of yours. Not sure 2 exposures are enough. You may want to look up the Mere Exposure Effect studied by Robert Zajonc. Wikipedia link (http://en.wikipedia.org/wiki/Mere_exposure_effect). Also lookup the field of Cognitive Lingustics. I had written a post on Cognitive Linguistics with links to Zajonc (Sorry for the shameless self plug) – Permalink http://sastwingees.blogharbor.com/blog/Management/_archives/2005/7/31/1094903.html
Thanks
Sukumar
Someone e-mailed me and refered also to Gladwell’s blink and how it covers situations where instant or instinctive responses to things are sufficent for action.
In reading the comments so far it’s obvious it depends on the situation – I used the music band example since it’s such a transient kind of recommendation. It’s not something you’re dependent on, but it is something you’re always looking for good recommendations for. So when you happen to hear the same thing mentioned, it’s enough to validate it. If it were something with more at stake, say, a possible spouse, a city to live in, a religion, it would take more investment before someone would be willing to make a decision.
But I’m surprised at how infrequently people are asked for references and how rarely references are contacted. We put a lot of trust in the opinions of people we don’t know.
I think the first of Gladwells books (The tipping point) is more relevant. One of the the themes is exactly how people get to choose different brands for things. He postulates three different kinds of people who are involved: The “mavens”, the “connectors” and the “salespeople”. The mavens are experts for certain things and other people trust their recommendations because of their expertise (its not only expertise but a certain mindset, wanting to help people make good decisions). These recommendations get picked up by salespeople and connectors who spread the word.
I found it very interesting to read.