Dogged determination is a great skill to have, especially doing something difficult that has no “right” answer (e.g. how do I create a new piece of software that a million people will use, enjoy and pay for!?). But there comes a time when either 1) people clearly have misunderstood what you said and it becomes clear there is no way your point can be understood easily and/or 2) you said something careless not supported by real data and/or 3) nobody cares about your data because it is far too limited in scope. Then you may just need to confess that you made a mistake.
We are very far from understanding the nonlinear ways in which human hearing and cognition work. What is clear though, is that human beings are really good at picking out voices from noise and in understanding speech. A great paper about this, The Weckud Wetch of the Wast: Lexical Adaptation to a Novel Accent says that:
listeners can adjust their interpretation of altered phonetic forms when listening to a talker who speaks a different dialect. Here we document that these adjustments can be made within 20 min of listening experience, as assessed by performance on a lexical decision task.
There is also lots of evidence to suggest that once you get familiar with one kind of dialect or accent, you can far more quickly pick up novel speakers using the same in future. So if you are in a situation where, for example, you have (a) never experienced that specific kind of accent before, and/or (b) only getting a very small dose of this accent, say a 2 and a half minute presentation or even a 45 minute pitch, you may miss some of it. But wait, that sounds like the kind of encounter someone would have with a founder or CEO of a small company looking to get funded (very short, infrequent interactions). Although it looks like if the person listening to these pitches gets more familiar with those classes of accents by spending more time LISTENING to them, they’ll have a much easier time understanding them. Certainly people with heavy accents can work to make themselves more understandable. I know that; many people love the South African accent many Saffas like I have but there are certain terms and words I’ve consciously come to avoid since they won’t make sense to most. But I’ve also spent a lot of time working with people with accents and I have little problem understanding Southern American accents (e.g. Georgia!), Russian, Chinese, Indian and other accents. Listen more to a more diverse set of accents, get better results.
With all the talk about “big data”, startups in general, but especially the business of funding startups is the world of small data. I spoke to a well-respected VC a few weeks ago that has done two deals in four years. Many VC firms do only a few deals a year as a group. Most of them won’t talk to a potential startup unless referred by someone they trust. There are plenty of rational reasons for this, but clearly this “how do I build a little company”, small number of data points, working-with-other-people-from-different-backgrounds stuff is hard and unpredictable. Paul Graham and other VCs who write and work hard to try to educate about fundraising and startups have lots of great advice to dispense, but like all of us these people make mistakes too. No blogger, VC or commentator deserves an automatic pass – and any data these people have to share should be scrutinized and not accepted on faith.
One of the biggest problems many smart people have, is that far too often their need to BE right outweighs the need for them to DO right. It’s great when the two are the same thing, but far too often they are simply not.
