5 ms·

anecdotal evidence: many of the bad decisions we've made have been data-driven. in other words, we should have listened to our guts more. i think the challeng

anecdotal evidence: many of the bad decisions we've made have been data-driven. in other words, we should have listened to our guts more. i think the challenge is knowing when to use data, and when to use gut. either way, making decisions is very hard. it's basically predicting the future based on flawed information about the past.

acbabis 10y ago
> many of the bad decisions we've made have been data-driven The author of the article would probably say that people's preconceptions influence their data collection in these cases.
- beachstartup 10y ago
  yeah, sure. i mean, how meta do you want to get? at some point you have to draw a line between data and feelings. you can't say that data is really just feelings, so everything is feelings.
afarrell 10y ago
You're still making a couple gut decisions when you use data: - This data was collected in a representative way. - The problem we are addressing is the same as the one this data investigates. The first one assessment is a pretty big one, given the possibility of survivorship bias and the silence of users who are only mildly satisfied/annoyed.
- mamcx 10y ago
  Is like: https://en.wikipedia.org/wiki/Hacker_koan https://en.wikipedia.org/wiki/Hacker_koan So Sussman began working on a program. Not long after, this odd-looking bald guy came over. Sussman figured the guy was going to boot him out, but instead the man sat down, asking, "Hey, what are you doing?" Sussman talked over his program with the man, Marvin Minsky. At one point in the discussion, Sussman told Minsky that he was using a certain randomizing technique in his program because he didn't want the machine to have any preconceived notions. Minsky said, "Well, it has them, it's just that you don't know what they are."
TeMPOraL 10y ago
A different anecdotal take - from my experience, the problem isn't in knowing when to use data (the answer is IMO "always, if you have it and can afford it") - it's in how to use data. The way I see it typically used is only as an argument to convince yourself you're right all along. You know, people proclaiming "the A/B test shows it works!", without any understanding of what "A/B test", "showing" and "working" even means.
tensor 10y ago
The article wasn't arguing against being data driven, it was warning how you can fool yourself into thinking you are being data driven when in fact you are simply following intuition.

acbabis 10y ago

> many of the bad decisions we've made have been data-driven The author of the article would probably say that people's preconceptions influence their data collection in these cases.

beachstartup 10y ago
yeah, sure. i mean, how meta do you want to get? at some point you have to draw a line between data and feelings. you can't say that data is really just feelings, so everything is feelings.

afarrell 10y ago

You're still making a couple gut decisions when you use data: - This data was collected in a representative way. - The problem we are addressing is the same as the one this data investigates. The first one assessment is a pretty big one, given the possibility of survivorship bias and the silence of users who are only mildly satisfied/annoyed.

mamcx 10y ago
Is like: https://en.wikipedia.org/wiki/Hacker_koan https://en.wikipedia.org/wiki/Hacker_koan So Sussman began working on a program. Not long after, this odd-looking bald guy came over. Sussman figured the guy was going to boot him out, but instead the man sat down, asking, "Hey, what are you doing?" Sussman talked over his program with the man, Marvin Minsky. At one point in the discussion, Sussman told Minsky that he was using a certain randomizing technique in his program because he didn't want the machine to have any preconceived notions. Minsky said, "Well, it has them, it's just that you don't know what they are."

TeMPOraL 10y ago

A different anecdotal take - from my experience, the problem isn't in knowing when to use data (the answer is IMO "always, if you have it and can afford it") - it's in how to use data. The way I see it typically used is only as an argument to convince yourself you're right all along. You know, people proclaiming "the A/B test shows it works!", without any understanding of what "A/B test", "showing" and "working" even means.

tensor 10y ago

The article wasn't arguing against being data driven, it was warning how you can fool yourself into thinking you are being data driven when in fact you are simply following intuition.