Google Ventures workshop: Lean analytics
HTML-код
- Опубликовано: 10 фев 2014
- Google Ventures | The Lean Startup is all about a constant cycle of learning. By measuring and analyzing as you grow, you can validate whether a problem is real, find the right customers, decide what to build, and how to monetize it. In this workshop, Alistair Croll talks about finding the One Metric That Matters to your startup right now.
- Наука
Really like this guys presentation style. Awesome talk!
This is one of the most informative and useful workshops I've listened to in a very long time.
Is there a place i can download the presentation itself?
Thank you.
This is a great video -- thank you for posting it.
Really good watch. Worth your time if you're interested in measuring web traffic, startup progression and/or care about modern (digital) business analytics.
impressive lectures
Dr. Siddig Ahamed Mohamed
Great presentation. Very insightful and easy to follow!
This was insightful
Thanks for the insightful presentation! Does anyone have other great videos on lean analytics you could recommend??
I didn't realise The Fat Duck was popular over the atlantic?
what is a cohort?
what is google?
Jamiel Cotman Cohort analysis is a kind of Group Analysis
hodor hodor
Subtitles please
Anybody knows the Name of the book he recommend at around min 31:00?
lmao this guy is funny. why is the audience so cold.....
You get the feeling this guy thinks everyone is dumb????
When he says patronizingly, "this is what we call science" -- well, that was not science. Science requires accurate repeatable prediction of hypothesis to form laws or theories, not simply measurement of various arbitrary tests. I think this is what we call pseudo-science.
Someone please give this man some sedatives. He might not have considered the audience experience. I guess.
That statistical boy/girl problem has to be bullshit. I can't explain why. But it has to be wrong.
What about this: I have 8 kids and they are all boys. My wife is pregnant right now. What are the chances that the baby will be a girl?
By his reasoning I suppose the answer is 1 over 8 factorial or something like that (forgive me, I haven't done this in a while). This cannot possible be right. Is it not still 50/50 that it will be a boy or a girl? I swear this is the same as the coin flipping problem. If you flip a coin 9 times and it was heads every time, then what are the chances it will be heads on the tenth flip? 50%.
Am I crazy?
He presented a different problem to what you understood it to be.
His question was if he had two children and at least one of them was a girl, what is the probability that the other child is a boy.
Combinations of two children: BB, BG, GB, GG.
Given that at least one is a girl, BB is excluded.
Clearly, two of the remaining three have a boy giving 2/3 the other child is a boy.
+Luke Hansen I'm genuinely baffled by it. I think he's using a sort of trick of language that involving permutations. Isn't the GB/BG distinction, although a possible combination, irrelevant in terms of chances? If one is a girl, it seems the other one simply has to be a boy or a girl.
BG,GB - they both are the same option. We are not talking about if the B or the G came first, but they are just there. So the options are BG or GG, and therefore the chances are still 50%. But you are right..there is a little bit of word-wrangling going on, and it's all what you hear.
You are absolutely right statistically
You're right, it's bullshit. He misunderstands probability. The sex of each individual child is irrelevant to the sex of another child. They are probabilistically independent and so must be evaluated individually, not as a probability of each event occurring together.
Croll is a little bit abnoxious trying to be smart all the time. he could have said the thing in 10 min.