Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Retail in real time

retale.com

Post reply on HN
Print view View original

1–8 of 8 posts

Re: Retail in real time

#3

Post by bruceb »

What bugs me about these things is that lots of people don't realize this an avg # sold in X time. I have heard people say well it has been X seconds so 1k colas have been sold. Well it is 3am in the morning so no they haven't. At 3pm 10k will sell in that time.

Maybe this is making a distinction with out a difference but I think the short cut of not saying avg leads to many thinking that literately every X seconds an event happens. I have seen so many news reports that use sloppy wording that leads to this kind of thinking.

Re: Retail in real time

#4

Post by maad »

josephlord wrote: post #2

1 in 10 Best Buy customers gets a virus removed??? Firearms spending 2/3 of toys?

I would assume viruses removed would be in bulk batches, so people with dozens of viruses on a single computer. As someone said earlier just taking the X number of things per day and then averaging that out over the day in real time is really misleading.

Re: Retail in real time

#5

Post by lnanek2 »

bruceb wrote: post #3

What bugs me about these things is that lots of people don't realize this an avg # sold in X time. I have heard people say well it has been X seconds so 1k colas have been sold. Well it is 3am in the morning so no they haven't. At 3pm 10k will sell in that time. Maybe this is making a distinction with out a difference but I think the short cut of not saying avg leads to many thinking that literately every X seconds a…

A lot of the time they are using global stats anyway that are pretty averaged out across hours as well. It is 6AM here, yes, and everyone is buying coffee on the way to work, but it is 9PM in Australia and people are buying drinks in bars. It is probably cola time somewhere else, like India.

I don't think that level of detail is needed for this anyway, though. The point is to make it clear a lot of these things are sold, which can then change the way you think. A programmer may never write an app related to baby food, then see how much baby food sells, then realize it could make sense.

Re: Retail in real time

#6

Post by pandatigox »

Do the stats revealed here really reflect the current retail? Or does it just take a stat trend (for example X sold in Y minutes/seconds)and apply a linear equation[1] to "predict" how much consumers would buy?

[1] Something like this?

   f(t) = m*t + c

Re: Retail in real time

#7

Post by bruceb »

lnanek2 wrote: post #5
bruceb wrote: post #3

What bugs me about these things is that lots of people don't realize this an avg # sold in X time. I have heard people say well it has been X seconds so 1k colas have been sold. Well it is 3am in the morning so no they haven't. At 3pm 10k will sell in that time. Maybe this is making a distinction with out a difference but I think the short cut of not saying avg leads to many thinking that literately every X seconds a…

A lot of the time they are using global stats anyway that are pretty averaged out across hours as well. It is 6AM here, yes, and everyone is buying coffee on the way to work, but it is 9PM in Australia and people are buying drinks in bars. It is probably cola time somewhere else, like India. I don't think that level of detail is needed for this anyway, though. The point is to make it clear a lot of these things are s…

Well in this case they say US retail and it does say real time hence one might believe that yes this is actually happening right now. Best Buy employees don't work at 3am unless it is BF.
Post reply on HN