You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
Forecasting is art, not science. I built a forecasting system that looks at the last 180 days of our sales. It weights all 180 days equally and takes into consideration seasonality. What it doesn't do is weight recent data (last 30 days) more heavily than older data (last 150).pic.twitter.com/OseWnrEVsd
Our best selling goodminton went from being a so so seller with a sales rank in the 5,000 to 10,000 range (lower sales rank is better. If you are #1, you are the best seller in your category) to #345 and lower. That's a MASSIVE increase in sales.
Unfortunately, because I'm a dumbass, I just followed our excel formulas and didn't realize that we should have weighted the more recent data more heavily and now we are going out of stock, resulting in massive lost revenue and another problem I cannot even talk about...grrrrrrr
Usually we have the reverse problem. During the Australian wildfires we were selling tons of Australian animals, but that was clearly a one-off. tl;dr Forecasting is art, not science. You must combine quantitative with qualitative to do a good job.
A most strange coincidence 
https://twitter.com/orrdavid/status/1259968701345837058?s=20 …
THERE IS A GOD.
You could use multiple methods/models and when they diverge, call in the humanpic.twitter.com/hJJvDjVYPd
I understand 50pcs of retail discretionary is tied to holidays esp Christmas etc. How does your model capture this?
Yup, got seasonality in there.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
