Hi,

I’m a math enthusiast and self-taught programmer with a passion for finance. Over the years I have observed the market be “surprised” by quarterly earnings reports resulting in big share price swings. Some of these “surprises” have in my mind been predictable, in particular for companies with large “online footprints”. Hence, I started this project where I map out the full “online footprint” of some listed companies. Using this data I build mathematical models to derive the quarterly reported key metrics. I publish these metrics in real time on this project website, in hope of it creating a more efficient market.

To get the project started, my focus has been on Facebook/META which poked my interest as they had very big swings in share price during their last three earnings reports (-24%, +16% and -10%), almost solely driven by surprises in reported user metrics. I currently have some basic models generating daily views on company reported user metrics (download raw data). The methodology is pretty robust though there are some blind spots (Whatsapp) and risk for bias (non-monetizeable users).

After completing Facebook/META, my plan is to move on to Google/SNAP/Pinterest, maybe some of the airlines (DAL, LUV, UAL), and possibly something specialized around Coinbase, Shopify or Airbnb. This is a passion project of mine, and I would love to get in contact with anyone finding this interesting. I’m very eager to get feedback, or maybe even a partner to help drive this project.

Thanks, and I hope you find this useful!

/b3

PS. To explain why these “surprises” still exists, it’s worth noting that most alternative data providers only use broadly applied techniques and do not go deep into company specific data gathering, and therefore are not able of making assertions avoiding the above-mentioned surprises for Facebook/META in 20222Q1 and 2022Q2. There are more specialized alternative data providers (e.g. Yipit, MScience, ThinkNum), but they mainly aggregate data sources in order to sell company specific analysis rather than themselves doing primary deep company-specific data gathering.

PS2. Also worth noting is that the recent Supreme court decision enables deeper and less restraint data gathering - previously arguable not pursuable.