The estimates use a layered model. For each platform you selected, we calculate a baseline footprint (the data stored about you when you create an account) and add a daily generation rate multiplied by the number of years you have plausibly used that platform.
The years figure is bounded by three things: how long you have been online, how long the platform has existed, and a typical adoption lag for your age cohort. A 55-year-old user who joined TikTok almost certainly did not join in 2018, so the model adjusts accordingly.
The daily generation rate is adjusted by a demographic-fit factor reflecting how intensively your age and gender cohort typically uses each platform, drawing on published research from Pew, Datareportal, GWI and platform transparency reports. Your stated overall usage intensity modulates this baseline but does not override it: a self-described light user from a heavy-usage demographic still generates more data than a self-described heavy user from a light-usage demographic.
A small country-level adjustment accounts for differences in digital economy maturity. A modest random variation is added so two similar users receive slightly different numbers, reflecting genuine individual variation.
A "data point" refers to any discrete piece of information a platform stores or derives about you: a single search query, a play event, a location record, a like, a scroll duration, an inferred attribute. The figures represent order-of-magnitude estimates, not precise audits. Your actual footprint is likely higher.
The breakdown by category (Identity, Behavior, Interests, Connections, Location) is derived from a per-platform distribution model. Each platform's footprint is divided across the five categories using fixed proportions reflecting that platform's primary function. Spotify and Netflix lean toward Interests and Behavior. Meta and LinkedIn lean toward Connections. Google and Apple distribute more evenly across all five. These proportions are informed estimates based on platform documentation, research into platform data practices, and reasonable inference about how different data types are generated on each service.
Brand icons are sourced from Simple Icons, an open-source project, and remain the trademarks of their respective owners. Their inclusion here is purely descriptive and does not imply any partnership, sponsorship or endorsement.
If you join the waitlist, your email address and questionnaire answers are stored securely and used solely to notify you when oMemory launches. We do not sell, share or transfer this information to third parties for any purpose.