Privacy Notice
Effective 20 October 2026
This notice covers www.scurvedata.com and the Forecasting workflow at /forecasting-agent/. S-Curve Data records how the product is used so it can tell whether the product works: whether forecasts succeed, how long they take, and whether people who try the workflow get a result. It is not used for advertising and is never sold.
What we collect
- Pages viewed on this site, and the hostname of the site that referred you (never the full address).
- How each visit arrived: the campaign tags in the link you followed (
utm_source,utm_mediumandutm_campaign), if it had any. - The country you are visiting from, as a two-letter code. Our hosting provider works this out from your IP address when the page loads; only the country code is passed to us, and your IP address is not recorded with these events.
- Opening the Forecasting workflow, and whether you chose a demo dataset or your own.
- Forecast runs: the method, whether the run completed or failed, how long it took, and a general error category if it failed.
- Viewing a forecast result.
- In the Experimentation workflow: opening your workspace, moving between its workflow steps (overview, design, health, analysis, decisions), starting a case (a demo, a template or an import, and which built-in demo or template), approving a plan, generating a design, interim or final report and its overall status, recording a decision (ship, hold, stop and so on), and downloading a report.
- In the Causal Inference workflow: opening the workspace, starting with a demo, an uploaded file or a generated dataset (and which built-in demo), reaching each stage of the workflow (understand data, define question, check design, estimate effect, interpret and export), and downloading a report.
- Whether each event came from an ordinary browser or from an automated client, such as a search-engine crawler. This is worked out from your browser's user-agent string when the event arrives; the string itself is not kept.
Identifiers
Your browser is given a random ID and a session ID, kept in local storage under keys starting
scurve.telemetry.. They are not cookies, and they are not linked to your name or email. When you
sign in to upload data, your workspace is recorded only as a one-way keyed hash, not as your email address.
What we do not collect
- The contents of data you upload, or the forecasts produced from it. In the Experimentation workflow, also not the names of your experiments, their metrics or results, or the reasons you write for a decision. In the Causal Inference workflow, also not your file names, column names, the questions you ask or the estimates produced.
- Your email address, error messages, or anything you type.
- Your browser's user-agent string, or your device or browser version.
- Your IP address with these events. Our hosting provider's request logs record IP addresses for security and operations, and keep them for about 30 days.
How long it is kept
Individual events are deleted automatically after 90 days. Aggregate figures computed from them, such as a daily forecast success rate, may be kept longer; they do not identify a browser.
Where it is stored
Events are stored in a database on Google Cloud in the United States.
Google Analytics
This site also uses Google Analytics, which sets cookies. See how Google uses information from sites that use its services. Google offers a browser add-on to opt out.
Your choices
Clearing this site's data in your browser removes the random IDs; blocking local storage for this site stops them being kept. To ask about the data or have it removed, use the contact form.