Common questions about causal analysis, methods, and CausalLens.
Causal inference estimates whether one thing causes another — not just whether they correlate. For example: does a drug cause lower blood sugar, or do healthier patients simply take it more often? Causal methods adjust for confounders to answer the cause question.
Yes. Go to Account and click Cancel Pro subscription — one step, no phone call. Your account stays on the Free plan. When PayPal/Stripe billing is live, cancellation stops future charges.
Guests can try sample datasets free. New accounts get a 14-day Pro trial (no credit card) with Pro limits — up to 10,000 rows per file, all methods, exports, and AI summaries. After that, Free allows up to 200 rows and 2 methods per run; Pro is $49/month for up to 10,000 rows and unlimited methods.
For developers, DoWhy and EconML are excellent Python libraries. For non-technical users, CausalLens wraps those libraries in a guided wizard with plain-English output, balance checks, and reports — with a generous free trial.
CSV and Excel files with column headers. You need a treatment column (intervention), an outcome column, and optionally confounder columns.
The 200-row upload cap applies only after your 14-day Pro trial ends and you stay on Free without upgrading. During the trial you have the same 10,000-row limit as Pro. Pro supports up to 10,000 rows per file — every populated row is used in analysis (the on-screen table previews up to 2,500 rows for performance).
A method that pairs treated and control subjects with similar probability of receiving treatment, reducing confounding bias in observational studies. Common in medicine and epidemiology.
A quasi-experimental method comparing how outcomes change over time between a treated group and a control group. Widely used in policy evaluation.
Uplift modeling estimates the incremental effect of a treatment on each individual — who responds because of the intervention. Essential for marketing campaign targeting.
No tool can prove causation from observational data alone. Causal inference estimates effects under explicit assumptions. CausalLens lists those assumptions and runs robustness checks to help you judge credibility.
When you run CausalLens locally, your data stays on your computer. Optional AI explanations require an OpenAI API key you configure yourself.
causaLens (decisionOS) is an enterprise platform costing $100k+/year. CausalLens is an accessible app for individuals and teams — start with a free trial, then Pro at $49/month.
No. The wizard guides you step by step. The tutorial explains concepts in plain English. Technical users can export Python code for reproducibility.