General

A practical, non-legal guide to figuring out how long to keep business data, from raw event logs to old customer records, and what actually breaks when you keep everything forever.

When you cannot run an A/B test, before and after comparisons are the fallback. Here is how to run one honestly, what it can tell you, and where it lies to you.

How to tell when a cohort table’s retention ‘improvement’ is a mirage caused by data lag, cohort size, or definition drift, not a real gain.

A plain-English walkthrough of how to size a market using free public data, without paying for a report or inventing numbers to sound confident.

A practical guide to picking the right date field for churn, MRR, cohorts, and attribution when your system tracks four different timestamps for the same event.

A plain English guide to tracking trial to paid conversion when your sales cycle runs weeks or months, including cohort tables, censoring, and what the number can’t tell you.

A practical framework for picking the handful of numbers that actually deserve a spot on your weekly metrics report, and why more metrics usually means less clarity.

A plain-English breakdown of rolling cohort analysis: what it catches that monthly cohorts miss, a small business example, and where the monthly view is still the better tool.

A solo operator’s guide to lead scoring: what it is, how to build a simple point system, and where it breaks down if you trust it too much.

A plain-English guide to deciding how much time, traffic, and money to spend on A/B tests when your business does not have enough users to make testing cheap.