Analytics glossary
Short, opinionated definitions of the analytics terms that get thrown around. Each entry has the canonical definition, when to use it, when not to, and the alternatives. Written by Dr Waqas Rafique — PhD in Statistical Machine Learning, prior research at Cambridge, Oxford and UCL.
Holt-Winters
Time series with both trend and seasonality — e-commerce GMV, SaaS MRR, energy demand, agency monthly billings. At…
RFM segmentation
Use percentile-based bucketing (PERCENTRANK in Excel) so each score 1-5 contains roughly 20% of customers. Equal-width b…
Cohort retention
A 95% monthly retention rate sounds great. But if your last cohort is retaining at 80% while older ones retain at 99%, t…
Anomaly detection
Z-score threshold of 2.5 corresponds to ~1.2% false-positive rate; 3 corresponds to ~0.3%. Use 2.5 for sensitive monitor…
KPI dashboard
A lagging indicator tells you what already happened (revenue, churn, active users). A leading i…
Related
Each glossary entry links through to a longer tutorial with worked Excel examples. For a full landscape of the tools that use these methods, see the 2026 comparison of AI tools for Excel.