Forecasting
Time Series Analysis
A statistical method that analyzes sequential data points (ad spend, revenue, conversions) over time to identify trends, seasonal patterns, and cyclical behavior. The foundation of most forecasting models. Common techniques include ARIMA, Prophet, and LSTM neural networks. For ecommerce advertising, time series analysis reveals hidden patterns like the lag between Meta spend increases and Shopify revenue impact, or the creative fatigue cycle for video ads.
Why it matters
Marketing data is ordered and autocorrelated, so methods that assume independent observations give confidently wrong answers on it.
In practice
Decompose before you model: separate trend, seasonality and residual. Most of what looks like a performance signal is one of the first two.
Related terms