What Is FRED and Why Should Marketers Care
FRED, the Federal Reserve Economic Data system, is a free public database maintained by the Federal Reserve Bank of St. Louis. It contains over 800,000 time series covering everything from money supply and interest rates to employment and GDP. Economists and traders use it daily. Marketing teams almost never touch it.
That is a missed opportunity. Every macroeconomic shift that moves ad auction prices, consumer confidence, and media costs is published in FRED weeks or months before it shows up in a marketing dashboard. The data is free, updated regularly, and programmatically accessible through the FRED API.
The Five FRED Series Every Marketing Team Should Monitor
You do not need all 800,000 series. Five are enough to build a working model of how macroeconomic conditions affect marketing performance:
- M2SL (M2 Money Supply): The broadest measure of money in the economy. When M2 grows, more dollars chase the same goods and services. Consumer spending rises. Ad engagement improves. When M2 contracts, the opposite happens.
- WALCL (Fed Total Assets): The Federal Reserve's balance sheet. Quantitative easing expands this. Quantitative tightening shrinks it. Changes here move bond markets, which move corporate earnings expectations, which move ad budgets across every industry.
- DGS10 (10-Year Treasury Yield): The benchmark risk-free rate. Rising yields mean tighter financial conditions. Falling yields mean easier conditions. This series has a documented correlation with quarter-over-quarter changes in total US advertising expenditure.
- UNRATE (Unemployment Rate): Employment drives consumer spending. But the unemployment rate is lagging. Watch initial claims (ICSA) for a leading signal of where unemployment is heading.
- UMCSENT (Consumer Sentiment): The University of Michigan sentiment index. Sentiment leads spending, which leads ad engagement. A sustained decline in sentiment predicts lower ROAS 4 to 8 weeks later.
How to Access the Data
FRED data can be accessed three ways. The simplest is the FRED website (fred.stlouisfed.org), where you can view charts and download CSV files. For automated workflows, the FRED API provides JSON and XML endpoints. You will need a free API key, which takes about 30 seconds to obtain from the FRED website.
For Python users, the fredapi package wraps the API in a clean interface. A basic pull of M2 money supply data takes three lines of code: import the library, authenticate with your key, and call get_series with the series ID. The data comes back as a pandas Series ready for analysis.
From Raw Data to a Decision Framework
Raw FRED numbers are not directly actionable. M2 at $21.2 trillion does not tell you whether to increase or decrease paid social spend. You need to compute rates of change and compare them to historical ranges.
The practical approach:
- Compute 4-week and 13-week percentage changes for each series.
- Convert those changes to percentile ranks relative to the trailing 5-year window.
- Combine the percentile ranks into a composite score weighted by predictive power.
- Map the composite score to regime thresholds (high expansion, moderate growth, neutral, contraction).
This is the calculation WhenBRRR runs automatically against 15 FRED series. The output is a single regime classification with confidence scoring and channel-level budget recommendations. The math is documented and deterministic: the same inputs always produce the same output.
What Marketers Typically Discover
Teams that start monitoring FRED data report three consistent findings. First, they gain 4 to 6 weeks of lead time on ad cost changes. Second, they can explain budget recommendations to finance teams using the same economic data that finance already watches. Third, they stop making reactive cuts because they have a framework for anticipating changes.
The hardest part is not the data. It is changing the organizational habit of making budget decisions based solely on last month's performance metrics. Once a team adds leading economic indicators to its planning process, the improvement is measurable and compounding.