Analytics
494 bites tagged Analytics — interview questions with model answers, and 60-second explainers.
Feature Adoption Rate: Measuring if New Features Deliver Value
Feature adoption rate measures if users actually use specific features, not just log in. It's vital for SaaS products to prove ongoing value for renewals. The footgun: a low rate means customers pay for unused bloat, which actively hurts perceived value and…
Activation Rate: Measuring the 'Aha!' Moment
Activation rate measures the percentage of users who experience your product's core value, not just sign up. It's a key metric for diagnosing onboarding effectiveness. The common mistake is tracking 'completed onboarding' instead of the 'aha!'
Databricks: The Unified Platform for Data and AI
Databricks unifies your data warehouse and data lake into a single 'Lakehouse' platform. It's used for building ETL pipelines, training ML models, and running BI queries on the same data. The main footgun is cost: its power can lead to surprise bills.
Apache Spark: A Unified Engine for Big Data
Think of Apache Spark as a general-purpose engine for large-scale data analytics. It lets you program an entire cluster of machines as one, automatically handling data parallelism and fault tolerance so you can focus on the analysis itself.
Snowplow: A Private Pipeline for Event Data
Think of Snowplow not as an analytics tool, but as a private pipeline you own for creating high-quality event data. It collects raw events, validates them against schemas, and loads them into your warehouse. The footgun is expecting turnkey dashboards.
Segment: The Universal Translator for Customer Data
Segment is a universal translator for customer data. Track an event once in your app, and Segment forwards it to all your marketing and analytics tools, saving you from building dozens of separate integrations.
dbt: Managing Data Transformations as Code
dbt treats your data transformations as a software project, letting you build, test, and version control your SQL. It's the 'T' in the modern ELT paradigm. Use it to create reliable data models in a warehouse. The footgun: dbt only transforms data.
Amplitude: Analytics for Understanding User Behavior
Think of Amplitude as a DVR for user actions, not just a traffic counter. It tracks what users *do* inside your app, letting you build funnels and segment users by behavior.
Google BigQuery: A Serverless Data Warehouse
Think of BigQuery as a massive SQL database you don't have to manage. It's a serverless data warehouse for analyzing huge datasets, like terabytes of application logs. The footgun is treating it like a regular database for real-time transactions.
Jupyter Notebooks: Interactive Code Sandboxes
Jupyter Notebooks are digital lab notebooks for running code, seeing output, and writing notes in one place. Data scientists use them for exploration, visualization, and prototyping.
Google Analytics 4
Google Analytics is a service for tracking user activity across websites and mobile apps. It helps measure marketing performance by tracking traffic and user 'events'. The footgun is focusing on raw traffic instead of the events that signal valuable actions.
Workforce Analytics: Data-Driven People Decisions
Workforce Analytics applies systematic data analysis to people-related decisions, moving beyond gut feelings for hiring and promotions. It's used to predict turnover or measure training ROI.
Business Analytics vs. Business Intelligence
Business Analytics predicts the future and prescribes actions, while Business Intelligence describes the past. BI reports last month's sales; BA forecasts next month's demand.
Marketing Mix Modeling (MMM): Isolating Marketing's Impact on Sales
Marketing Mix Modeling (MMM) statistically links aggregate marketing efforts to sales outcomes over time. It's used to determine the ROI of past campaigns, like a TV ad blitz. The main footgun: the model is only as good as the historical data you feed it.
Sales Velocity: How Fast Your Pipeline Makes Money
Think of sales velocity as the speedometer for your revenue engine, measuring how quickly your pipeline generates money. Sales leaders use it to forecast revenue and find bottlenecks.
Customer Lifetime Value (CLV): A Customer's Total Worth
Customer Lifetime Value (CLV) predicts the total net profit a customer will generate, not just a single sale's revenue. It's used to set acquisition budgets and guide retention efforts. The footgun is using revenue instead of profit, leading to overspending.
DAU/MAU Ratio: Measuring Product Stickiness
The DAU/MAU ratio measures product stickiness by comparing daily to monthly users. Social media apps aim for high ratios (50%+), while e-commerce expects lower ones. The footgun is comparing ratios without context; a 'good' number varies by product type.
Customer Retention Rate: Your Leaky Bucket Metric
Customer Retention Rate measures how many customers you keep over a period, showing how "leaky" your business's bucket is. It's vital for subscription services and e-commerce to gauge loyalty and predict revenue. A high overall rate can hide dangerous churn.
Net Promoter Score (NPS): A Snapshot of Customer Loyalty
NPS distills customer loyalty into one number by asking, "How likely are you to recommend us?" It's a quick pulse check for product teams. The footgun is treating the score as a diagnosis; it tells you *what* customers feel, but not *why*.
Customer Acquisition Cost: What It Costs to Win a Customer
Customer Acquisition Cost (CAC) is the total price you pay in sales and marketing to get one new customer. Businesses use it to see if their model is viable by comparing it to customer lifetime value (LTV). The footgun is forgetting to include all costs.
Ensemble Forecasting: Predicting with a Crowd of Models
Instead of one 'best guess,' ensemble methods generate many forecasts to map the range of possibilities. This is crucial for complex systems like weather prediction, where a single model is misleadingly precise.
Granger Causality: Forecasting, Not Causing
Granger Causality tests if one time series can forecast another, not if it causes it. It's used in econometrics to see if money supply changes predict inflation. The footgun is the name itself: it only shows predictive power, not true cause-and-effect.
ARIMA: Forecasting by Modeling Autocorrelation
ARIMA models forecast a time series by learning its "memory"—how past values influence the next. It's used for forecasting sales or server load where patterns are driven by internal dynamics.
Prophet: Automated Time Series Forecasting
Prophet treats forecasting like decomposing a signal, breaking a time series into trend, seasonality, and holiday effects. It excels at predicting business metrics with strong seasonal patterns, like web traffic.
Get Analytics bites daily.
Five a day, five minutes, offline. With quizzes so it sticks.
Open testing — you’ll join as an early tester.