Data, marketing, and analytics glossary
Here you find terms that appear in the GinnyTech course, explained practically. Use it when you encounter an acronym, a metric, or a technical concept and want to quickly understand its purpose.
How to use it
- Start from the short definition.
- Look at the context: it tells you in which area of the course you will find that concept again.
- When a term seems abstract, always link it to a concrete question: what does it measure, what changes, or what decision does it help to make?
Terms A-Z
A
A/B Testing Experimental method to compare two versions of a variable (webpage, email, feature) to determine which performs better against a specific goal. It is the gold standard for establishing causality. Context: Product Analysis / Marketing Data Science
ACID (Atomicity, Consistency, Isolation, Durability) Fundamental properties of transactions in relational databases (and some modern Data Lakes) that guarantee data validity even in case of errors or blackouts. Context: Data Warehousing / S3 Data Lake
Actionable Metrics Metrics that link specific actions to observable results and help make concrete decisions. They contrast with Vanity Metrics. Context: Data Driven Management
Activation Rate Percentage of users who reach the “Aha!” moment or complete a key action indicating they have obtained value from the product for the first time. Context: Product Analysis / Marketing Analytics
ARPU (Average Revenue Per User) Metric indicating the average revenue generated by each active user in a given period. Essential for calculating LTV. Context: Fundamental Metrics / Marketing Data Science
ARR (Annual Recurring Revenue) The annualized value of recurring contracts. It is the main metric for evaluating SaaS companies. Context: Fundamental Metrics
Athena (AWS) Serverless query service that allows analyzing data directly in Amazon S3 using standard SQL. Context: S3 Data Lake
Attribution Models Rules that determine how credit for a sale or conversion is assigned to various touchpoints in the purchase journey (e.g., Last Click, First Click, Data-Driven). Context: Marketing Data Science
B
Backpressure Mechanism in streaming systems to handle situations where the producer sends data faster than the consumer can process it, slowing the flow to avoid crashes. Context: Real-time Analytics / Kafka Streaming
Backward Compatibility Ability of a system (or data schema) to operate with inputs generated by previous versions. Crucial in Kafka to allow schema evolution without breaking existing consumers. Context: Kafka Streaming (Schema Registry)
Bounce Rate Percentage of visitors who leave the site after viewing only one page. A high bounce rate may indicate irrelevant content or technical issues. Context: Marketing Analysis
Broker (Kafka) Single server within a Kafka cluster responsible for receiving, storing, and distributing messages. Context: Kafka Streaming
C
CAC (Customer Acquisition Cost) Total cost incurred to acquire a new paying customer (marketing + sales expenses / new customers). Must be balanced with LTV. Context: Fundamental Metrics / Marketing Data Science
Card Sorting UX Research technique where users are asked to organize topics into groups. Useful for designing information architecture and navigation menus. Context: Product Analysis (UX Research)
CDC (Change Data Capture) Technique to identify and capture changes made to data in a database (insert, update, delete) to replicate them in real time to other systems (e.g., via Debezium to Kafka). Context: Real-time Analytics / Data Collection
CDP (Customer Data Platform) Software that aggregates and organizes customer data from various sources (web, mobile, email, CRM) to create a unified and actionable user profile. Context: Data Collection / Marketing Analysis
Churn Rate The percentage of customers or subscribers who stop using the service in a given period. It is the opposite of Retention. Context: Fundamental Metrics / Product Analysis
ClickHouse Open-source columnar OLAP database designed for real-time analysis on large volumes of data. Famous for its speed in executing aggregate queries. Context: Real-time Analytics / Data Warehousing
Cohort Analysis Behavioral analysis method that groups users based on common characteristics shared within a time interval (e.g., signup date) to observe how their behavior evolves over time. Context: Product Analysis / Marketing Data Science
Columnar Storage Data storage format (e.g., Parquet, ClickHouse) that saves data by column instead of by row. Optimized for analytical queries reading few columns of many rows. Context: Data Warehousing / S3 Data Lake
Consumer Group (Kafka) Group of consumers collaborating to read messages from a Kafka topic. Ensures scalability and fault tolerance in data reading. Context: Kafka Streaming
Conversion Rate (CR) Percentage of users who complete a desired action (e.g., purchase, signup) compared to total visitors. Context: Marketing Analysis / Fundamental Metrics
D
Data Contract Formal agreement between data producers and consumers defining schema, semantics, SLA, and data evolution rules, treating data as a product. Context: Kafka Streaming / Data Engineering
Data Lake Centralized repository that allows storing all structured and unstructured data at any scale. Data is saved in the original raw format. Context: S3 Data Lake
Data Warehouse System used for reporting and data analysis, considered the central component of Business Intelligence. Data is typically structured, cleaned, and optimized for SQL queries. Context: Data Warehousing
DAU/MAU Ratio Ratio between daily active users and monthly active users. Measures product “stickiness,” i.e., how often users return. Context: Fundamental Metrics / Product Analysis
Dead Letter Queue (DLQ) Special queue where messages that cannot be processed correctly are sent, allowing later analysis without blocking the main flow. Context: Kafka Streaming
Debezium Distributed open-source platform for Change Data Capture (CDC). Converts transaction logs from existing databases into Kafka event streams. Context: Real-time Analytics
Dimensions Qualitative attributes that describe data (e.g., “Country,” “Browser,” “Product Category”) and allow segmenting and filtering metrics. Context: Data Driven Management / Data Warehousing
E
ELT (Extract, Load, Transform) Modern variation of ETL where data is loaded into the target (e.g., Data Warehouse or Lake) before being transformed. Leverages the computing power of the target (e.g., BigQuery, Snowflake). Context: Data Warehousing / Data Collection
Embeddings Vector representations of words or concepts in a multidimensional space, used in Machine Learning to capture semantic meaning and similarity. Context: Marketing Data Science
ETL (Extract, Transform, Load) Traditional data integration process that extracts data from sources, transforms it on an intermediate server, and loads it into the target system. Context: Data Warehousing
Event Properties
Specific metadata associated with a single tracking event (e.g. price o product_name for the event Purchase).
Context: Data Collection / Product Analysis
Exactly-Once Semantics Guarantee in streaming systems that each message will be processed exactly once, without duplicates or losses, even in case of failures. Context: Kafka Streaming
F
First-Party Data Data collected directly by the company from its own customers (website, app, CRM), for which it has consent and ownership. Context: Marketing Analysis / Privacy
Funnel Analysis Analysis that maps the steps a user must take to reach a goal, identifying where drop-offs occur. Context: Product Analysis / Marketing Analytics
G
GDPR (General Data Protection Regulation) European regulation on data protection and privacy imposing strict rules on the collection, consent, and management of personal data. Context: Data Collection / Privacy
Gestalt Principles Visual psychology principles (e.g., proximity, similarity) explaining how the human brain organizes visual elements into unified groups or shapes. Fundamental for Data Visualization. Context: Dashboard Visualization
Glue (AWS) AWS serverless data integration service that facilitates data discovery, preparation, and combination for analysis (includes Data Catalog and ETL). Context: S3 Data Lake
Granularity The level of detail of stored data. Fine granularity (e.g., every single event) allows for deeper analysis compared to aggregated data. Context: Data Warehousing
H
HADI Cycles (Hypothesis, Action, Data, Insight) Iterative cycle used in Growth Hacking and Agile to validate hypotheses through rapid experiments and data analysis. Context: Marketing Data Science / Data-Driven Management
Heatmap Graphical representation of data where values are shown by colors. In UX, it shows where users click, scroll, or move the mouse. Context: Product Analysis (UX Research)
Hive Style Partitioning
File organization strategy in the Data Lake where directory paths include partition column names (e.g. year=2023/month=05/).
Context: S3 Data Lake
I
IaC (Infrastructure as Code) Practice of managing and provisioning IT infrastructure through machine-readable definition files (e.g., Terraform) rather than manual configuration. Context: Infrastructure Ops
Iceberg (Apache Iceberg) Open table format for huge analytical datasets that brings SQL features (transactions, schema evolution) to files in the Data Lake. Context: S3 Data Lake
Idempotence Property of an operation that can be applied multiple times without changing the result beyond the initial application. Essential in data pipelines to handle retries. Context: Kafka Streaming / Data Engineering
J
JTBD (Jobs To Be Done) Theory that customers don’t buy products but rather “hire” them to perform a specific job or solve a problem in their lives. Context: Product Analytics / Data-Driven Management
K
Kafka Connect Component of the Apache Kafka ecosystem to connect Kafka with external datastores (databases, key-value stores, search indexes, file systems) in a scalable and reliable way. Context: Kafka Streaming
Key Result (KR) Key Result in the OKR framework. It is a quantitative metric that measures progress toward the Objective. Must be measurable and verifiable. Context: Data-Driven Management (OKR)
KPI (Key Performance Indicator) Key Performance Indicator. A measurable metric that demonstrates how effectively a company is achieving its main business objectives. Context: Data-Driven Management / Metrics Fundamentals
L
Lagging Indicator Outcome indicator that looks at the past (e.g., last month’s revenue). Confirms a trend that has already occurred but is difficult to influence directly. Context: Data Driven Management
Leading Indicator (Driver) Predictive indicator that anticipates future results (e.g., number of demos booked). It can be directly influenced by the team’s actions. Context: Data Driven Management
LTV (Lifetime Value) Estimate of the total net revenue a company expects to earn from a single customer over the entire relationship. Context: Fundamental Metrics / Marketing Data Science
M
Materialized View Database object containing precomputed results of a query. In ClickHouse, they are incrementally updated upon data insertion for extreme performance. Context: Real-time Analytics / Data Warehousing
MergeTree engine The most powerful and robust ClickHouse table engine, designed to insert large volumes of data into a single table and query them quickly. Context: Real-time Analytics
Metric Any quantitative measure used to track and evaluate the status of a specific process. KPIs are a selected subset of metrics. Context: Fundamental Metrics
Monitoring The process of collecting, aggregating, and analyzing metrics on IT systems to monitor their health and performance (e.g., CPU, RAM, Latency). Context: Infrastructure Ops
MRR (Monthly Recurring Revenue) Normalized monthly recurring revenue. It is the standardized measure of predictable revenue for a subscription-based company. Context: Fundamental Metrics
N
North Star Metric The single metric that best captures the core value the product delivers to its customers. Aligns the entire company toward a long-term goal. Context: Data Driven Management
NPS (Net Promoter Score) Metric measuring customer loyalty by asking: “How likely are you to recommend our product to a friend?”. Ranges from -100 to +100. Context: Product Analysis / Marketing Data Science
O
Observability Measure of how well the internal state of a system can be understood based solely on its external outputs (Logs, Metrics, Traces). Goes beyond simple monitoring (“why is it broken?”). Context: Infrastructure Ops
Objective (OKR) The Objective in the OKR framework. A qualitative, memorable, and ambitious description of what you want to achieve. Context: Data Driven Management
OKR (Objectives and Key Results) Goal-setting framework used to define measurable objectives and track their outcomes. Encourages alignment and commitment around measurable goals. Context: Data Driven Management
OLAP (Online Analytical Processing) Data processing approach focused on multidimensional analysis and complex read queries (e.g., Data Warehouse). Context: Data Warehousing
OLTP (Online Transactional Processing) Data processing approach focused on executing a large number of short atomic transactions (e.g., operational databases, e-commerce). Context: Data Warehousing
Onboarding The process through which new users learn to use the product and perceive its initial value. Critical for retention. Context: Product Analytics
P
Parquet Open-source columnar storage file format optimized for use with big data processing frameworks. Offers efficient compression and schema evolution. Context: S3 Data Lake
Partitioning Technique of dividing a table or dataset into smaller, manageable parts (e.g., by date) to improve query performance. Context: Data Warehousing / Kafka Streaming
PII (Personally Identifiable Information) Any data that could potentially identify a specific individual (e.g., Name, Email, IP, Tax Code). Requires special protection under GDPR. Context: Privacy / Data Collection
PMF (Product Market Fit) The point at which a product meets strong market demand. Occurs when customers buy, use, and recommend the product as fast as you can produce it. Context: Product Analysis / Marketing Data Science
Precision Tracking Data tracking philosophy that involves explicitly and deliberately defining the events to track, as opposed to indiscriminate Autotrack. Context: Data Collection
Privacy by Design Systems engineering approach that promotes privacy and data protection from the start of the design cycle. Context: Privacy
Product Analytics Quantitative analysis of how users interact with a digital product, used to improve user experience and guide the roadmap. Context: Product Analytics
R
Retention Rate The percentage of users who continue to use the product after a certain period. It is the main indicator of a product’s long-term health. Context: Product Analysis / Marketing Data Science
ROAS (Return On Ad Spend) Marketing metric that measures the effectiveness of a digital advertising campaign (Revenue from ads / Cost of ads). Context: Metrics Fundamentals / Marketing Analytics
ROI (Return On Investment) Performance metric used to evaluate the efficiency of an investment. (Gain from investment - Cost of investment) / Cost of investment. Context: Fundamental Metrics
S
S3 (Simple Storage Service) Object storage service offered by AWS that provides industry-leading scalability, data availability, security, and performance. Common foundation for Data Lakes. Context: S3 Data Lake
Schema Registry Component that stores and manages versions of data schemas (e.g., Avro) for Kafka topics, ensuring compatibility between producers and consumers. Context: Kafka Streaming
Segmentation The process of dividing a large base of consumers or companies into subgroups of consumers (segments) based on shared characteristics. Context: Data-Driven Management / Marketing Analytics
Server-Side Tracking Data collection method where data is sent from the web server (backend) directly to the analytics platform, bypassing the user’s browser and AdBlockers. Context: Data Collection
SLA (Service Level Agreement) Contract between a service provider and a customer that specifies, usually in measurable terms, which services the provider will deliver (e.g., 99.9% uptime). Context: Infrastructure Ops
Snowflake Schema Multidimensional data model that is an extension of the Star Schema, where dimension tables are normalized (split into additional tables). Context: Data Warehousing
Star Schema The simplest data warehousing model, consisting of a central fact table surrounded by denormalized dimension tables. Context: Data Warehousing
Streaming Processing The continuous processing of data streams in real time as they are generated, enabling immediate analysis and reactions. Context: Real-time Analytics / Kafka Streaming
SUS (System Usability Scale) A standardized (“quick and dirty”) tool to measure system usability, based on a 10-question questionnaire. Context: Product Analysis (UX Research)
T
Terraform Open-source Infrastructure as Code (IaC) tool that allows defining and provisioning data center infrastructure using a declarative language. Context: Infrastructure Ops
Time to Value (TTV) The time it takes for a new customer to realize value from the product after purchase or signup. Reducing TTV is crucial for activation. Context: Product Analytics
Topic (Kafka) Category or feed name where records are published in Kafka. Topics are always multi-subscriber. Context: Kafka Streaming
Tracking Plan Document or repository that defines which events to track, their properties, and triggers, serving as a “single source of truth” for analytics. Context: Data Collection
U
Unit Economics Direct analysis of revenues and costs associated with a single unit business model (e.g., a single customer or product). Context: Fundamental Metrics / Marketing Data Science
Usability Testing Technique used in user-centered interaction design to evaluate a product by testing it with users. Context: Product Analysis (UX Research)
User Properties Attributes that describe segments of the user base, such as language preferences, geographic location, or subscription status. Context: Data Collection / Product Analysis
UX Research (User Experience Research) The systematic study of target users and their requirements to add realistic context and insights to design processes. Context: Product Analytics
V
Vanity Metrics Metrics that look impressive but do not necessarily correlate with business success and do not provide guidance on future decisions (e.g., Number of Likes, Total Pageviews). Context: Data Driven Management
Viral Coefficient (K-factor) Average number of new users each existing user manages to convert. If K > 1, exponential viral growth occurs. Context: Fundamental Metrics
W
Windowing Technique in stream processing to group events based on time (e.g., Tumbling window, Hopping window, Session window) for aggregations. Context: Real-time Analytics
Z
Zookeeper Centralized service to maintain configuration information, naming, distributed synchronization, and group services. Historically used by Kafka (now being removed with KRaft). Context: Kafka Streaming
Acronyms
| Acronym | Full Meaning | Reference Area |
|---|---|---|
| ACID | Atomicity, Consistency, Isolation, Durability | Database |
| API | Application Programming Interface | Development |
| ARPU | Average Revenue Per User | Business Metrics |
| ARR | Annual Recurring Revenue | Business Metrics |
| AWS | Amazon Web Services | Cloud |
| CAC | Customer Acquisition Cost | Marketing |
| CDC | Change Data Capture | Data Engineering |
| CDP | Customer Data Platform | MarTech |
| CI/CD | Continuous Integration / Continuous Deployment | DevOps |
| CMP | Consent Management Platform | Privacy |
| CPC | Cost Per Click | Advertising |
| CPM | Cost Per Mille (Impressions) | Advertising |
| CRM | Customer Relationship Management | Sales/Marketing |
| CTR | Click-Through Rate | Advertising |
| DAU | Daily Active Users | Product Metrics |
| DLQ | Dead Letter Queue | Data Engineering |
| DPA | Data Processing Agreement | Legal/Privacy |
| ELT | Extract, Load, Transform | Data Engineering |
| ETL | Extract, Transform, Load | Data Engineering |
| GA4 | Google Analytics 4 | Analytics |
| GDPR | General Data Protection Regulation | Legal/Privacy |
| IaC | Infrastructure as Code | DevOps |
| JSON | JavaScript Object Notation | Data Format |
| KPI | Key Performance Indicator | Strategy |
| LTV | Lifetime Value | Business Metrics |
| MAU | Monthly Active Users | Product Metrics |
| MRR | Monthly Recurring Revenue | Business Metrics |
| NPS | Net Promoter Score | Product Metrics |
| OKR | Objectives and Key Results | Strategy |
| OLAP | Online Analytical Processing | Database |
| OLTP | Online Transactional Processing | Database |
| PII | Personally Identifiable Information | Privacy |
| PMF | Product Market Fit | Product Strategy |
| ROAS | Return On Ad Spend | Marketing |
| ROI | Return On Investment | Business |
| S3 | Simple Storage Service | Cloud Storage |
| SLA | Service Level Agreement | Operations |
| SQL | Structured Query Language | Database |
| SUS | System Usability Scale | UX Research |
| UX | User Experience | Design |