Data platforms, analytics and the pipelines that turn information into decisions.
What dbt does, how models, tests and docs work, a simple model example you can follow, what changed with dbt v2 and the Fivetran merger, and how to start.
A plain-English guide to the 4 statistics ideas behind most business reports, averages, variance, correlation and significance, with worked examples and Excel functions.
A hands-on pivot table tutorial that builds a summary from a simple sales table, then adds filters, date grouping, slicers and a PivotChart.
Choose Power BI if your company runs on Microsoft 365, Windows and Azure and wants low per-user costs. Choose Tableau if you need flexible visual exploration, Mac support …
Embedded analytics means putting dashboards, charts and reports directly inside a software product or internal app, so users see insights where they already work instead of opening a …
BI tools cost anywhere from free to thousands of dollars a month. Most charge per user, with published list prices from about $3 to $75 per user per …
Excel is best for ad hoc analysis, modeling and one-off reports. Power BI is best for shared, automatically refreshed dashboards that combine several data sources. Move to Power …
To choose a BI tool, start with your data sources, who will build reports versus only view them, how much governance you need and how each vendor charges. …
To build a dashboard people use, start with the decisions it supports, not the charts. Pick 5 to 9 metrics tied to those decisions, define each one, and …
Business intelligence (BI) is the set of processes and tools that collect data from across a business, clean and store it in one place, and turn it into …
Self-service analytics lets business users explore data, build reports and answer their own questions without waiting on IT or a data team. It works when a central team …
Big data is data so large, fast-moving or varied that ordinary databases and spreadsheets can't store or analyze it efficiently, so it needs distributed systems that spread the …
A metric is any number you track, like website visits or calls answered. A KPI, or key performance indicator, is one of the few metrics tied directly to …
Data analytics is the process of turning raw business data into answers you can act on. You start with a clear question, collect and clean the relevant data, …
The 4 types of data analytics are descriptive (what happened), diagnostic (why it happened), predictive (what is likely to happen next) and prescriptive (what you should do about …
Data-driven decision making means choosing actions based on agreed metrics and evidence rather than gut feel alone. Build the habit by tying a handful of KPIs to clear …
Structured data follows a fixed format of rows, columns and defined fields, like sales records or customer tables, so it's easy to search with SQL. Unstructured data has …
Predictive analytics uses historical data, statistics and machine learning to estimate what is likely to happen next, such as which customers will churn, how much stock you'll sell …
Data analytics uses existing data to explain what happened and guide current decisions, mostly with SQL, spreadsheets and BI dashboards. Data science builds predictive models and new methods, …
dbt (data build tool) is a transformation framework that turns raw data already in your warehouse into clean, tested tables. You write SQL select statements, and dbt builds …
A data pipeline is a series of automated steps that moves data from where it's created, such as apps, databases and APIs, to where it's used, such as …
Batch processing collects data and handles it in scheduled chunks, with results minutes to days later. Stream processing handles each event as it arrives, with results in seconds …
ETL (extract, transform, load) cleans and reshapes data on a separate engine before loading it into a warehouse. ELT (extract, load, transform) loads raw data first and transforms …
A data warehouse stores cleaned, structured data in defined tables for fast SQL reporting and BI. A data lake stores raw data of any type, such as files, …
Snowflake is usually the simpler choice for SQL analytics, BI and data sharing, with compute billed in credits and little infrastructure to manage. Databricks is usually stronger for …
A data lakehouse is a data platform that stores all your data in low-cost cloud object storage, like a data lake, but adds warehouse features such as ACID …
A data warehouse is a central database built for analysis. It pulls data from systems like your POS, CRM, ecommerce store and accounting software, organizes it into consistent …
To improve data quality, pick the datasets that drive decisions, define what "good" means for each using 6 dimensions (accuracy, completeness, consistency, timeliness, uniqueness and validity), fix problems …
To clean data, back up the original, profile it to find problems, remove duplicates, handle missing values, fix formatting such as extra spaces, text case, dates and numbers …
Synthetic data is artificially generated data that mimics the patterns of real data without copying actual records. Teams use it to test software, share data more safely, train …
A data catalog is a searchable inventory of a company's data assets, such as tables, files, dashboards and models, plus the metadata that explains them. It shows what …
Data governance is the set of roles, rules and decisions that control how your organization's data is defined, accessed, protected and kept accurate. A practical starting model has …
To create a data retention policy, inventory the data you hold, group it into categories, set a retention period for each based on legal requirements and real business …
Master data management (MDM) is the practice and technology of keeping 1 trusted version of your core business records, such as customers, products, suppliers and locations, across every …
A data mesh is an organizational model in which business domains own their data and publish it as products under shared governance rules. A data fabric is a …
Google Sheets is better for real-time collaboration, browser-first teams and quick shared trackers. Excel is better for large datasets, complex models, Power Query, Power Pivot and offline desktop …
To build a pivot table in Excel, format your data as a table with one header row, click any cell, choose Insert > PivotTable, then drag fields into …
The statistics you need for most business data analysis are the mean and median (what is typical), variance and standard deviation (how spread out the numbers are), correlation …
SQL is the standard language for asking questions of data in a database. Beginners need about 8 building blocks to answer most business questions: SELECT, WHERE, ORDER BY, …
Choose a chart by the question you're answering. Use a bar chart to compare categories, a line chart to show change over time, a stacked bar or a …
A Jupyter notebook is a document that mixes runnable code, results, charts and written notes in one file, saved with an .ipynb extension. You write Python in small …
For most people starting in data analysis, Python is the better first choice because it's widely used beyond analytics, has strong libraries such as pandas and scikit-learn, and …
Retail analytics uses sales, inventory, customer and traffic data to decide what to stock, how to price and how to staff. The metrics that matter most are conversion …
FP&A analytics is the use of financial and operational data to plan budgets, forecast results and explain variances between plan and actual. It turns accounting data into forward-looking …
Small businesses can use data analytics by starting with data they already have in their POS, accounting software and website tools. Begin with 5 starter analyses: best and …
Healthcare data analytics uses clinical, operational and financial data from EHRs, claims, scheduling and devices to improve patient care and run practices more efficiently. Common uses include quality …
People analytics is the use of employee and HR data to make better workforce decisions about hiring, retention, engagement, pay and performance. It combines data from HR systems, …
Supply chain analytics is the use of data from purchasing, inventory, warehousing and shipping to measure and improve how goods move from suppliers to customers. Core metrics include …
Comparing two versions of a web page, email, or feature with different groups of users to see which performs better.
Big DataDatasets so large, fast-moving, or varied that traditional tools struggle to store and process them efficiently.
Business IntelligenceTools and practices that collect, analyze, and present business data, usually through reports and dashboards, to support decisions.
Customer Data PlatformSoftware, known as a CDP, that combines customer data from many sources into unified profiles that marketing and other tools can use.
Data CleansingFinding and correcting errors, duplicates, and inconsistencies in data so it is accurate and usable for analysis.
Data GovernanceThe policies, roles, and processes that determine how an organization's data is managed, protected, accessed, and kept accurate.
Data LakeA central repository that stores large volumes of raw data in its original format until it is needed for analysis.
Data LakehouseA data architecture that combines the low-cost, flexible storage of a data lake with the structure and query performance of a data warehouse.
Data MeshAn approach to data architecture that gives business domains ownership of their own data, shared as products across the organization.
Data PipelineAn automated series of steps that moves data from one or more sources, transforms it, and delivers it to a destination for use.
Data VisualizationPresenting data as charts, graphs, maps, or other visuals so patterns, trends, and outliers are easier to understand.
Data WarehouseA system that stores cleaned, structured data from many sources, optimized for reporting and business analysis.
Digital TwinA virtual model of a physical object, process, or system that is updated with real-world data to simulate and predict its behavior.
ETLExtract, Transform, Load: a process that pulls data from sources, cleans and reshapes it, and loads it into a destination such as a data warehouse.
First-Party DataInformation a business collects directly from its own customers and audience, such as purchases, sign-ups, and website activity.
Jupyter NotebookAn open-source, browser-based environment that mixes live code, results, charts, and explanatory text in a single document.