Effective data wrangling starts with data collection. Make sure to think about Front-end validation, limiting free text, GDPR, security, accessibility, inclusion, and designing to minimise the amount of data collected. The next part is effective data storage. You can store data in relational or non-relational data stores. Relational data solutions in Azure include Azure SQL Database, Azure Database for MariaDB, and Azure Database for PostgreSQL. Non-relational data solutions include Azure Storage and Azure Cosmos DB. You might wrangle data during a data movement process between systems. A batch process moves multiple records, typically on a schedule. A streaming process operates on each individual event/message as it happens. In ETL (Extract Transform Load) you run a process to retrieve data from a system, process it, and then insert it into a new data store. In ELT (Extract Load Transform) you run a process to retrieve data from a system, insert it into a new data store, and the process it. The modern data warehouse stack (used to combine data from multiple sources) includes Azure Data Factory, Azure Databricks, Azure Synapse Analytics and Azure Analysis Services. Power Query is a useful language for processing data and can be used in Excel, Power BI, and Azure Data Factory. Power BI is a self-service data modelling and visualisation tool that can connect to data stored in lots of different places. It has a data modeling area where Power Query can be used to build a dataset. The reports allow you make multi-page analysis of a dataset. A dashboard combines visuals from one or more reports to provide a broader and simpler view. SQL is a language that works with many relational and non-relational systems. You can use it interact with data, data structures, and the access model for the data objects. Finally you can write code to wrangle data with Python, R, and dotnet being common data wrangling languages.