Glossary
What is data dictionary?
A data dictionary is an overview that describes in plain language, for each table and column, what it contains, in which unit and with which valid values. It helps both people and AI to interpret data correctly.
Data dictionary explained
A data dictionary records what each column really contains. Is "amount" including or excluding VAT? Does "status 3" mean delivered or cancelled? Does a credit note count as negative revenue? Agreements like these often exist only in one colleague's head.
For AI a data dictionary is especially valuable, because a language model cannot guess what an organisation means by a column. With a good description, the model more often picks the right column and the right filter.
A data dictionary is not the same as a data catalogue. The dictionary describes the contents of tables and columns, while a catalogue gives an overview of all data sources, reports and owners in the organisation. A catalogue often includes a dictionary, but is broader.
Example
Acme Groothandel's data dictionary describes orders.total_amount as "order amount in euros, excluding VAT and shipping costs, after discount". Both the controller and the AI then know they must add shipping costs separately if those count as revenue.
Data dictionary in ENABLE
ENABLE compiles a data dictionary of the data warehouse for each organisation, with tables, columns, data types, relationships and an estimated row count. It is sent with the question to the language model in the SQL Explorer, without any rows of data.
FAQ
Frequently asked questions
What is the difference between a data dictionary and a data catalogue?
A data dictionary describes what each table and column contains. A data catalogue is a broader, searchable overview of data sources, reports and owners, often including a dictionary.
What does a data dictionary contain?
For each column: the name, data type, a plain-language description, the unit and the valid values. Sometimes also the source, the owner and notes on special cases.
Does a data dictionary help AI?
Yes. With good descriptions, a language model that writes SQL picks the right columns and filters more often, because it does not have to guess what a column means.
Further reading
- ResourcesText-to-SQL explained: AI on your databaseText-to-SQL is a technique in which an AI model converts a question in plain language into an SQL query, which is then run on a database. The language model writes the query; the database supplies the figures. How well that works depends mainly on the quality of your metadata and on the safeguards around execution, such as read-only access, timeouts and row limits.Read more
- PlatformSQL Explorer with AI: ask your database questionsAsk your data warehouse in plain language. The AI writes the SQL, you stay in control.Read more
- SolutionAI governance: control over data, AI and accessWho sees what, what the AI receives and who approves: practical AI and data governance.Read more
- ResourcesGlossary of data, AI and Power BI termsWhat is a service principal, row-level security, a semantic model, text-to-SQL or RAG? Short, clear definitions of terms around data, Power BI and AI.Read more
Use AI on your own data. Securely, and without the headache.
In an online demo we show how ENABLE makes your data securely available and how you build on it: describing an app in the AI Builder, plain-language questions, dashboards and APIs. And how we set it up and manage it for you.