Stockertim Dictionary · Artificial intelligence · Technology
Natural language query
Also called natural language queries, NLQ, plain-language query, natural language querying
In short
A question put to a data tool in everyday words rather than code, which the tool translates into a query and answers from the data it holds.
- Beginner
- Technology
- Checked
In detail
A natural language query lets someone ask a data tool a question the way they would ask a colleague: "which products sold most in March" or "show me late deliveries by depot". The tool interprets the words, works out which data and calculation are meant, runs a query and returns a number, table or chart.
Behind the scenes the tool maps words to fields and measures. Some tools use defined business terms set up by a data team, so that "revenue" always means the same calculation; others rely on a language model to guess the mapping. Defined terms give more consistent answers, while open guessing handles more varied wording but makes more mistakes.
Everyday words are often ambiguous. "Last month" may mean the last calendar month or the last 30 days; "customers" may mean accounts or people; "sales" may include or exclude refunds and tax. A good tool shows how it read the question, such as the filters and the measure it used, so the user can spot a misreading.
Natural language queries widen access to data, but the answer is only as reliable as the interpretation. Checking a new kind of answer against a known report is a sensible habit.
Worked example
A manager at Lagoon Logistics types into a dashboard: "How many deliveries were late last month?"
- The tool answers 340 and shows that it used the last 30 days and counted deliveries more than one day after the promised date.
- The monthly report, which uses the calendar month and any delay at all, shows 410.
- The manager rephrases: "How many deliveries in October arrived after the promised date?"
- The tool now answers 410, matching the report.
Both answers were correct for the question as the tool read it; showing its interpretation let the manager see why they differed.
In practice
Business intelligence tools, spreadsheets and support desks offer plain-language questions so non-specialists can explore data. Data teams improve reliability by defining business terms and by checking common questions against official reports.
Watch out
- Ambiguous words such as "last month" or "customers" can be read in more than one reasonable way.
- A confident chart is not proof the tool understood the question; read its stated filters.
Learn it properly
Official sources
- AI Risk Management Framework · National Institute of Standards and Technology
- AI for small business · US Small Business Administration
Facts last checked on 3 November 2026.
Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.