Summary  
This chapter demonstrates how to prompt a language model to transform a broad topic into one clear main research question and several supporting questions, structuring them for inclusion in a research plan.

General domain of usage  
Customer support

A broad topic gives Claude a general direction, but research questions define exactly what it should investigate. For example, the topic **AI in customer support** could be divided into questions such as:

* How are companies currently using AI in customer support?
* What tasks can AI automate?
* What are the main benefits and limitations?
* How does AI affect customers and support employees?
* What results have companies reported?

Good research questions should be specific, relevant, and possible to answer using reliable sources.

> ### Task
> 
> * Choose a research topic and ask Claude to turn it into one main research question and several supporting questions.
> 
>   > **My research topic is how AI is changing customer support. Create one clear main research question and five supporting questions. Save them to the research plan in my workspace.**

Build a complete AI-powered research system using Claude Code. You will learn how to turn a broad topic into focused research questions, create a structured plan, find and evaluate reliable sources, and organise evidence in a reusable workspace. Through practical AI Lab chapters, you will create source libraries, compare conflicting information, store findings in a research database, and produce clear research briefs with accurate citations. By the end of the course, you will have a repeatable Claude Code workflow that can be used for future research projects.

Learn what makes a research system reliable and how to prepare a structured workspace before collecting information. You will define research questions, set the scope of the project, and create a clear research plan for Claude to follow.

Learn how to find relevant information and judge whether a source is trustworthy. You will remove weak, outdated, and duplicate sources before organising the strongest results into a structured source library.

Turn collected sources into clear and usable evidence. You will extract key findings, compare sources that disagree, and build a research database that keeps claims, evidence, and source details connected.

Use the collected evidence to develop supported conclusions and create a concise research brief. You will review claims and citations for accuracy before turning the complete process into a reusable Claude Code workflow.