Synthesising Information: The Core Research Skill

Synthesis — taking many disparate sources and pulling out a coherent picture — is one of the most time-intensive research activities. It is also one where AI provides the most dramatic efficiency gains, when used correctly.

The Paste-and-Query Method

The most reliable approach to AI synthesis: you provide the sources, AI processes them.

1. 1.Gather your sources (articles, reports, transcripts, notes)
2. 2.Paste the relevant text into your AI conversation
3. 3.Ask specific synthesis questions

"Here are four analyst reports on the state of the electric vehicle market. Based only on the information in these documents: What are the points all four agree on? Where do they disagree? What does each report say about consumer adoption barriers?"

The key phrase: "Based only on the information in these documents." This keeps the AI from mixing in its training data (which may be outdated) with your curated sources.

Thematic Analysis Prompts

For qualitative research — interview transcripts, focus group notes, customer feedback — AI can identify themes systematically:

"Here are 15 customer interview transcripts. Identify the five most common themes across them. For each theme, quote at least two specific passages that illustrate it. Note any outliers or surprising minority views."

This takes hours manually. AI does it in seconds, though you should verify the theme assignments on at least a sample.

Contradiction and Gap Identification

"Based on the sources I have provided, where do the sources contradict each other? What important questions do none of the sources address? What would a skeptic say is missing from this body of evidence?"

This is valuable for academic writing, policy analysis, and due diligence work.

Building Structured Summaries

"Summarise this report using the following structure: > 1. Key findings (3-5 bullet points) > 2. Methodology (2-3 sentences) > 3. Limitations acknowledged by the authors > 4. Implications for [YOUR SPECIFIC CONTEXT] > 5. Questions this report does not answer"

The final point — questions this report does not answer — is particularly valuable for identifying research gaps.

Competitive and Market Research

AI is useful for structuring competitive analysis:

"Here is publicly available information on three competitors: [PASTE]. Analyse them across: pricing model, target customer segment, key differentiators, apparent weaknesses, and recent strategic moves."

The output gives you a structured framework to build on — even if you need to update specific facts from verified sources.