DeepSeek · DeepSeek Fundamentals
Effective Prompting Techniques
Discover prompt-layering and comparisons
Layering and Comparison Techniques
Ready to elevate your AI skills?
In this lesson, we'll explore two powerful prompting techniques that help get deeper, more comprehensive insights from DeepSeek. Let's dive in!
Understanding Advanced Techniques
Before diving into practice, let's understand what makes these techniques advanced and why they're worth mastering.
The thing is that while basic prompts are good for simple tasks, complex research or analysis needs a more sophisticated approach.
Technique 1: Prompt Layering
Layered prompting is like having a conversation that gets gradually deeper. Instead of asking everything at once, you break your query into connected steps.
Why does this work better? Because AI, like humans, processes information more effectively when complex problems are broken down into smaller parts.
This is especially true for DeepSeek's Reasoning Mode, which shows its step-by-step thinking — layering your prompts matches how DeepSeek naturally processes complex queries.
So, instead of asking one big question, break your research into connected steps. This helps DeepSeek focus on each aspect separately while maintaining context.
Choose one
What's the main advantage of breaking down complex queries?
Layering in Action
Here's how to analyze a market opportunity using layered prompts:
- Layer 1: "What is the current market size for mobile banking services in North America in 2026?"
- Layer 2: "Based on this market size, who are the top 3 providers and their market share?"
- Layer 3: "For these providers, what are their primary user demographics?"
Try it yourself with another layered prompt in the next exercise!
practice preview
Interactive practice
Fill in the blank
Fill in the gaps to complete the first part of the layered prompt.
practice preview
Interactive practice
Fill in the blank
Complete the second part of the prompt.
practice preview
Interactive practice
Fill in the blank
Finalize the prompt layering-based analysis with the prompt below.
Well done! As you can see, each layer narrows the focus while building on previous information, leading to a specific, actionable conclusion.
Otherwise, if the prompt was not layered, the answers would not have been as focused as they are now.
Technique 2: Comparative Analysis
Our second advanced technique is comparison prompting. This method helps uncover insights by examining differences and similarities between elements.
It works particularly well with DeepSeek's Search Mode, which can pull information from multiple sources to build comprehensive comparisons.
Why compare? It is because seeing how things relate to each other often reveals insights that studying them in isolation might miss.
For example, instead of just asking about a product:
”Compare the adoption rates of electric vehicles in urban vs. rural areas in 2026, considering:
- Infrastructure availability
- Consumer demographics
- Price sensitivity”
Choose one
What makes comparative prompts valuable?
Combining Both Techniques
Now, let's see how layering and comparison work together:
- Layer 1: "Compare market penetration of mobile payment systems in North America vs. Europe in 2026"
- Layer 2: "For both regions, analyze user demographics and preferred payment scenarios"
- Layer 3: "Based on this data, predict growth trends for the next two years"
This combined approach helps you build a thorough understanding while keeping each step focused and manageable.
Choose one
Now, what's the best way to combine layering and comparison?
Good Job!
Now, you are ready to apply these techniques to your own projects. Yet, before doing so, let's recap the key tips from this lesson.
- Start with broader questions before diving deep
- Use each response to inform your next prompt
- Keep comparisons focused and relevant
- Build analysis step by step