How Many PYQs Should You Solve for UPSC Prelims?

The honest answer is that the number matters less than what you do with each question — but aspirants still need a concrete target. Here is a realistic plan that balances coverage with deep analysis, so you are not just farming volume.

The target: depth before count

A practical benchmark:

• Minimum: the last 10 years of Prelims GS Paper-I — roughly 1,000 questions — solved and analysed.
• Ideal: 15–20 years for high-weight, stable subjects (Polity, Geography, Environment), where old patterns still repeat.
• Each question twice: once cold, once again in revision. Solving 1,000 questions once is weaker than solving 700 questions twice.

If you go back to 1995 for the durable subjects, you are well past what most successful candidates manage — and that is enough.

One PYQ analysed properly teaches more than ten PYQs rushed. Volume without analysis is just typing.

Why "just do more" backfires

Beyond a point, extra questions add little. The traps and themes repeat, so your 1,200th question rarely teaches you something your first 800 didn't — unless you analysed those 800 well. The bottleneck is almost never the number of questions seen; it is the depth of review. This is the core idea in our PYQ strategy guide.

A revision cadence that works

1. Week 1–N: solve one subject's PYQ set cold, then spend double the time analysing every option.
2. Maintain an error log: one line per mistake, capturing the exact misconception.
3. Spaced re-attempts: redo a set after ~3–4 weeks; you should now be eliminating traps automatically.
4. Final phase: stop solving new sets and instead revise your error log and re-attempt the questions you got wrong.

How to know you've solved "enough"

You are ready not when you hit a number, but when you can predict the answer pattern: you read a question and instinctively spot which statement is the trap. When new PYQs stop surprising you, the marginal value has dropped — shift your remaining energy to revision and full-length timed papers.

For where to concentrate this effort, pair this plan with the subject-wise weightage analysis so your question count is skewed toward the subjects that actually carry the paper.