Strategy

Portfolio Rebalancing 101: When to Act and What to Sell

By Crito Team  •  June 5, 2026  •  10 min read

Most investors know they should rebalance their portfolios. Few do it systematically. Fewer still do it well. The ones who do consistently outperform over long periods — not because they're smarter, but because they're more disciplined.

This guide explains when to rebalance, how to decide what to sell, and how AI makes the entire process faster and more rigorous.

What Is Portfolio Rebalancing?

When you build a portfolio, each position has a target weight — say, 15% in Reliance Industries, 10% in TCS, 8% in HDFC Bank. Over time, prices move and these weights drift. Reliance might run up to 22%. TCS might fall to 7%. Your portfolio no longer reflects your original thesis.

Rebalancing is the process of returning to your target weights — selling some of the outperformers and buying more of the underperformers (or adding new positions entirely).

Done well, rebalancing enforces a sell-high, buy-low discipline that's psychologically difficult to maintain without a systematic process.

The Two Main Rebalancing Approaches

Calendar-based rebalancing

Rebalance on a fixed schedule: quarterly, semi-annually, or annually. Simple. Easy to stick to. The downside is that you might rebalance when markets are fine and miss a rebalancing event when markets move sharply between schedule dates.

Threshold-based rebalancing

Rebalance when any position drifts beyond a set threshold — say, when a holding's weight moves more than 5 percentage points from its target. More responsive to market events, but requires more monitoring.

Most professional portfolio managers use a hybrid: they rebalance on a calendar schedule, but also trigger off-cycle rebalancing when a threshold breach occurs.

💡 Crito's approach: Run a rebalancing analysis whenever something meaningful changes — a big earnings surprise, sector-level news, or quarterly as a baseline check. The AI compares your current holdings to a fresh ideal portfolio and flags meaningful drift.

How Much Drift Is Too Much?

Portfolio sizeRecommended drift thresholdWhy
Under ₹5 lakh10%+ per positionTransaction costs eat into small rebalances
₹5–50 lakh5–7% per positionMeaningful enough to matter, manageable costs
Over ₹50 lakh3–5% per positionPrecision pays at this scale

These are starting points, not rules. Your tolerance for drift should also reflect your transaction costs, tax situation, and conviction in each position.

What Should You Sell When Rebalancing?

This is where most investors struggle. The instinct is to hold winners and sell losers. The math says the opposite: trim winners that have grown beyond their target weight.

When deciding what to sell, ask four questions:

  1. Has the investment thesis changed? If yes, reduce more aggressively regardless of price. If no, trim back to target weight only.
  2. Is the valuation stretched? A stock at 25x P/E that ran to 35x during a market frenzy is a stronger sell candidate than one at 20x that grew to 25x on earnings growth.
  3. What are the tax implications? Short-term capital gains in India (held under 1 year) are taxed at 15%. Long-term gains over ₹1 lakh are taxed at 10%. Timing your sells around the 1-year mark can meaningfully reduce your tax bill.
  4. Is there a better use for the capital? If your AI analysis identified a stronger candidate than the stock you're selling, rebalancing into that new position is more valuable than just reducing weight.

Tax Considerations for Indian Investors

Indian tax law distinguishes between short-term and long-term capital gains on equity:

Practically, this means:

⚠️ Important: This is general information, not tax advice. Consult a chartered accountant (CA) for guidance on your specific situation. India's tax rules change regularly.

How AI Improves Rebalancing Decisions

Traditional rebalancing is mechanical: if a position is 5% over target, trim 5%. That's fine for passive portfolios, but misses crucial information available about each position's current quality.

AI-assisted rebalancing asks a harder question: given today's market data, is this the right composition, not just the right weight?

When you run a rebalance analysis in Crito, the AI:

  1. Re-evaluates every position against your original investment thesis using current market data
  2. Identifies if any holdings have fundamentally weakened since your last analysis
  3. Suggests replacements for weak positions — not just weight adjustments
  4. Provides a ranked comparison of current holdings vs. the AI's fresh ideal selection
  5. Flags positions where the gap between current weight and ideal weight is most significant

This is more thorough than traditional rebalancing and turns the process into a real portfolio review, not just a weight-normalization exercise.

The Rebalancing Mistakes to Avoid

Let AI handle your rebalancing analysis

Build your basket on Crito and re-run the analysis quarterly. The AI compares your current holdings to the ideal portfolio and tells you exactly what to change.

Start for free →

Further Reading