Four practical filters for making one of investing’s most neglected decisions.
Buying a stock gets plenty of attention. Selling one rarely does.
When markets fall, the familiar advice is to buy more, average down, or add to existing positions. Analysts, experts and television commentators spend far more time discussing what to buy than when to sell. Yet deciding when to part with a stock—or a mutual fund—is just as important as deciding when to own it.
The question also comes up repeatedly among investors. It was one of the most frequent requests from the community around a YouTube channel that recently crossed 50,000 subscribers, roughly a year after it began. Starting the channel had never been part of a grand plan, and reaching that milestone so quickly was unexpected. More exciting than the number, though, was the conversation developing around it: community posts asking for views were met with suggestions for what the next video should explore.
One of its recurring requests was simple: When should an investor sell?
There is no exhaustive list of reasons. Circumstances differ, and any framework leaves room for exceptions. But four filters can bring some discipline to the decision.
1. Has the original investment thesis changed?
Every stock enters a portfolio for a reason—or at least it should.
Some investors do their own research. Others act on an analyst’s recommendation, a television discussion, or the decision of a portfolio manager to add a company. Even when the original idea comes from somebody else, it is worth writing down three or four specific reasons for buying the stock.
Perhaps a company is about to launch a new product. Perhaps its fundamentals are intact but its valuation has fallen well below historical levels. A business that has traded at a price-to-earnings multiple of about 25 times over the past ten years may now be available at 15–16 times earnings.
That gap may be the reason for buying. The next question is whether the reason continues to hold.
Consider ITC, Infosys and HDFC Bank. These are examples of stocks that have frustrated many investors by going nowhere for three, four, or even five to six years. A long period of weak share-price performance naturally creates doubt. But frustration is not, by itself, an investment thesis.
The more useful questions are:
- What was the company doing five years ago that it is not doing today?
- Has something fundamental changed, or is the business dealing with an external slowdown?
- In ITC’s case, is the weakness connected to the broader slowdown in FMCG?
- In HDFC Bank’s case, did the merger compress net interest margins, and will the recovery simply take time?
- In Infosys’s case, is artificial intelligence creating a genuine, lasting threat to the business?
These questions are difficult. Different investors may examine the same information and disagree: one may see permanent damage, while another sees a temporary problem. That disagreement is precisely why analysis matters.
It also helps to have a community of thoughtful people—friends or fellow investors—who can challenge an opinion. The aim is not to let somebody else make the decision, but to understand how the same business looks from another perspective.
One particularly useful exercise is to pretend the stock is not already in the portfolio.
Forget the price at which it was purchased. At today’s price, would an investor buy it, sell it, or stay away? Ask somebody else the same question. If the answer is that the stock would not be bought today—and might actually be sold—the original thesis deserves closer examination. If something has permanently changed for the worse, that is a good reason to sell.
The important distinction is between a company whose business has changed permanently and one whose share price has merely disappointed.
The examples of ITC, Infosys and HDFC Bank are deliberate. Strong performers in the portfolio, such as Laurus Labs, Hero MotoCorp and Federal Bank, could have been used instead. But the less exciting examples make the point better: a stock can do very little for years without the underlying business necessarily suffering permanent damage. The assessment made here is that the fundamentals of these three companies have not changed sufficiently to justify selling them; external circumstances have weighed on performance. On that view, the decision is to keep holding them for the long term.
The first filter, then, is not whether the investor has made money. It is whether the reason for owning the business still exists.
2. Has the portfolio moved away from its intended allocation?
Selling is not always a judgment against a company. Sometimes it is simply a matter of maintaining portfolio discipline.
Before building a portfolio, an investor should decide how much to allocate to large-cap stocks, mid-cap stocks, small-cap stocks and fixed-income investments such as bonds. Once that allocation is set, market movements should not be allowed to change it indefinitely.
Take a portfolio designed to hold 60% in equities and 40% in fixed income.
If the stock market falls by 10%, 20% or 30% over a year, the equity portion will shrink relative to fixed income. That can be an opportunity to sell some fixed-income investments and add to equities, restoring the intended 60:40 balance.
The reverse applies after a strong market rally. When equities have grown beyond their intended weight, selling some stocks and moving the proceeds into fixed income brings the portfolio back to its original allocation.
But a practical problem remains: Which stock should be sold?
Sometimes a portfolio contains 25 stocks, all of them good businesses. This happens in portfolio consultations too. A client pays for an analysis, sends across a portfolio, and the honest reaction is that it is brilliant. There may be nothing useful to recommend. Why suggest a change merely because somebody paid for a review? Activity for its own sake would serve no purpose.
Suppose the investor needs ₹50 lakh to purchase land or make another investment. There is no automatic reason to sell one excellent company in its entirety. A more balanced approach may be to sell proportionately across the holdings. In a 25-stock portfolio, for example, trimming 4% of each position would release 4% of the portfolio’s value; where that produces the required amount, the investor can raise the cash without eliminating a holding.
The same principle applies when rebalancing after a rally. A reduction across the portfolio can make more sense than forcing a complete exit from a business that remains attractive.
This discipline becomes especially important when markets are rising sharply. Although the idea of a powerful rally may feel distant during a sluggish market, such periods return. Investors become greedy. Instead of selling equities to maintain their allocation, they may begin selling fixed-income investments to buy even more stocks, leaving themselves overcommitted to the market.
That is exactly when discipline matters most: not only between equity and fixed income, but also among large-, mid- and small-cap stocks.
A sale can be sensible even when nothing is wrong with the stock. The portfolio’s allocation, rather than the company’s quality, may be the reason.
3. Has the valuation become unattractive—or returned to fair value?
Of the four filters, valuation deserves particular attention.
One practical approach is to begin with a company’s historical valuation. This is especially useful for large-cap businesses and more established mid-cap companies. At the time of the discussion, much of the fresh buying was concentrated in that part of the market precisely because it had gone nowhere for some time. Historical comparisons are less useful for very small or rapidly expanding businesses—roughly those below ₹15,000–20,000 crore in market value—where investors may be buying for explosive growth and the past valuation may offer little guidance.
For established companies, the process is more straightforward. Study the multiple at which the stock has historically traded, then consider what could justify a return to that level.
Take HDFC Bank as an illustration. If a stock has historically traded at a valuation multiple of around 26–28 but is now at 14–15, an investor may work out the price at which a return to the old multiple would be justified and wait for the gap to narrow. In that framework, a recovery in the valuation multiple can lift the share price.
The estimate can also change with the business. If the company is developing new revenue streams, acquiring businesses or doing something that improves its growth prospects, an allowance of about 20% above or below its historical valuation may be used.
Then comes the harder question: How long should an investor wait?
A useful thumb rule is 36 months—three years. That is generally enough time to observe whether the expected development is beginning to play out.
Of course, this immediately raises an objection. What about stocks such as ITC or HDFC Bank, which may have gone nowhere for three years or longer?
The answer circles back to the first filter: the original thesis.
HDFC Bank’s merger with a housing-finance company was expected to put pressure on margins. Recovering from that pressure could take time. In ITC’s case, the stated concerns include a substantial increase in GST on cigarettes and a severe slowdown in the FMCG sector since the implementation of GST, affecting its sizeable non-cigarette FMCG business as well.
Yet these are established companies whose products and services remain visible in everyday life. They have not stopped operating. They are not fly-by-night businesses that may suddenly disappear.
The central question is whether their valuations can eventually return to levels at which they traded in the past. One investor may conclude that they can; another may think that the old valuation will never return. There is no substitute for making that judgment independently.
The three-year rule is a guide, not an automatic sell order. If the original thesis still stands, a stock’s failure to perform within that period need not end the investment.
At the same time, constantly buying and never selling is not necessarily sound investing. The amount of selling matters too, and that leads to the final filter.
4. Is there a good reason to sell at all?
Before selling a stock or mutual fund, an investor should ask a question that sounds almost too simple:
What will happen to the money after the sale?
If a business has been held for only two years, it may not have had enough time to deliver on the original thesis. If the investor does not need the money, why sell merely for the sake of doing something?
Every sale creates a second decision. The proceeds must go somewhere. They may be invested in another stock, moved into fixed income, or used for some other purpose. If the existing holdings no longer look attractive and fixed income appears preferable, that is a different and legitimate reason to sell.
But the intended use of the proceeds should be considered before the sale, not discovered afterward.
A simple way to measure selling: portfolio churn
Portfolio churn is a useful check on whether an investor is trading too much—or perhaps too little.
For this purpose, the churn ratio is calculated as:
Churn ratio = (Value of stocks sold during the year ÷ Portfolio value at year-end) × 100
The year-end portfolio value includes investments purchased during the year.
Suppose ₹10 lakh worth of stocks were sold over a year, and the portfolio was worth ₹1 crore at year-end. The churn ratio would be:
(₹10 lakh ÷ ₹1 crore) × 100 = 10%
A churn ratio of around 10% is a useful target, and preferably not more. In an exceptional year, particularly after strong portfolio performance, it might rise towards 20%. But routinely turning over 10–20% or more—whether in an individual portfolio or a managed fund—is considered high in this approach. Too much selling can itself become a problem.
The bigger problem observed, however, is often at the other extreme. Some investors hardly ever sell, even when their stocks have risen considerably or when something has gone seriously wrong with a holding.
The goal is not to trade constantly. Nor is it to treat selling as a failure. A sensible middle ground allows room for occasional exits while giving investments time to work.
The hardest discipline may be doing nothing
One reason investors struggle with selling is the sheer volume of noise surrounding the markets.
There are news updates, apps, alerts and notifications arriving throughout the day. Much of this information creates the feeling that something must be done immediately. One almost wishes for an app that could be installed on every phone and computer simply to block the news and stop the notifications.
The pressure to act appears even after a portfolio has been properly reviewed. An investor may receive a clear assessment, a list of stocks to hold or buy, and a plan for systematic investments—and then immediately ask how soon the next consultation should take place.
Sometimes the most sensible answer is: if the review is in 2026, perhaps look closely at it again in 2030. In the meantime, follow the investment plan and stop watching the portfolio constantly.
That may sound strange coming from someone whose livelihood requires following financial news and markets every day. But watching the market and changing a portfolio are two different activities.
For perspective, consider the portfolio behind this discussion. Although following markets every day was part of the speaker’s livelihood, the portfolio itself had not been changed between March and October 2026. The last adjustment, in March, was simply to sell some gold and buy stocks. There was no predetermined date for the next change.
The broader point is simple: doing too much can damage results.
Even a skilled investor will ordinarily find some reason to sell a portion of the portfolio over time—perhaps because a holding has appreciated significantly, or because new information reveals a serious problem with a company. That is why annual selling around 10%, and occasionally 10–20%, can be a useful reference point. But activity for its own sake is not the objective.
A final check before pressing “sell”
Selling deserves the same care as buying. Before exiting a stock or mutual fund, consider four questions:
- Thesis: Has the reason for buying fundamentally changed?
- Allocation: Has the portfolio drifted away from its intended mix of equities, fixed income, and market-cap categories?
- Valuation: Is the stock priced at a level that no longer makes sense relative to the business and its history?
- Purpose: Is selling necessary, and is there a clear plan for the proceeds?
Then check the churn ratio. Too much selling can be a problem; never selling can be a problem too.
The aim is not to find a clever reason to transact. It is to make sure every sale has a reason.
These filters can be put into practice and debated. There may be other good reasons to sell, and those experiences and perspectives are worth sharing in the comments. They are part of what makes a thoughtful investing community useful in the first place.
Source video: When Should You Sell a Stock or Mutual Fund?
