How to sell something cheap: require a named replacement, separate a stalled business from a low price, locate the damage in the right segment, and publish the order with a limit.
1. Never sell into cash — make the sale compete against a named replacement
The repeatable method
- Frame the decision as a switch: what you own versus one specific alternative, sized the same.
- Compare the two on the same three axes — valuation trend, fundamentals, and the path of intrinsic value.
- Only sell if the alternative wins on all three. If it does not, the position stays.
- Sequence the orders so the purchase is funded by the sale, which prevents cash drag and forces the comparison to be real.
Here: both companies are charted anonymously before either is named — "Orange line: the company we're selling. Blue line: the company we're buying… The company we're buying became 61% (!) cheaper since 2022. It's now cheaper than the company we're going to sell" — followed by a fundamentals comparison and an intrinsic-value comparison. Then the orders: sell 1,000 OTCM at a $54 limit, and "after we sold OTC Markets, I will enter an order for Q 440 at a limit price of $123" — $54,000, almost exactly the proceeds.
Watch for
- The replacement being chosen after the decision to sell. Here it was already argued Positive on 11 January, which is the right order.
- A comparison that only works because the two are in different currencies of measurement. Both are compared on forward PE and on intrinsic-value growth.
2. Sell on the business, never on the price — and say which of the two you are acting on
The repeatable method
- Run the valuation work first, and note honestly whether the position is cheap.
- Then ask a separate question: has anything permanent changed in the earnings power?
- List the structural reasons explicitly. If the only reason is the price, do not sell.
- Where the business has changed, act regardless of how attractive the multiple looks — a low multiple on permanently stalled earnings is the definition of a value trap.
Here, three reasons, none of them valuation. "
Growth has stalled and it looks like it's a more structural problem than initially thought. Financial data are becoming more and more a
commodity product and OTC Markets seems to be struggling from increased competition. The high level of
Stock-Based Compensation (23.4% of Net Income) is also something I don't like." Seven days earlier the
same position passed every valuation test — 19.6x against a 21.6x average, an 11.9% modelled return, a reverse DCF needing 6.2% against 8.0% expected — and was still only a Hold. The sale is consistent with the standing rule: sell when the case breaks, not when the price is high.
Watch for
- "Dead money" as a reason by itself. Twenty-seven months of nothing is a prompt to re-examine, not a verdict — the verdict has to come from the business.
- Commoditisation arriving slowly. It is only visible in pricing and retention, both of which lag.
3. Apply the stock-compensation threshold as a hard disqualifier
The repeatable method
- Compute stock-based compensation as a percentage of net income, and compare it with a fixed bar (this archive uses <10%).
- Above roughly 20%, treat it as decisive rather than as a deduction — that share of the profit belongs to employees, not owners.
- Check the five-year average, not just the current year, so a single grant does not mislead.
- Apply it symmetrically: to names you want to sell and names you want to buy.
Here: OTCM's "
23.4% of Net Income" is the third reason for the sale. Four months later almost exactly the same figure — "
SBC as a % of Net Income equals 22% (!)" — is what keeps
FICO out of the portfolio despite a 7.8/10 quality score. And
LeMaitre, sixteen days earlier, is marked ❌ at 14.0% (15.6% on a five-year average) without that being decisive. Three applications, one bar, and the outcome scaling with the number.
Watch for
- Buybacks that merely offset the dilution. At 23% of profit, an unchanged share count is a cost, not a return.
- The charge being deducted in one model and not in the headline multiple — the inconsistency flagged in the LeMaitre case.
4. Locate a disappointing result in a segment before deciding it is temporary
The repeatable method
- Date the fall to a specific event, and quantify it.
- Split the company into its segments and establish which one missed.
- Classify that segment: cyclical or structural, and what share of revenue it is.
- Buy only if the miss sits in the smaller, cyclical part and the larger, structural part is untouched — and say so, so the claim can be checked next quarter.
Here, from the newspaper interview quoted in the issue: "The trigger was the earnings report in November 2025. That turned out less positive than expected, causing the share price to drop by about 17 percent." Then the split: "Zoetis consists of two segments: pet care and farm animal products… the weakness stemmed primarily from the latter. Those activities are also more dependent on the economic cycle", against a business where "a bit below 70% of its revenue is generated via the pet market." The miss is in the cyclical 30%.
Watch for
- A cyclical segment that is cyclical for longer than expected. Farm economics can stay weak for years.
- Guidance cuts that recur. One is a segment problem; three is a company problem.
5. Ask who actually pays — a business with no intermediary payer keeps its pricing power
The repeatable method
- Trace the money from the end user to the company and count the parties in between.
- Identify any professional negotiator in the chain — an insurer, a benefit manager, a state procurement office — because that is where margin is extracted.
- Prefer businesses where the end user pays directly and the purchase is emotionally rather than economically motivated.
- Confirm with realised pricing over several years, not with the theory.
Here: "
Unlike human healthcare, most customers pay out of pocket, so Zoetis doesn't have to deal with insurance companies putting pressure on prices." Supported by the demand argument — "people have fewer and fewer children and more and more pets, and they are treated as full-fledged family members" — and one anecdote that carries it: "her dog was sick, and she paid
$25,000 for medication, uninsured." The exact inverse of the
Novo Nordisk situation, where a pharmacy benefit manager's formulary decision moves the volumes and price is conceded to win access.
Watch for
- Pet insurance penetration rising. The moment an intermediary payer appears, this advantage starts to erode.
- Out-of-pocket demand being called recession-resistant. It is discretionary spending defended by emotion, which is strong but not tested at scale in a deep downturn.
6. Publish the order with a limit, and treat illiquidity as part of the decision
The repeatable method
- Before selling, check average daily volume against your position size.
- Set a limit price rather than selling at market, and state the minimum you will accept.
- Commit in advance to waiting rather than chasing the price down, and say so publicly if you have an audience.
- Sequence dependent orders explicitly, so the purchase cannot run ahead of the funding sale.
Here: "Please note that the liquidity in OTC Markets is limited. Hopefully we don't influence the stock price of OTC Markets too much. If we do, I won't be in a hurry to sell. I want to sell at the right price. We currently own 1.000 shares worth $55.470… the current stock price equals $55.5 and I don't want to sell it for much less. That's why I'm entering my sell order for Q 1.000 with a limit price of $54." Then the dependent buy: "after we sold OTC Markets, I will enter an order for Q 440 at a limit price of $123."
Watch for
- The conflict this creates. Announcing a sale in an illiquid stock to a large subscriber base moves the price against you and against readers; a limit and a stated willingness to wait mitigate it, they do not remove it.
- Corroboration from inside the same organisation. "TJ… already bought Zoetis for the Compounding Dividends Portfolio" is agreement, not independent confirmation.
Methods distilled from the archived Compounding Quality post for personal study. Not investment advice.