Actionable insights — Best Buys January 2026

1. Do the quality work once, then shop only from the resulting universe

1. Pre-qualify a fixed "investable universe" of companies that already pass the quality screen — moat, high ROIC, consistent free cash flow.
2. Each month, re-rank that universe on price rather than re-running the whole search. The monthly work is valuation, not discovery.
3. Keep the universe stable so that month-to-month price moves become the signal.

• A name entering the "cheap" list because the business deteriorated — the universe needs periodic re-qualification, not just re-pricing.

2. Start from the month's worst performers, not the best

1. Sort the universe by trailing one-month return and read the losers first: "The cheaper we can buy great companies, the better."
2. For each faller, identify the single reason given by the market, then judge whether it changes the ten-year cash flows.
3. Treat the winners list as information, not as a source of ideas.

• Repeat appearances on the losers list month after month — persistent underperformance usually means the market knows something the screen doesn't.

3. Size the news against the price move — the noise test

1. Quantify the disappointment in the units the business actually operates in (a growth rate, a guidance change), not in the units of the share price.
2. Set that magnitude against the market-cap move. A rounding-error miss producing a large decline is a sentiment event.
3. Then zoom out to intrinsic value — book value, NAV, cash flow per share — and check it is still compounding.

• The opposite case: a small headline that is really a leading indicator (a churn uptick, a first price-led slowdown) — the test is whether the metric is a level or a trend break.

4. Use the reverse DCF as a hurdle, and compare it to delivered growth

1. Solve for the free-cash-flow growth rate the current share price implies.
2. Line that number up against three references: the 10-year realised CAGR, the 3-year expected CAGR, and the long-term EPS estimate.
3. Buy when the implied requirement sits comfortably below all three; skip when it sits above them.

• A cheap-looking implied growth rate resting on a peak-margin FCF base — normalise the starting cash flow before trusting the hurdle.

5. Follow the disciplined operator onto the register — buy the playbook, not just the business

1. Track stake-building by acquirers with a proven operating playbook, and note the direction of travel (a stake rising over quarters, not a one-off block).
2. Write down the specific improvements the playbook prescribes — cash collection, value-based pricing, cost discipline, return-gated M&A — and turn them into a margin gap you can quantify.
3. Look for the first evidence the culture has landed inside the target (a cost review, a pricing change, a personnel move) before assuming it will.

• Whether the acquirer gains actual influence (board seats, governance) — a large minority stake without control can stall for years.

6. Invert the sign on a falling price when the company is a cannibal

1. Identify companies whose main use of cash is buying back their own shares.
2. For those, treat a lower price as an operational input, not just a mark-to-market loss: the same buyback budget retires more shares, so per-share value builds faster.
3. Confirm the buyback is funded from free cash flow rather than debt, and that share count is genuinely falling net of issuance.

• Buybacks financed with leverage into a cyclical downturn — the mechanism only works if the cash flow holds.

7. Separate a cyclical trough from a structural decline before buying the fall

1. Trace the demand shock to its cause and date it (a pull-forward, a stimulus, a one-off build cycle).
2. Ask what the boom left behind that generates recurring revenue afterwards — an installed base needing consumables and service is the tell.
3. Buy when the market is extrapolating trough earnings as permanent and the moat (scale, distribution) is untouched.

• Whether the installed base is actually growing; and whether the trough is being deepened by a competitor rather than by the cycle.

8. Publish candidates, not the book — and say why

1. Keep the monthly write-ups to names outside the current portfolio, so the analysis is a genuine evaluation rather than a defence of an existing position.
2. Treat the five as a ranked shortlist of candidates, not as a model portfolio.

• Which candidates later enter the portfolio — the conversion rate is the honest test of whether the monthly list is conviction or content.