The Full Investment Memo — A Complete Worked Example

First published 26 Aug 2026 · Last verified 29 Aug 2026

The Full Investment Memo — A Complete Worked Example

You have now spent ninety-eight chapters building tools. You have a Canon Score that rates a company across several dimensions. You have valuation methods from Chapters 45 through 49 that tell you what a business might be worth. You have sector-specific accounting rules that tell you what to trust and what to distrust in a bank's balance sheet versus a hydropower company's balance sheet versus an insurance company's balance sheet. You have governance checks that catch related-party mischief before it catches you. You have tax rules that tell you what you actually keep after NRB and the tax office take their share. And you have position-sizing rules that tell you how much of your precious capital to risk on any single idea.

The problem is that tools sitting in separate drawers do not build anything. A carpenter with a hammer in one room, a saw in another, and nails in a third has not built a chair. This chapter is about pulling every drawer into one workbench: a single document called the investment memo, written before you buy a single share, that forces every tool to work together at once.

Think of the memo the way a Kathmandu Valley engineer thinks of a structural drawing before a building goes up. The engineer does not carry the load calculations in her head and the soil report in a separate file and the client's budget in a third place, trusting memory to reconcile them at the moment concrete gets poured. She puts it all on one drawing, checked against itself, before anyone touches the ground. An investment memo is your structural drawing. If the numbers do not reconcile on paper, you find out in your notebook, not in your portfolio.

Lesson 99.1 — Why a Memo, and Why Now

A lot of Nepali retail investors buy shares the way people buy vegetables at Kalimati market: someone in a Viber group says "iyo ramro cha" (this one is good), the price is moving, and within twenty minutes an order is placed through Meroshare. There is no document. There is no thesis. There is a feeling.

Feelings are not free. A feeling-based purchase cannot be argued with later, because there was never an argument in the first place — only an impulse. When the stock falls 15 percent, the investor has no way to check whether the original reasoning has broken, because there was no original reasoning written down to check against. So the only available response is emotional: panic-sell at the bottom, or stubbornly hold because "it will come back," which is not analysis, it is hope wearing analysis's clothes.

A memo fixes this by making you write down, in advance, why you are buying, what you expect to happen, and at what price the thesis would be proven wrong. This is the single most useful habit in this entire book, more useful than any individual ratio, because it is the habit that makes every other tool actually count for something.

You do not need a memo for every trade. If you are doing short-term technical trading with strict stop-losses on a small trading sub-account, a full memo is overkill — Part XIV already gave you the tools for that game, and it plays by different rules. But for anything you intend to hold for more than a few months, anything that will occupy more than 2 or 3 percent of your portfolio, and certainly anything you are tempted to call a "core holding," a memo is not optional. If you cannot write one, you do not understand the company well enough to own it. That sentence is worth underlining in your notebook.

A full memo has six sections, and this chapter walks through each one, then shows you a complete example from beginning to end so you can see how the pieces click together in practice.

Lesson 99.2 — Section One: The Snapshot and the Thesis

The first section of a memo is short — half a page, no more — and it does one job: it forces you to say, in plain words, what the company actually does and why you are spending your evening on it instead of on one of the other 270-odd companies listed on NEPSE.

Start with the boring facts. Company name, ticker symbol, sector classification as NEPSE defines it, paid-up capital, current market price, and the date you are writing the memo (prices move daily; an undated memo becomes useless within a month). Then, in two or three sentences, describe the business the way you would describe it to your father over dinner, without jargon. If a company's business cannot be explained in three plain sentences, that itself is information — either the business is genuinely complex (rare on NEPSE outside a few conglomerates) or you do not understand it well enough yet to own it.

Then write the thesis: one or two sentences stating why this stock, why now. "I believe NGTL is undervalued because the market is pricing it as a low-margin commodity trader, but its LPG distribution license and captive retail network give it a moat the market is not paying for" is a thesis. "It has been going up" is not a thesis, it is a chart observation, and charts are the subject of a different Part of this book, not this one.

This section should also record how you found the idea. Did a screen flag it on low P/E? Did you notice the product in daily life — LPG cylinders with the company's logo showing up more often at your neighbourhood shop? Did a sector you already understood (because you or your family works in it) throw up a name you recognised? Recording the source matters because some sources are more reliable than others, and over years of memo-writing you will learn which of your own idea-generation habits actually produce winners and which ones just produce activity.

Lesson 99.3 — Section Two: The Canon Score and Qualitative Diligence

This is where the bulk of the qualitative work from earlier in this book gets assembled into one scorecard. Recall that the Canon Score is not a single number pulled from a formula — it is a disciplined walk through several dimensions of business quality, each scored honestly, so that a strong score in one area cannot quietly hide a rot in another.

For a NEPSE company, the dimensions worth scoring in every memo are:

Business durability — does this company sell something people will keep needing in five years, regardless of who is in government or what the monsoon does? A cement company selling into a country still building roads and buildings scores well here. A company selling a single fad product scores poorly.

Management and promoter quality — has the promoter family or leadership shown, across cycles, that they treat minority shareholders as partners rather than as a source of cheap capital to be diluted or ignored? Do they communicate honestly in AGM reports, or do they bury bad news in footnotes?

Governance and related-party exposure — this is worth its own line, separate from general management quality, because Nepal's related-party disclosure norms leave real gaps. Score this by actually reading the related-party transaction notes in the annual report, not by assuming.

Financial strength — debt levels appropriate to the sector (a bank naturally carries far more leverage than a trading company, and judging both by the same debt-to-equity yardstick is a rookie mistake), working capital discipline, and consistency of cash generation versus reported profit.

Growth runway — is the addressable market still expanding (more remittance-fed consumption, more hydropower demand, more insurance penetration as awareness grows), or is this company competing for a shrinking pie?

Sector and macro sensitivity — how exposed is this business to NRB policy shifts, monsoon timing, import restrictions, or exchange rate movement? A company financing itself in dollars for imported inventory carries a risk a purely domestic retailer does not.

Liquidity and tradability — how easily could you actually sell this position at a fair price if your thesis breaks? A thinly traded counter where your own order would move the price by 3 percent is a real cost that a memo must acknowledge, not a footnote to ignore.

The governance dimension deserves particular care for any company controlled by a single family group, which describes a large share of NEPSE's trading, manufacturing, and hospitality sectors. Check promoter shareholding percentage and whether it has been quietly rising (accumulation, generally a good sign) or falling (promoters cashing out, a signal worth investigating) over the last several AGM disclosures. Check whether board seats include genuinely independent directors or merely family members and their business associates. Check whether related-party transactions — goods bought from or sold to a sister company under the same promoter group, loans given to associate companies, rent paid to a promoter-owned building — are disclosed at arm's-length pricing or simply disclosed as a number with no comparison to market rates.

Lesson 99.4 — Section Three: Valuation — Bringing Chapters 45 Through 49 to the Same Table

This is usually the longest section of the memo, and it is where investors most often cheat themselves by running one method, liking the answer, and stopping. The discipline this book has tried to build in you is triangulation: run at least two, ideally three, independent valuation methods, and treat large disagreements between them as information rather than as noise to be averaged away.

Relative valuation (Chapter 45) starts with the simplest question: what multiple of earnings or book value is the market paying for comparable companies in the same sector, and how does this company's multiple compare? For a trading company, price-to-earnings and price-to-book are both usable, but earnings for a low-margin trading business can swing sharply on inventory gains or losses from currency movements and commodity price cycles, so a single year's P/E can be misleading. Better practice: compute P/E on a three-year average normalised earnings figure, smoothing out one unusually strong or weak year.

Dividend discount modelling (Chapter 46) asks what a share is worth as a stream of future cash dividends, discounted back to today at a rate that reflects the riskiness of those cash flows. This method works best for companies with a long, stable dividend history — banks, insurers, hydropower companies with power purchase agreements — and works poorly for companies that reinvest most earnings into working capital rather than distributing them, which describes many trading companies. Use it, but weight it lightly for this sector.

Discounted cash flow (Chapter 47) is more work but more honest for a working-capital-heavy business: project free cash flow (operating cash flow minus capital expenditure) forward, discount it back, and add a terminal value. For a trading company, the key driver is not revenue growth so much as working capital efficiency — how much cash gets tied up in inventory and receivables to support each rupee of sales growth. A trading company growing revenue 15 percent a year while inventory grows 25 percent a year is burning cash to grow, and a DCF will catch that even when a simple P/E multiple will not.

Asset-based or net asset value approaches (Chapter 48) matter more for companies where the balance sheet itself, not the earnings stream, is the main source of value — real estate holding companies, some finance companies, and mutual funds, where net asset value per unit is close to the entire investment case. For an operating trading company this method is a sanity check rather than a primary tool: does the market price sit reasonably above tangible book value, given that book value alone ignores brand, distribution network, and licenses?

Sector-specific and sum-of-the-parts adjustments (Chapter 49) matter whenever a company is not really one business but several stapled together — a trading house with a hotel subsidiary, or a manufacturer with a captive finance arm. Value each piece with the method appropriate to its own sector, then add them up, rather than applying one blended multiple to a business that does not deserve one uniform multiple.

Do not skip the sanity check of comparing your multiple to the sector's own history and to comparable listed peers. A trading company multiple that looks cheap against the NEPSE market average but expensive against its own five-year trading range, or against two or three direct peers in the same trading and distribution business, is not actually cheap — it is cheap relative to the wrong benchmark.

Lesson 99.5 — Section Four: Risks, Taxes, and Position Sizing

Every memo needs an honest risk register — not a generic list copied from the front page of an annual report's "risk factors" boilerplate, but the two or three things that would specifically break your thesis if they happened. Write them as trigger conditions, not vague worries. "NRB tightens import margin requirements on this product category" is a risk you can watch for in the news. "Things could go wrong" is not a risk, it is an admission that you have not thought hard enough.

For a company with import exposure, currency risk deserves its own line: since the Nepali rupee is pegged to the Indian rupee but Nepal's trade with third countries (China, and increasingly cross-border trade settled in other currencies) exposes importers to exchange rate movement that the peg does not fully absorb, a trading company's margins can be squeezed by a depreciating currency raising the landed cost of imported goods faster than retail prices can adjust.

The taxes-and-costs section is the one investors skip most often, and it is the one that quietly eats the most return over a holding period of several years. Two tax lines matter for a NEPSE equity holding: capital gains tax and dividend tax.

Capital gains tax in Nepal is charged on the profit from selling shares, and the rate depends on how long you held the shares and, for individuals, differs from the rate applied to institutional or corporate holders. Shares held longer than the threshold that separates short-term from long-term treatment attract a lower rate than shares sold within a shorter window, which is one more reason a memo written with a multi-year holding horizon in mind, rather than a trading mentality, tends to compound better after tax. Dividend income is subject to withholding at source — the company deducts tax before the dividend reaches your Demat-linked bank account, so what lands in your account is already net of that withholding, and no further separate filing is typically needed for that income alone if it is your only source subject to final withholding, though your overall tax situation should always be checked against current Inland Revenue Department rules, since rates and thresholds do shift with each budget.

DP (Depository Participant) fees, broker commission on both the buy and the sell leg, and SEBON-mandated regulatory fees are smaller than tax but not zero, and a memo on a stock you plan to trade in and out of frequently should account for round-trip costs eating into any expected edge, particularly on a name where the expected mispricing is modest.

Position sizing is the last piece, and it is where all the analytical work above gets converted into an actual number of shares to buy. The core idea, covered in earlier Part XVII chapters on the Investment Constitution, is that conviction and risk should scale your position size, not your excitement level. A stock where your valuation work shows a wide margin of safety, a strong Canon Score, and a business you deeply understand earns a larger position than a stock that merely looks statistically cheap on one multiple with a weak governance score.

A practical sizing method many disciplined NEPSE investors use is a simple conviction ladder: assign the position an initial size based on the combination of Canon Score and margin of safety (a high score plus a wide safety margin justifies a fuller initial position; a merely adequate score or a thin safety margin justifies a starter position, perhaps half the eventual target size), then set specific price levels at which you would add to the position (if it falls further while the thesis remains intact) or trim it (if it approaches your fair value estimate and the safety margin has closed). Writing these trigger prices into the memo before you buy — not deciding them emotionally in the moment three months later — is what turns a memo from a research document into an actual operating system for your money.

Lesson 99.6 — The Complete Worked Memo: Nepal General Traders Limited

Everything above is now assembled into one continuous document, exactly as an investor would write it in a notebook the evening before placing an order. The company below, Nepal General Traders Limited, trading under the ticker NGTL, is a composite built for teaching purposes, styled on the kind of diversified trading house long listed on NEPSE that distributes essential commodities — edible oil, sugar, LPG cylinders, and construction-grade steel — through a national network of depots and retail tie-ups. The figures are illustrative, built to be realistic for the sector, not a live quote; always pull current audited figures from the company's own disclosures and the NEPSE/Merolagani data feeds before acting on any real memo.

Section One — Snapshot and Thesis

Company: Nepal General Traders Limited (NGTL). Sector: Trading. Paid-up capital: Rs 1.2 arba. Current market price: Rs 486. Shares outstanding: 1,20,00,000. Market capitalisation: Rs 5.83 arba. Date of memo: Ashoj 2082 (mid-September 2026).

Business description: NGTL imports and distributes edible oil, sugar, packaged LPG cylinders, and construction-grade steel through nine regional depots and a network of roughly 4,000 retail dealer tie-ups across all seven provinces. It holds an LPG bottling and distribution license, owns its own bonded warehouse facility near the Birgunj dry port, and operates a small captive customs-clearing subsidiary that handles roughly 60 percent of its own import volume, with the remainder cleared through third-party agents.

Thesis: The market prices NGTL as a plain commodity trader on a P/E multiple near the bottom of the trading sector's range, but this ignores two durable advantages — the LPG distribution license, which is not easily replicated given current licensing constraints, and the owned bonded warehouse and partial in-house customs clearance, which gives NGTL a real cost and speed advantage over competitors who rely entirely on third-party clearing agents, especially during the pre-Dashain and pre-monsoon stocking seasons when Birgunj customs congestion is worst. This operational moat is not visible in a simple P/E comparison against peers and is why the stock screens as merely average when it is, on inspection, better positioned than average.

Idea source: Noticed NGTL-branded LPG cylinders becoming more common at retail shops in a mid-hills district over the past year, prompting a look at the company's distribution footprint disclosure in its most recent annual report.

Section Two — Canon Score and Diligence

Overall Canon Score: 3.1 out of 5, weighted toward business durability and financial strength, held back by governance and macro sensitivity. This is a solid, unglamorous business, not an exceptional one — a distinction the sizing decision in Section Six will respect.

The related-party item flagged above deserves a specific note: the customs-clearing subsidiary that handles 60 percent of NGTL's import volume is majority owned by the same promoter family outside the listed company. The annual report discloses the value of clearing fees paid to this subsidiary each year but does not benchmark that fee against what an independent clearing agent would charge for comparable volume. This is not proof of overcharging, but it is a gap in disclosure that a memo must record rather than gloss over, and it is worth a direct question to the company secretary at or before the next AGM.

Section Three — Valuation

Three-year average normalised EPS: Rs 34.20 (smoothing out an unusually strong year driven by a one-off inventory gain during a period of rising edible oil prices, and a weak year affected by Birgunj customs disruption). At the current price of Rs 486, this gives a normalised P/E of 14.2x, compared to a trading-sector average on NEPSE closer to 16-17x over the trailing three years, and compared to the two closest listed peers trading at 15.8x and 17.1x respectively on the same normalised basis.

Book value per share: Rs 268. Price-to-book of 1.81x, in line with sector peers running between 1.6x and 2.0x, so book value alone does not suggest either a bargain or an overpayment — this method is a sanity check here, not a primary driver, exactly as expected for an operating trading business rather than an asset-holding one.

Dividend history: NGTL has paid a cash dividend in nine of the last ten years, averaging a payout ratio near 40 percent of normalised earnings, with the one skipped year explained by the customs disruption event above rather than by a change in policy. Applying a dividend discount approach with a conservative long-run growth assumption of 7 percent (roughly tracking nominal GDP growth plus a modest margin for volume growth in LPG and packaged foods) and a discount rate of 13 percent (reflecting the sector's macro sensitivity noted in the Canon Score) produces a fair value estimate near Rs 470 in the base case, Rs 410 in a conservative case assuming slower dividend growth, and Rs 560 in an optimistic case assuming faster LPG segment expansion.

Discounted cash flow, built on free cash flow after accounting for the working-capital drag of inventory and receivable growth (both trending up over the last two years, a point flagged in the financial strength row above and treated here as a real cost, not ignored), produces a base-case fair value estimate near Rs 455, with the model notably more sensitive to the working-capital assumption than to the revenue growth assumption — confirming that the real investment question for this company is working-capital discipline, not top-line growth.

The three independent methods (relative, DDM, DCF) cluster in a fairly tight base-case range of roughly Rs 455 to Rs 480, against a current market price of Rs 486. This is the most important sentence in the entire valuation section: the stock is trading almost exactly at the base-case fair value cluster, not meaningfully below it.

Section Four — Risks

The three risks that would most directly break this thesis, in order of how closely the company monitors them internally versus how much is outside its control:

First, an NRB directive raising cash margin requirements on LC-financed imports for edible oil or LPG-related equipment would raise NGTL's working capital costs with little lead time, squeezing margins in the following one to two quarters until pricing adjusts.

Second, sustained rupee depreciation against currencies used for the portion of imports sourced outside India would raise landed costs faster than retail prices can be adjusted, particularly painful in the steel segment where competition is more price-sensitive than in packaged LPG.

Third, the unresolved related-party question around the customs-clearing subsidiary could, in a worse scenario than currently assumed, represent a slow transfer of value out of the listed entity — a risk that is not currently large enough to override the thesis but that should be re-checked at every annual report cycle, specifically watching whether clearing fees as a percentage of import value are rising over time.

Section Five — Taxes and Costs

Holding NGTL with a multi-year horizon in mind (this memo's stated intention, consistent with the thesis in Section One) places any eventual sale in the long-term capital gains category rather than the short-term category, provided the holding period at time of sale exceeds the threshold in force under the fiscal year's Inland Revenue rules — a rate meaningfully lower than the short-term rate, and worth confirming again at the time of sale since budget announcements can shift these thresholds. Dividends received will arrive net of withholding tax deducted at source by the company before credit to the linked bank account. Broker commission and DP charges apply on both the entry and the eventual exit trade and are treated in this memo as a modest, fixed drag rather than a factor influencing the buy decision itself, since they are small relative to the position size being considered.

Section Six — Decision and Sizing

Fair value cluster (Section Three): approximately Rs 455 to Rs 480 across three independent methods. Current market price: Rs 486. The stock sits essentially at fair value, with the base case showing no meaningful discount and the low case across methods sitting below the current price, meaning a genuinely conservative read of the numbers suggests the market is not offering a margin of safety today.

Canon Score (Section Two): 3.1 out of 5 — a sound, durable business with a real but narrow operational moat, held back by a governance question that remains unresolved and by real sensitivity to NRB policy and currency movement.

Recommendation: Hold for watching, not Buy at the current price. This is not an Avoid — the business quality supports a position in the portfolio at the right price, and the thesis (an underappreciated LPG and logistics moat) remains intact and worth tracking. But the valuation work does not show the margin of safety this book has argued you should require before committing fresh capital, and the unresolved related-party question further argues against sizing up at a price offering no cushion for being wrong.

Action: Add NGTL to a watchlist with a target entry zone of Rs 410 to Rs 430 — near the conservative-case low end across the DCF and DDM methods — where a genuine margin of safety would open up. If the price reaches that zone without a change in the underlying thesis, initiate a starter position sized at roughly 3 percent of portfolio, reflecting the moderate (not high) Canon Score, with a plan to review sizing upward toward a fuller 5 to 6 percent position only after the next annual report resolves the related-party clearing-fee question with clearer disclosure. Set a specific re-check trigger: revisit this memo immediately if NRB issues any new import margin directive affecting edible oil or LPG-related categories, and revisit it at the next AGM specifically to reread the related-party transaction note.

This is what a complete memo looks like: a snapshot and thesis that state plainly what the bet is, a Canon Score that scores the business honestly across several dimensions rather than one favourable headline number, a triangulated valuation that uses at least two independent methods and reports where they agree and where they diverge, a risk register with specific trigger conditions to monitor rather than vague worries, a tax section that reflects what you actually keep, and a sizing decision that respects the conviction level the rest of the memo actually earned rather than the excitement level the idea generated. Every tool from the previous ninety-eight chapters shows up somewhere in these six sections. That is the entire point of the memo format: nothing you learned in this book is meant to be used alone.