The series

The Exceptional Company Series

EXCEPTIONAL CLASS OF 2026 5/5

What does exceptional actually look like in a company's earnings? These are the businesses that answer it — profits that have compounded durably, consistently, and for reasons that will still hold in ten years. Each one shows the full working. An assessment of business quality, never a valuation or a recommendation to buy.

Not a buy list. Exceptional can also be expensive. Why →

What exceptional means here

It is a measurement, not an opinion, and it is about earnings — whether profits have compounded durably rather than whether a company is admired. Every business is put through the same five structural tests, and passing all five earns admission.

The series is narrower. It takes only those that also clear an elevated bar on the record itself — at least ten years of earnings history, rising in at least 80% of year-on-year comparisons, a worst fall of no more than 25%, and compounding at 12% a year or better — and are strong today across those same five checks. Both, not either: a spectacular recent run with a scarred history does not qualify, and neither does a long clean record from a business that has since weakened. The full working →

The members

Written in the order we got to them — not a ranking, and not the complete list. The companies that clear the same bar and have no piece yet are named below.

Clears the same bar, not yet written about

33 companies meet the same twelve-year earnings record as the pieces above. They are not lesser, and they are not rejected — there is simply no piece yet. How every company is measured →

AAPLAAPLADBEADBEAMTAMTAZOAZOBROBROCOSTCOSTCPRTCPRTEMEEMEEVVTYEVVTYFICOFICOFIXFIXGMEDGMEDGOOGLGOOGLINTUINTULOWLOWLULULULUMETAMETAMPWRMPWRMTDMTDMYRGMYRGNVDANVDANVONVOODFLODFLPAYCPAYCROLROLSPSCSPSCTDYTDYTMETMETSMTSMTTDTTDTYLTYLVVVEEVVEEV

Eleven of the twenty-one companies written up on this basis so far. Twenty-seven have a profile on this site; of the other six, four are assessed against different tests entirely and two have too short a reported record for any grade to be computed — which is a fact about the test rather than about them. The series is meant to stay small; if it stopped being small it would stop meaning anything.

Written, not yet exceptional

Pieces exist for these two, and the measurement does not (yet) call them exceptional — each one's own line says exactly why. Not a shortlist: two of the 13 companies written up so far, no more special than the rest.


Earnings record of the pieces published so far

Each written company's trailing twelve-month earnings per share, rebased to 100 at the common start quarter, on a logarithmic scale so equal vertical distance means equal percentage growth. Covers only the 11 pieces published to date, not the wider set that clears the same bar — a company's absence from this chart says nothing about its earnings, only that nothing has been written about it yet. Not a portfolio, not a performance claim, and not a view on any share price.

Multiples of each company’s own starting earnings, trailing twelve months, on a log scale. The average is a compound one — the growth rate the members share, not the sum of their multiples divided by four, which one exceptional member would dominate. Amber marks show when we published — everything to the right of one is what happened after.

TickerCompanyEarnings growth
NFLXNetflix58.3×
FTNTFortinet36.3×
ANETArista Networks31.5×
IDXXIDEXX Laboratories7.7×
MSFTMicrosoft7.4×
EXLSExlService7.4×
ASMLASML6.9×
ISRGIntuitive Surgical6.2×
MAMastercard5.5×
CWCurtiss-Wright4.7×
RMDResMed4.2×

Earnings, not price. This series judges business quality and takes no view on share prices, so that is what the chart shows.

Not a portfolio. Earnings per share cannot be added across companies, so each member is indexed to its own starting point and the two series lines are the mean and the median of those. It holds nothing and assumes nobody bought anything.

No benchmark, deliberately. Members are picked for their earnings record, so comparing that record to an index would only prove the selection selected. The honest test is what happens after we publish — those windows have only just opened.

Survivorship. These are today’s members. One that lost the grade would leave, and this chart would flatter what remains. Past growth is not a forecast.


What this is not

So what is it for? The business question and the price question only work in that order, and only the first one keeps: understanding a company takes weeks and stays true for years, while what the market will pay for it can change in a month. Doing the durable half in advance is what makes it possible to think clearly when a price eventually does something dramatic — rather than meeting a company for the first time through a falling chart and a headline. How a great business can still lose you money →

What the grade means

Every company we assess goes through the same five structural tests, applied identically by Legend, our own rules-based framework:

  • Compounding ability — is growth real, and is it repeatable?
  • Earnings quality — are the earnings cash-backed, or engineered?
  • Balance sheet strength — could it survive a bad decade unaided?
  • Capital allocation — is cash deployed sensibly? If this series has a precedent, it is The Outsiders — eight companies chosen not for their industries or their stories but for what management did with the cash. Exceptional businesses, selected for much the same reasons.
  • Per-share value creation — does the compounding actually reach the shareholder, per share, after dilution?

Passing all five earns admission. The series is narrower than that: it takes the companies that also clear an elevated bar on two further measures — how strong the business is today across those same checks, and how consistent its earnings record has been across twelve years. Both, not either. A company with a spectacular recent run and a scarred earnings history does not qualify, and neither does a serene record that has gone quiet.


Why a company you expected is not here

Most companies we write about are not in this series, and that is the point rather than an oversight — a framework that admitted everything would not be measuring anything. Every company page states its own status plainly. The common reasons:

Banks
Assessed on capital adequacy and book-value compounding. The exceptional grade is built on the earnings line of a compounder, so a bank cannot earn it — not because it is a lesser business, but because the test does not fit.
Holding companies
Assessed on net asset value per share rather than earnings.
Funds and vehicles
Not single companies, so the structural tests do not apply.
Companies under ten years
The grade reads a twelve-year earnings record. A shorter history cannot be assessed for it, however strong it looks.
Assessed and not graded
The largest group: businesses that pass some or all of the structural tests without clearing the elevated bar. Most good companies land here, and saying so is the point.

Our research covers all of these. Some of the work we are proudest of is on businesses that will never carry this grade.


If a word here is doing work you cannot see

Compounding, earnings quality, per-share value creation — every one of them means something specific here, and none of them means what a broker means by it. The glossary defines the twenty-eight terms this site uses, in plain English and without assuming you already know.

And the tests are only half of it. The other half is which markets are worth looking at in the first place: the seven forces we underwrite against, also available as a five-page sheet.


A point in time, not a permanent record

The assessment is recalculated as companies report, and can change on any earnings release or other fundamental disclosure. A company can lose the grade. When that happens we will say so rather than quietly remove it — a membership list that only ever grows is not telling you anything. Nothing here updates itself.

The Conviction Play publishes general commentary and analysis for information only. It is not investment advice, not a personal recommendation, and not an offer or invitation to buy or sell anything. Capital is at risk and the value of investments can fall as well as rise.

© The Conviction Play 2026 · @theconvictionplay · Spotify · Subscribe · Privacy · Terms