Every ink and toner box in the store carries a magic number — ~300 pages, ~2,600 pages, up to 6,000 pages. It looks like a promise. It’s actually the output of a standardized lab test, and understanding what that test does — and doesn’t — measure is the difference between trusting a marketing number and knowing what you’ll actually spend.

The three standards behind the number

Page yield isn’t self-reported guesswork; it’s defined by ISO/IEC test methods, one per technology:

standard	covers	test basis
ISO/IEC 19752	monochrome laser toner cartridges	standard test page, ~5% black coverage
ISO/IEC 19798	color laser toner cartridges	ISO/IEC 24712 test suite, ~5% per color channel
ISO/IEC 24711	inkjet cartridges	same 24712 suite, inkjet-specific end-of-life rules

The details differ — inkjet rules allow defined cartridge-shake points to simulate a user coaxing out the last drops, laser methods specify their own end-of-life criteria — but the skeleton is the same: print the standard pages continuously, on standard paper, until the cartridge declares itself empty, then declare that page count the yield. Tests run across multiple cartridges and multiple printers, and the published figure is the statistical average, not the best unit.

The “5%” you hear about is the test page itself: a standard business-letter-style page (the ISO/IEC 19752 chart for mono work; the five-page ISO/IEC 24712 suite for color and inkjet) whose printed area works out to roughly 5% coverage per channel. That basis is the whole reason yields are comparable — an Epson number and an HP number and a Brother number were all earned on the same page.

What 5% coverage looks like — and what it doesn’t

Five percent coverage is a plain page of double-spaced text. If your printing is letters, invoices and emails, you’re near the test basis and the rated yield is meaningful. Everything denser scales down from there:

• A dense report with headers, a logo and a chart or two: 10–20% coverage → your yield is roughly half to a quarter of rated.
• A full-page photo on an inkjet: effectively ~100% coverage on every channel → the cartridge that was rated for 250 pages is a ~12-page cartridge.
• Edge case nobody quotes: draft mode and nearly-blank pages (a form, a signature line) run above rated yield — your 300-page cartridge becomes a 500-page one.

This is exactly the model the calculator implements: effective yield = rated yield × 5 ÷ your coverage percent. It is a linear approximation — the honest caveat is that real cartridges do slightly worse at high coverage, because printheads still clean themselves and fusers still warm up regardless of what the page looked like. For ranking two printers it is accurate enough; for promising a client their per-page invoice, pad it by 10–20%.

Why your real yield still differs at the same coverage

Even printing the exact ISO test page, you won’t hit the rated number in real life, for three structural reasons:

1. The test prints continuously. Real users print a page here, a page there. Every time an inkjet wakes up it can run a maintenance cycle that spends ink cleaning the head — ink that never lands on any page. A cartridge printed out in one afternoon out-yields the same cartridge spread over three months of Tuesdays. Laser toner doesn’t dry or purge, which is one of its quiet advantages for sporadic printers.
2. The environment fights you. Temperature and humidity shift ink viscosity and toner behavior; the standard specifies conditions your office doesn’t.
3. “Declared yield” hides two flavors. Some makers quote the average of the test distribution; others quote variants or add qualifiers like “in accordance with ISO/IEC 19752” while quietly using different paper or settings. The closer to a plain “ISO/IEC 19752” claim, the better.

And two non-yield drains nobody measures:

• Starter cartridges. The cartridges shipped inside a new printer are typically filled to 30–60% of a standard fill — the printer’s page-one economics are worse than its replacement economics.
• Separate drums. On printers where the imaging drum is a distinct part, it lasts several toner cartridges and then becomes a second consumable line item. Price it per page the same way: drum cost ÷ drum yield.

Using yield honestly

The rated yield is not a lie — it’s a benchmark. Two rules keep it honest in practice:

• Compare like with like. “ISO/IEC 19752 ~2,600 pages” vs “ISO/IEC 24711 ~6,000 pages” is a legitimate cross-technology comparison — same test philosophy, different machines. “Up to 50,000 pages monthly duty cycle” is a different number entirely (how hard the engine can run, not how much a cartridge holds) and means nothing for cost.
• Recompute for your coverage. The number on the box is measured at 5%; your pages are whatever they are. Thirty seconds in the calculator with a realistic coverage figure converts marketing yield into your yield — which is the only version that predicts a bill.

The benchmarks article puts these yields against real street prices to show what the numbers mean by printer class.

Frequently asked questions

Is a 2,600-page cartridge guaranteed to print 2,600 pages?

No — it's a standardized test result, not a floor. At exactly the ISO test's coverage (~5% per channel) you should land near it; denser pages, frequent small jobs and cleaning cycles pull you below, sparser pages push you above. Treat it as a comparison figure between cartridges more than a delivery estimate.

Why do starter cartridges run out so fast?

They're deliberately under-filled — typically 30–60% of a standard cartridge — to keep the printer's sticker price down. The box usually discloses a lower yield number for the bundled cartridges in the fine print. Replace them with standard or high-capacity cartridges and the quoted yield jumps accordingly.

What's a 'high-capacity' or 'XL' cartridge?

Same physical shell, more fill. An XL version might cost 1.5× a standard cartridge but carry 2–3× the yield — almost always a better cost per page, which is exactly the math the calculator is built to run. The catch: inkjet ink can expire, so XL carts only pay off if you print enough to finish them.

Do drum units count in the yield math?

Only if they're inside the cartridge. Many laser printers separate toner and drum — the drum lasts 3–5 cartridges before it needs replacing. When it does, its cost belongs in your CPP too: drum price ÷ drum yield, added to the toner term. Combined toner+drum cartridges already fold it into the cartridge price.

Who actually runs the ISO yield test?

The manufacturers themselves, on the standard test files and conditions — but the numbers get audited and contested (rival makers routinely buy each other's cartridges and re-test). It's not a perfect system, but it's a shared one, which is why yields across brands are meaningfully comparable in a way 'up to 50,000 pages duty cycle' never is.