Editorial standards

Every price on this site carries the month we checked it. That is the shortest way to explain how this publication works: nothing ships unless we can say where it came from and when.

The longer version is below. Hold us to it.

Numbers

Prices come from the provider's live pricing page, quoted with an "as of" month, and rechecked quarterly (and again whenever a comparison republishes). A price without a date is a rumor, so we do not print one.

External claims carry a named source and a date. When we write that Gmail hard-rejects unauthenticated mail, we cite Google's sender guidelines and the November 2025 change that made it so, not "recent updates." If we cannot name the source, the claim gets cut. "Studies show" does not appear on this site.

Our own measurements are labeled as ours. When we say Gmail rejected 73% of the mail we forwarded in June 2026, that figure comes from our delivery logs, and we tell you so. Our numbers describe our system at our volume. We do not dress them up as industry benchmarks, because they are not.

Experience

We publish only what we run. LetterDuck has carried a real media site's mail since June 2026: every address on the domain, the replies, the newsletter. Small volumes. Real recipients. Real consequences when something breaks.

The operational detail in our articles comes from that system, not from rephrasing the other nine search results. The canary probe that walks our inbound path every 2 hours is real and running. The throttle that pauses sends at a 0.3% complaint rate is code, not advice. When an article describes a suppression check that runs before every single send, it is describing our production send path.

Where our experience ends, we say so. We have never run a dedicated IP, so when we write about dedicated-IP thresholds we quote Postmark's and AWS's published guidance and mark it as theirs.

No affiliate links. Ever.

Nobody pays for placement here. Not for a ranking, and not for a green cell in a feature table. Our outbound links earn us nothing: no tracking parameters, no commissions.

We sell one thing, a flat-rate email workspace. That is the entire business model, and it is exactly why we can tell you when a competitor is the better pick. A publication funded by referral fees cannot afford that sentence.

AI assistance

Drafting here is AI-assisted, and we would rather say so than have you guess. A model helps produce the first pass; a person reviews every page, and checks each number against its source, before it goes live. Every page also has to clear a written checklist: dated sources, no invented tests or experience, no filler paragraphs that survive deletion. Parts of that checklist are enforced mechanically. The site will not build with an em-dash in the body copy, and that is a literal statement about our build script, not a figure of speech.

If a page is wrong anyway, the failure is ours, not the tool's. We own everything we publish.

Corrections

Found an error? Email hello@letterduck.com. We fix it and note the fix on the page rather than quietly rewriting history. Pricing mistakes get priority, since prices age fastest.

Why we compare honestly

A comparison that hides where competitors win is an ad, and readers can tell. So our comparison pages run the other way: the feature tables give competitors their genuine wins in green, and every verdict box names who should not pick LetterDuck. Google Workspace comes with docs, storage, and video calls. We are email only, and our comparisons say so on the first screen.

This is not generosity; it is the only way a comparison stays useful once the reader has three competitor tabs open. It is also the standard we will keep as the site grows, because trust compounds slower than traffic and is worth more.

Hold us to all of it. The address is above.