Editorial Policy

Independence

These are commitments, not aspirations.

• No press samples.We do not accept free or loaned units from manufacturers or retailers. If we own a machine, we bought it at retail, and we say so with the date. A free machine is not free — it is paid for with the reader's trust.
• No paid placement. No one can buy a recommendation, a ranking, a mention, or a score. There is no rate card, because there is nothing for sale.
• No pre-publication review.No manufacturer, retailer or affiliate partner sees a page before it goes live, and none of them gets a right of reply built into the process. They can email us like anyone else, and if they show us we're wrong, we'll correct it like anyone else.
• Commission does not affect rankings. Our scoring componentsare published in full, and "what it pays" is not one of them. Where the better product for the buyer earns us less — or nothing — we recommend it anyway. See the affiliate disclosure for concrete examples, including a grinder we recommend that we earn nothing from.

Sourcing standards

Every factual claim on a product page has to trace back to something you can check, or to reasoning we've shown you. In practice that means:

• Specificationscome from the manufacturer's own product page, spec sheet or instruction manual, and the source is linked under every spec table. Never from a retailer listing, never from another review site, never from memory.
• Prices come from the Amazon API with a visible timestamp. A price is never typed into an article — our product database has no field to store one in.
• Reliability and behaviour patternsare aggregated from what owners publicly report, and are described as "owners report" — never as "we found."
• A verdict is ours, is labelled as ours, and shows its reasoning.

If a claim cannot be traced to a linked source or to our own stated reasoning, it does not ship.Where a manufacturer simply doesn't publish something, we print that they don't publish it, rather than filling the gap with a number from somewhere else.

What we will never publish

• A testing claim for testing we did not do.
• A review, rating, testimonial or owner quote that we invented.
• A specification we could not source to the manufacturer.
• A price we did not pull live from the retailer.
• A rating in our structured data that implies hands-on testing we haven't done. Our schema carries the same editorial verdict the page shows you, and nothing more.

Corrections

We are going to get things wrong. Here is what happens when we do.

• Tell us: info@theespressoreport.com, or the contact form. Replies go straight back to you.
• We fix it within 48 hours of confirming the error.
• We note the correction on the page, with a date. We do not silently edit a page and pretend the mistake never happened. A site that quietly rewrites its history is a site you cannot trust about anything else either.
• If we got something badly wrong, we'll say so plainly rather than burying it in a footnote.

Updates, and not faking them

Every commercial page — reviews, comparisons, roundups, pairing guides — gets reviewed quarterly. Specs get checked when a model is refreshed or replaced. Prices are live by construction, so they never go stale.

We do not fake freshness.The "last updated" date on a page changes only when the page substantively changed. We are not going to bump a date to 2026 to look current to Google when nothing about the page moved. That trick is common, it's transparent, and it's a small lie told to a search engine at the reader's expense.

The corollary: if a page here says it was updated in July 2026, something on it actually changed in July 2026.

On AI

We use software to gather manufacturer specifications and to keep prices current — that is what the price pipeline is. Editorial judgment, the verdicts, and the writing are ours, and every claim goes through the sourcing standard above regardless of what tool helped find it. A citation to a manufacturer's spec sheet is a citation to a manufacturer's spec sheet, and an unverifiable claim is an unverifiable claim; where it came from doesn't change either.

Who's accountable

Stephen V. — a coffee enthusiast, with no lab and no certifications, which is stated plainly on the about page and in the byline of every article. There is nobody else to blame here, which is rather the point.