Editorial policy
Editorial policy
What we promise about accuracy, how corrections work, and why the dates on this site mean something.
If we are wrong
Corrections
Email contact@psiandhose.com with the page URL and the figure you believe is wrong. You do not need to explain why - pointing at it is enough, and we will check it against the manufacturer's current specification sheet.
If the manufacturer's figure and ours disagree, theirs wins. Our numbers are dated to September 2026 and manufacturers revise spec sheets, which makes specification drift the most likely kind of error on this site rather than an unlikely one.
Corrections are made on the page itself and the modified date is updated at the same time. We do not publish a separate corrections log, because a correction buried on a log page while the error stays live on the article is worse than useless.
If a correction changes a recommendation - a pick that no longer clears the bar, a spec that moves a product down the ranking - we change the ranking rather than leaving it and footnoting the problem.
How often pages are checked
Review cadence
Different content decays at different speeds, so it is reviewed on different schedules. Product availability moves fastest; engineering standards barely move at all.
| Content type | Cadence | Why |
|---|---|---|
| Roundups | Quarterly | Stock and model availability shift fastest, and we have no automated availability feed |
| Job pages | Quarterly | Same product exposure as roundups |
| Reference charts | Six months | Specifications are stable; a reported correction is the real trigger |
| Comparisons | Six months | Product generations change and can make a comparison stale |
| Guides | Annually | Largely evergreen engineering |
| The calculator | Annually | A formula audit rather than a data refresh |
| Trust pages | Annually, or on any policy change | These describe how we work, so they change when that does |
What actually gets checked
The quarterly roundup check
- 01Is every pick still sold? A discontinued pick is removed rather than left with a dead link, and the ranking is rebuilt without it.
- 02Does every outbound link still resolve, and does it still carry the correct affiliate tag?
- 03Do the published specifications still match the manufacturer's current sheet? This is where most corrections come from.
- 04Has a new model appeared that beats an existing pick on value? If so it goes in, and the displaced pick either moves down or comes out.
- 05Is the modified date updated - and only if something in the above list actually changed?
No fake freshness
Dates mean something here
The modified date on a page changes only when the content substantively changes. Fixing a typo, swapping an image or adjusting a heading does not move it. Correcting a specification, changing a recommendation, adding a section or rewriting an argument does.
This is deliberate and it costs us something. A site that re-stamps every page with today's date looks fresher to a crawler, and several competitors in this category are visibly ranking with titles bearing a stale year. We would rather show an honest older date than a dishonest recent one - and a sitemap where every entry claims to have changed on every deploy trains crawlers to ignore the field entirely, which throws away the one freshness signal we have.
Our sitemap's last-modified values are read from each page's real content date for the same reason, never from the build timestamp.
Not waiting for the cadence
What triggers an immediate update
- A pick is discontinued or becomes permanently unavailable.
- A published specification we rely on turns out to be wrong, or the manufacturer revises it in a way that changes a recommendation.
- A reader reports a factual error and we confirm it.
- A safety-relevant error - a pressure rating, an electrical figure, a filtration limit. These are corrected as soon as we see them, ahead of anything else.
- A change in how the site is funded or how picks are chosen. That changes the trust pages the same day.
How we keep it honest
Independence from commission
Commission rates do not enter the ranking logic. Picks are ordered by the cheapest unit that clears a job's published CFM demand plus our margin, and that ordering is computed from specifications rather than from what pays best.
The checkable evidence is in the content rather than in this promise. The tire page's top pick is a 12V inflator rather than a compressor. Multiple pages recommend an electric sander or an electric drill we earn nothing on. Several roundups have a section explicitly arguing you should spend less. A site optimizing for commission would not contain any of those.
We also do not accept payment for placement, do not run sponsored posts, and do not publish anything supplied by a manufacturer as editorial content. If that ever changes, this page changes with it.
Hard lines
What we will never publish
- Fabricated reviews, testimonials, ratings or review counts. There are none on this site because there are none to report.
- Claims of testing we have not done. No 'in our testing', no 'after 40 hours', no 'on our bench'.
- Invented credentials, licenses or years of experience.
- A price we cannot verify. There is no live price feed here, so no price appears anywhere.
- A specification we estimated. Not published means not published.
- A stock photograph captioned or implied to be our own product photo. Our imagery is licensed stock and is labeled as illustrative on product cards.
Sources
Where these numbers come from
- Compressed Air & Gas Institute - Compressed Air & Gas Handbook
The industry standards body for compressed air terminology and system design.
- Ingersoll Rand - Air Consumption Chart for Automotive Service Shops
Figures are for average service-shop usage. The chart carries its own caveat: "Always check with tool manufactures for actual air consumption of tools being used. The above is based on averages and should not be considered accurate for any particular make of tool."
- Engineering ToolBox - Air Tools, Air Consumption
Tool-class consumption at roughly 90 psi. The table states its own basis: "The ratings are based on 25% load factor - or tools running 25% of the time. Full performance requires 4 times the volume."
Accessed September 2026. If a figure on this page does not match the manufacturer's current sheet, tell us at contact@psiandhose.com and we will correct and date the change - see our editorial policy.