Who writes for this site, how methods are tested, how sources are cited, and what happens when we get something wrong. The shorter version of the long version of how this works.
Every piece on Window Washing Guide carries a byline, and every byline links to a contributor page on our masthead. Contributors are assigned a beat — regional, technical, or product — and pieces published under a byline are researched and reviewed in collaboration with the Giordano Inc. editorial team.
We do not buy syndicated content and we do not republish material from other sites.
If you want to know more about a contributor before trusting their work, every byline links to a full biography page that includes the beat they cover, how their pieces are researched, and our internal record of every piece they have written for us.
When a piece on this site recommends a method, a product, a ratio, or a tool, we have either tested it ourselves or relied on a documented source we can cite. The line between those two cases is drawn explicitly inside each article.
Where a claim rests on a documented source, that source is named in the article's Sources block and linked wherever the document can be opened. Where it rests on general trade practice rather than a specific document, the article says so rather than dressing the claim in a citation.
Where a source turns out not to exist, or to exist but not to support the claim it was attached to, we remove it rather than quietly swapping in the nearest real document by the same publisher.
Articles cite sources inline by footnote and aggregate them in a Sources block at the foot of the piece. Citations include manufacturer Safety Data Sheets and technical bulletins, peer-reviewed materials-chemistry and conservation literature, federal and state agency publications, and trade-organization standards (IRATA, IWCA, ANSI Z97.1, CPSC 16 CFR 1201, ASTM, NGA, FGIA).
When we make a claim that depends on a specific document, that document is cited, and we link it wherever the publisher makes it reachable. A citation we cannot open is not treated as provenance.
We do not cite Reddit threads, AI summaries, or unsourced TikTok videos as primary references, even when they are the proximate cause of a question we are answering.
Our contributors disagree, sometimes loudly, about substantive matters of trade practice. The longest-running disagreement is the vinegar question, which Mara and Easton have been litigating in some form for nine years and which is unlikely to ever fully resolve.
We treat this as a feature, not a bug. When contributors disagree on a topic that matters, the disagreement is stated explicitly in the relevant article rather than papered over with a consensus statement that nobody on staff actually believes. The article will name who holds which position and why.
We will not invent a fake unanimity for the sake of editorial smoothness. The trade is one in which reasonable working professionals look at the same problem and reach different conclusions. We owe readers an accurate picture of that, not a fabricated one.
When we get something wrong, we correct it in the original article and add a dated correction note at the foot of the piece. We do not silently rewrite history. We do not delete the wrong claim and replace it with the right one without acknowledgment.
If a correction substantially changes the recommendation in the article — for example, from "yes, you can use this on tinted glass" to "no, you cannot" — we update the headline, the social cards, and the search-engine description, and we add a banner at the top of the article calling the change out for at least thirty days.
If you find an error, write to us at corrections@windowwashingguide.com. We acknowledge corrections inside three business days, and we generally publish them within a week.
Every correction to a published figure or claim is listed, dated, in the corrections log.
Some product links in our articles are affiliate links, meaning that if you buy a product through one of them, we receive a small commission at no additional cost to you. When this is the case, the article discloses it.
We do not accept money from manufacturers in exchange for coverage. We do not run sponsored content disguised as editorial. We do not allow advertisers to read or influence pieces before publication. A product's presence in an affiliate slot does not buy it a recommendation in the prose.
If a recommendation in our coverage stops being one we stand behind, we change the recommendation, even when an affiliate link is in place. The link comes off. The honest answer goes up.
Window Washing Guide is produced by the Giordano Inc. editorial process, and that process uses AI tooling. We are not going to pretend otherwise. The claim we think readers should actually be able to check is not who typed a given sentence, but whether it is true and whether the document behind it opens.
So the standard we hold ourselves to is the one in section 03: a claim that depends on a specific document names that document, and the document is linked wherever its publisher makes it reachable. Sources that turn out not to exist are deleted rather than quietly swapped for something adjacent, and that sweep is run over the whole corpus repeatedly. It has removed real entries, including ones that had been sitting here for a long time.
If our policy on this changes, we will say so on this page. We will also say so in the articles affected. Readers should not have to guess about which words on a trade website are written by a person who has done the work and which are not.
Every figure in a city's water panel comes from the operator's own published document — its annual water quality report (the Consumer Confidence Report), or a lab sheet or water-quality page the utility itself publishes — or, for the source and treatment entries, from the federal record in EPA's Safe Drinking Water Information System. Figures from water-softener dealers, filter vendors and data aggregators are never used.
Each panel covers seven items across eleven fields: total hardness; pH, alkalinity and a corrosion index; iron and manganese; sodium, chloride and total dissolved solids; the disinfectant; the source and treatment; and whether results are reported for the whole system or per plant, zone or source. A field is either captured with the number the document prints, or marked not published after the document was read and the value was not in it. We never estimate a missing value.
Where a report prints a highest reading and a range, we store the range rather than presenting the highest reading as an average; where it prints an average, we show it. Units are converted only where the conversion is unambiguous — micrograms to milligrams, or grains per gallon to milligrams per litre at 17.12 — and where a report contradicts itself, the record says so rather than choosing the convenient number.
The hardness bands used for sorting and colour across the site follow the USGS thresholds. Where the operator publishes no hardness figure, the city page, the state page and the Water Atlas say so and print no band or grains figure for that city. The full method, including what each number means at the glass, is on the methodology page.
Some utilities cannot be read: the site refuses automated access, or the figures exist only behind an address lookup. Those cities are recorded as closed with the reason and every attempt listed, never filled in, and they reopen the moment a readable copy turns up. Every panel names its document and edition, and our build refuses to publish a captured field that lacks a value or a named source.
Tell us. Editorial concerns go to editorial@windowwashingguide.com. Factual corrections go to corrections@windowwashingguide.com. Both inboxes are read by a human within three business days.
ALL CONTACT METHODS →