Editorial Policy

Last updated: 5 August 2026

This policy states how figures on this site are sourced, what the commercial relationship behind it may and may not influence, how corrections are handled, and which editorial decisions are ruled out entirely. It is written to be checkable: every rule below produces a visible consequence somewhere on the site, and where it does not, the rule is not worth publishing. Readers who want to test it should open Best Ad Networks and try to find a figure that cannot be traced to a platform's own documentation.

Source hierarchy

Facts are ranked by source, and the ranking decides whether a number can be published as fact, published as contested, or left out. This is the single most important rule on the site because it is the one that produces different output from every competing page.

• Tier 1 — the platform's own documentation. Help center articles, terms of service, payment and pricing pages. Figures published without qualification come from here, with the date they were read recorded on the page. Dates matter because these documents change silently and carry no version history.
• Tier 2 — directories and third-party reviews, flagged. Used only to establish that a figure is disputed. When a directory and a platform disagree, both versions are published with a note about which we would act on. A Tier 2 figure is never presented as settled.
• Tier 3 — excluded entirely. Other comparison articles are not a source. They are read to understand which questions readers ask and which are answered badly, never to lift facts. Copying between comparison pages is what puts stale figures into circulation and keeps them there for years.

Some platforms publish no figure at all. In that case the page says so. Absence is reported as absence and never filled with an estimate, an industry average or a number borrowed from a directory entry.

Handling contradictions

Networks contradict themselves more often than readers expect. One platform in the main comparison publishes four account tiers under one set of names on its homepage. Its blog prints the same four tiers under older names. A legacy page on the same domain lists three tiers and a lower entry deposit. All three were live on the same afternoon.

The editorial response is to publish the contradiction. Choosing the convenient version would be an editorial judgement dressed up as a fact, and the reader loses the information that matters most: that the platform's own material cannot currently be relied on without checking which page it came from. Where a judgement is unavoidable, it is stated as a judgement, with the reason attached.

Experience claims and first-person writing

First-person statements on this site are restricted to research that was actually performed and can be repeated by a reader at no cost: reading a help center on a stated date, opening a registration flow to see whether a deposit is demanded before pricing appears, comparing a network's pages against one another.

Measured claims — payout arrival times, fraud filter rejection rates, conversion rates by GEO, campaign results of any kind — are not published unless a campaign was actually run and recorded. None of that kind is currently published here. Some pages would be stronger with that data in them. The gap is left visible anyway. A fabricated measurement is checkable by anyone with an account, and one exposed falsehood devalues every honest figure beside it.

The commercial relationship

Sign-up buttons on this site point to AdsCompass and carry tracking parameters identifying this site as the referrer. An account opened through one may earn a commission. There is no other revenue model: no display advertising, no sponsored posts, no paid placement in tables, no paid links.

Three constraints follow. Each one shows up in the published output, where a reader can check it:

• Factual columns are immutable. Launch year, minimum deposit, minimum withdrawal, payout schedule, pricing models and formats are what the platform's documentation says, for every row including the first one. Any reader can check each cell against the source in a few minutes, which is precisely why altering them would be self-defeating as well as dishonest.
• Drawbacks are named, not softened. The main comparison states that the commercial partner's publisher withdrawal threshold is ten times higher than two competitors, that its default payout cycle is slower than the weekly networks, that it publishes no site-approval criteria, and that it offers no automated CPA-goal bidding. Those sentences cost conversions and stay in.
• Competitors get identical treatment. Every platform in a comparison is described from the same tier of sources, on the same criteria, with its strengths stated as plainly as the partner's.

What the relationship legitimately affects is emphasis. Where the data supports a recommendation, the recommendation is stated openly, and the reasoning behind it is shown so a reader who weighs the criteria differently can reach a different conclusion from the same table. Hiding a commercial preference behind artificial neutrality would be less honest than stating it.

No scores, ratings or rankings

This site publishes no star ratings, numeric scores, "best of" positions or ordered rankings. A single order implies a single metric, and no single metric serves both an advertiser choosing on format coverage and a publisher choosing on payout speed.

Tables are ordered for readability and the reasoning is written out in prose so readers can apply their own weighting. Recommendations are scoped to a situation: a first test on an unfamiliar GEO, a small site living on weekly cash flow. No universal verdicts.

Corrections procedure

Every substantive correction is logged with a date on the page it affects. Nothing gets edited silently. That log already records three corrections from the initial research. Two of them concern figures still circulating elsewhere in outdated form.

Corrections sent in with a link to primary documentation are normally applied within one working day. Disputes about a figure are settled by re-reading the platform's own material; if that material contradicts itself, both versions go in; picking one would hide the part worth knowing. Requests to remove an accurately reported drawback, or to add a rating, are declined — the site publishes no ratings at all. Corrections can be sent through the contact page.

Updates and review

Advertising terms move faster than comparison content does. Pages carry the date figures were checked so readers can judge how much weight to give them. Material revisions go back to the source document. This site makes no promise of a fixed review cadence, because a promised quarterly review that silently lapses is a worse trust signal than none at all.

The person applying these rules, and the limits of what he claims to know, are described on the author page; the funding model and scope of the site are set out on the about page. Where editorial standards end and legal obligations begin is covered by the terms and conditions, and data handling by the privacy policy.