The Draft That Starts the Review
A writer has drafted a credit card hub, added an “easy approval” line, and placed a related-search unit beside an offer table. The editor’s question is not whether credit card content can be monetized. It is narrower: which choices fall under personalized-advertising limits, and which wording falls under AdSense for Search promise rules? Merging them causes over-correction or missed risk. This article maps the draft to both, without product advice.
Gate One: Targeting, Not a Content Ban
In the US and Canada, Google’s personalized-advertising rules prohibit serving ads for credit-related products or services based on audience gender, age, parental status, marital status, or ZIP code. Banking products and services and certain financial planning services sit in the same category, and the policy names credit cards and loans as examples. Read this as a targeting restriction, not a content ban: the rule attaches to personalization signals, not the topic. A credit card hub may still be reviewed for content and layout compliance; what changes is audience selection. If a draft proposes “women aged 25–34 in one ZIP code” as the audience for a credit-card offer, remove those dimensions and document why, rather than delete the page’s credit content.
Creditworthiness Signals in Audience Lists
The second gate is data. Google’s policy treats negative financial-status information as sensitive, including signals that a user has a low credit rating or a high debt burden. That information may not select or target personalized ads or be collected through audience lists such as cookie lists. For a publisher, this is list hygiene: a remarketing segment built around visits to high-debt articles rests on a sensitive financial signal, not an ordinary interest. Retire such segments, keep rating or debt burden out of list logic, and define first-party audiences by neutral page context rather than inferred financial condition.
Promise Wording and the Strike Arithmetic
Many credit-card drafts fail at wording, not targeting. The AFS strikes table separates concrete promises the publisher cannot fulfill from vague promotional promises. A concrete, explicitly stated, impossible-to-fulfill promise is an egregious example: one violation equals one strike. Examples include free or cash offers, promising loans, unreasonably cheap offers, and a named account “with no credit check.” A promise that is not clear or explicit, or that offers specific information without supplying the real list, is a normal example: five violations equal one strike. That arithmetic is the only quantified enforcement detail in the evidence; it is not a forecast of outcomes. The comparison below uses only the policy’s own classifications and stated arithmetic.
| Promise pattern on the page | Classification in retrieved policy text | Stated consequence |
|---|
| Concrete, explicit promise outside publisher control (e.g., free or cash offers, promising loans, unreasonably cheap offers, named account “with no credit check”) | Egregious example of an impossible-to-fulfill promise | 1 egregious violation = 1 strike |
| Promotional phrasing with unclear or non-explicit terms, or a promise of specific information without the real listing | Normal example (promise made but not clear or explicit) | 5 policy violations = 1 strike |
Revision direction matters more than examples. Separate publisher-controlled facts from reader outcomes: a promise of approval, a named institution’s decision, or an account-opening result sits outside publisher control and is the riskier form. A vague “better options” line may be weaker, but it remains in the normal violation column and is not a safe harbor. Prefer scope-limited statements about what the page covers, what it links to, and what a reader can verify. Because no approval, rate, or fee data was verified here, do not manufacture such figures to make copy sound concrete.
Creative Descriptions and Query Origin
Two shortcuts need scrutiny. The referrerAdCreative (RAC) parameter must describe the upstream ad creative accurately; omitting or misrepresenting key details that meaningfully alter Google’s understanding is egregious, while a paraphrase that keeps every key detail is normal. Shortened credit-offer descriptions are risky when they drop defining terms. AFS queries must come from explicit user input, with no pre-filled terms, and publishers may not incentivize searching. Units belong on search results pages, at most two per page, and the AFS code must not be modified.
When Layout Makes the Unit the Story
Product-Integrated Feature (PIF) units may not sit on content that violates publisher policies, may not use misleading text such as “click here for the best offer” to attract interaction, and may not be the page’s focal point. Ad-interference rules add that Google-served ads must not overlap navigation or action items, severely interfere with content use, appear on no-exit screens, or be obscured by content. A credit-card hub that stacks offer tables above explanation can drift toward making the unit the page. Keep ads separate, keep explanation primary, and flag prompts that imitate editorial advice.
Pre-Publication Checks and Evidence Limits
Use these checks before publishing:
- Audience settings: no gender, age, parental status, marital status, or ZIP code for credit, banking, or covered planning content.
- Audience lists: no low-credit-rating, high-debt, or similar financial-status signals.
- Promises: remove approval, account-opening, no-credit-check, free/cash, loan, and below-market claims.
- Rewrite vague offers to page contents, links, or verifiable documents.
- RAC: preserve all key creative details; remove meaning-altering omissions.
- Queries: user-initiated, not pre-filled, not incentivized.
- AFS placement: search-results pages only; two boxes maximum; code unmodified.
- PIF: not focal, no misleading interaction prompts, not on violating content.
- Layout: no overlap with navigation/actions, no obstruction, no no-exit screens.
Evidence limits: no APR, fee, interest-rate, approval-rate, or credit-score data was verified, so none appears here; consult issuer disclosures. The targeting rule is US/Canada-specific. Google is consolidating publisher policies, so verify current text. This maps elements to categories, does not certify compliance, and is not legal or financial advice. No author, reviewer, or financial-expert credential is claimed. Sources: Google Publisher Policies and AFS strike tables.
研究の知恵