Where Personal Loans Fit in the Policy
Google's personalized advertising rules name "credit-related products or services, banking products and services, or certain financial planning and management services" as a category with extra limits. The policy text gives examples: credit cards and loans, bank and checking accounts, debt management products. A personal loan campaign therefore sits inside a named category rather than on its margin, and two separate obligations follow from that: a category-specific targeting restriction that applies in the United States and Canada, and a broader sensitive-data restriction that applies wherever personalized ads run.
Readers often assume the two are the same rule read twice. They are not. The first restricts who you may target. The second restricts what data you may use to decide who to target. A campaign can pass one test and fail the other. Treating the two tests as one rule hides the second failure mode; each needs its own fix.
The Five Signals You Cannot Use in the US and Canada
Within the United States and Canada, the policy prohibits advertising credit-related products, banking products and services, and certain financial planning and management services on the basis of an audience's gender, age, parental status, marital status, or ZIP code. The wording is a list of signals, not a list of ad formats, so the practical question is where those signals enter your targeting: demographic segments, inferred age bands, household or life-stage audiences, and geo-segments cut to postal code rather than to a broader area.
A realistic case: a lender wants to reach "recently divorced homeowners in three ZIP codes" for a debt-consolidation loan. Each of those three descriptors — marital status, ZIP code, and, depending on how it is defined, homeownership-linked life stage — has to be checked against the list. A campaign targeting "adults 25-54 in the Chicago metro" fails on age; the same creative aimed at a broad metro area without demographic narrowing does not fail on that ground.
The exception written into the policy concerns employment advertising for certain pre-determined US government advertisers using bona fide occupational qualifications. It does not extend to credit products, and it should not be read as a general licence to segment.
A Second, Broader Ban: Negative Financial Status Data
Separately, personalized ads may not be selected or targeted using, and audience data such as Cookie lists may not be collected or used from, a set of sensitive sources. The list includes activity on adult, gambling, or government sites and apps; activity of users known to be under 13; and inferred or actual sensitive information. Among the sensitive categories the policy names are "negative financial status information or other detailed information related to an individual's financial situation, such as information indicating a low credit rating or a high debt burden."
That sentence is the crux for personal loan marketing. A campaign built from a list of people showing signs of credit stress, or from browsing behaviour on pages about debt problems, is using precisely the data the policy excludes — even if no demographic signal is involved. It also matters for audience building: a Cookie list assembled from a debt-relief site inherits the problem. Audience size does not neutralize the issue.
Consent, Cookie Lists, and the Interest-Based Notice
Three obligations attach to personalized advertising when Google platform products or Google ad code are used. The advertiser must hold every right needed to use audience data such as Cookie lists. The advertiser must add a notice in the ad stating that it is tailored to user interests — the policy cites an "Ad Choices" icon as an example. And all parties must follow applicable industry codes, such as the Digital Advertising Alliance's Self-Regulatory Principles for Online Behavioral Advertising or IAB Europe's EU Framework for Online Behavioral Advertising.
These are process expectations, not creative choices, and they are the part most often left until launch. A useful check: if you cannot name the source and the permission behind every audience list you upload, the rights requirement is not yet satisfied.
Statements That Escalate a Violation
The policy's enforcement examples separate ordinary violations from severe ones. Ordinary examples involve promises that are made but not clear or explicit. Severe examples include concrete promises that cannot be fulfilled — among them "free or cash offers," "unreasonably cheap offers," and, named directly, "promising loans." A related severe example is a specific promise outside the advertiser's control, such as opening a high-yield savings account with no credit check.
For loan advertising, the practical line is approval language. "Guaranteed approval," "everyone approved," or an offer presented as certain when approval depends on the lender's review is the kind of concrete, unqualified promise the policy treats as egregious rather than ordinary. Separately, publisher-side rules on acquiring traffic forbid traffic sources that promise products, services, or promotions the destination page does not contain or makes hard to find — relevant when a lead-generation page is promoted through third-party placements.
A Working Checklist
A practical sequence, in the order the checks usually matter:
- Map every targeting input to one of the five signals; document the mapping.
- Strip demographic and postal-code narrowing from loan campaigns in the US and Canada.
- Audit audience lists for financial-hardship indicators before upload.
- Confirm rights, the interest-based notice, and the industry code obligation.
- Rewrite approval claims into conditional language.
- Keep a change log, because the policy is revised over time.
Boundaries Worth Stating
This summary rests on Google's published policy text, which is a platform rule and not a statement of US lending law. Federal and state credit rules, licensing, and disclosure requirements are outside that document, and the two frameworks are not substitutes. Wording, category definitions, and enforcement examples can change, and the rules differ by country, so confirm the current text before launch and take advice from qualified counsel for legal questions.
研究の知恵