How to Vet a Google Local Service Ads Agency for Admissions

The FTC’s 2022 Health Products Compliance Guidance requires that any health-related advertising claim be supported by competent and reliable scientific evidence, which for efficacy claims typically means randomized, controlled human clinical testing ¹. That standard applies to addiction treatment the same way it applies to a supplement label or a digital therapeutic. An LSA ad headline promising lasting recovery, a landing page citing a success rate, or a call script describing detox comfort all sit inside that scope.

The 2022 guidance is not a minor refresh. It consolidates lessons from more than 200 FTC health-product advertising enforcement actions and explicitly extends the agency’s framework from dietary supplements to all health-related products and services, including treatments and apps ². An LSA agency working in addiction services should be able to explain, without prompting, how that scope change affects copy review.

Lead generators are inside the enforcement perimeter, not adjacent to it. In December 2024, FTC staff sent warning letters to 21 health-care plan marketers and lead generators flagging misrepresentations of benefits, comprehensiveness, costs, and incentives ³. An LSA agency occupies that same lead-generator category when it writes ad copy, owns the landing page, or scripts the intake call.

The diagnostic question is procedural. A defensible agency maintains a claim substantiation file: each material claim in active ad copy and landing pages mapped to a specific source document, with the level of evidence noted and the reviewer identified. If the agency cannot produce that file on request, its copy has not been pressure-tested against the FTC standard, and the treatment center inherits the exposure when the ad runs under its Google Screened profile.

Two practical consequences follow. First, outcome language (‘proven results,’ ‘highest success rate,’ ‘medically supervised comfort’) needs documented backing or it should be cut before launch. Second, implied claims matter as much as explicit ones; FTC reviews the net impression a reasonable consumer takes from the ad, not the literal sentence. Agencies that only review explicit wording are reading half the rulebook.

HIPAA’s marketing rule is narrower than most agencies assume and broader than most centers expect. HHS defines marketing as communication about a product or service that encourages the recipient to use it, and with limited exceptions, any use or disclosure of protected health information for that purpose requires prior written authorization from the individual ⁴. The exceptions for face-to-face communications or nominal promotional items do not cover the workflows that sit underneath a Google Local Services Ads program ⁵.

Four LSA touchpoints regularly cross into PHI territory without the center noticing.

• Call recordings are the first. Google’s LSA platform records calls by default, and the moment a caller states a diagnosis, medication, substance use history, or insurance information tied to that condition, the recording contains PHI. If the recording is stored on the agency’s systems, reviewed by agency staff for quality scoring, or used to train sales scripts, the agency is handling PHI on behalf of the center. That is a business associate function, and it requires a Business Associate Agreement before the first call lands.
• CRM imports are the second. When intake notes, call transcripts, or lead-disposition tags flow from the agency’s tracking platform into the center’s CRM or admissions system, the data set traveling back and forth almost always contains PHI. The vetting question is whether the agency’s CRM, call platform, and any sub-processors are covered by the BAA chain.
• Retargeting pixels are the third. Pixels fired from a landing page that already collected a phone number, name, or condition-specific intake field can transmit identifiers to ad platforms that are not business associates. HHS guidance treats disclosure of PHI to another entity for that entity’s own marketing as a use requiring authorization ⁴.
• Lookalike audiences built from converted-caller lists are the fourth. Uploading a list of admitted patients to seed a lookalike model is a disclosure of PHI for marketing, and the authorization standard applies.

An agency that cannot map each of these touchpoints to a written control fails the HHS pressure test.

HHS OIG flags marketing and referral payment arrangements as a recurring category of suspect remuneration, with Special Fraud Alerts and related guidance identifying structures that present heightened enforcement risk under federal fraud and abuse laws ⁶. The label on the invoice does not control the analysis. A fee called ‘marketing’ that functions like a per-patient payment for steering admissions reads to OIG the way a referral payment reads.

Peer-reviewed research on for-profit substance use disorder treatment documents how patient brokering and financial inducements have caused direct patient harm, with study participants linking brokering practices to fatal and non-fatal opioid overdoses ⁷. That evidence base is why the compensation structure inside an LSA agency contract is a clinical risk question, not only a finance question.

Three common LSA fee structures sit at different points on the risk curve.

• Pay-per-lead, the native LSA model, charges for a qualifying inbound call regardless of admission outcome. The fee is tied to the lead event, not the patient decision, which keeps it closer to a media cost than a referral payment, though ad copy quality and dispute integrity still drive FTC exposure.
• Flat retainer compensates the agency for media management, creative, and reporting on a fixed monthly basis. It is the cleanest structure from a remuneration standpoint because no payment varies with admission volume.
• Performance fee per admission ties agency compensation directly to admitted patients. That structure most closely resembles the arrangements OIG has historically scrutinized, and it sits adjacent to the brokering patterns documented in the SUD treatment literature ⁷.

A defensible agency can describe its fee structure in one sentence, identify which federal risk surface that structure creates, and explain what its contract does to mitigate it. Agencies that obscure the compensation model behind ‘success-based’ or ‘admissions-aligned’ language without specifying the trigger fail this test on the first question.

The claim substantiation file is the single artifact that separates a defensible LSA program from a hopeful one. It is a working document, not a brochure. Every material claim in active ad copy, landing pages, intake scripts, and Google Screened profile content should appear in the file with three columns: the claim as written, the source document that supports it, and the reviewer who signed off.

Source documents need to actually substantiate the claim. A press article quoting the center’s own marketing language is not evidence. Peer-reviewed outcome data, internal chart-audited completion rates with documented methodology, or licensure and accreditation records are. The FTC’s framework calls for competent and reliable scientific evidence, and for efficacy claims that standard typically points to controlled human clinical testing rather than testimonial collections ¹.

The landing page evidence log extends the file to every page a click can land on. Hero headlines, statistic callouts, before-and-after framing, and clinician credentials each need a row. Implied claims, including imagery that suggests a clinical setting the facility does not operate, belong in the log too. An agency that produces a clean file in under a week has the workflow in place. An agency that asks what the file is reveals that copy has been shipping without one.

Three documents anchor the HIPAA portion of the request. The Business Associate Agreement comes first. It needs to name the agency, every sub-processor that touches lead data (call tracking platform, CRM, analytics tool, ad platform integration, transcription service), and the safeguards each party maintains. A BAA that lists only the agency while sub-processors run unnamed in the background is incomplete. HHS guidance treats disclosure of PHI to another entity for that entity’s marketing as a use requiring authorization, which means the chain matters as much as the top-level contract ⁴.

The call recording consent script comes second. LSA call recordings begin the moment a caller connects, and addiction-treatment callers routinely volunteer diagnosis, substance, and insurance information within the first thirty seconds. The agency should provide the verbatim opening that intake staff use, the disclosure language it places on landing pages near the click-to-call element, and the retention schedule for recordings. Without authorization or a BAA covering the recording pathway, the storage and review of those calls sits outside HIPAA’s narrow marketing exceptions ⁵.

The pixel inventory comes third. The agency should produce a written list of every tracking pixel, conversion tag, and audience-sync integration firing on landing pages and thank-you pages, with notes on what data each one transmits. Pixels that send hashed phone numbers, email addresses, or condition-specific URL parameters to ad platforms turn a marketing page into a PHI disclosure point. Agencies that cannot produce the inventory in a spreadsheet have not audited their own implementation.

The lead dispute log shows whether the agency is managing the LSA channel or just paying its invoices. Google credits charges for leads that fall outside the service area, target an unrelated job type, or come from spam. A working log captures every disputed lead, the reason, Google’s resolution, and the recovered credit. Centers reviewing the log learn two things: how aggressively the agency challenges bad leads, and what the actual cost per qualified call looks like once credits are netted out.

Google Screened verification proof is straightforward but easily skipped. The agency should produce the license documentation, background check confirmations, and insurance certificates submitted during onboarding, along with the date the profile achieved Screened status. Profiles operating without current verification can lose ad eligibility without warning, which collapses the pipeline overnight.

The compensation disclosure closes the request. The agency should describe its fee structure in writing, identify the trigger event for each fee (impression, call, qualified call, admission), and acknowledge in the contract which structure applies. OIG guidance flags marketing arrangements that function as referral payments as suspect regardless of how they are labeled ⁶. A clear written disclosure is what allows counsel to evaluate the arrangement against that standard before the first lead is paid for.