Google Ads for Loan Officers: Compliance, Targeting, and Attribution That Actually Works
Most Google Ads guides for loan officers were written by generalists who do not work in mortgage. They tell you to target your ZIP code. That instruction gets your account suspended. Mortgage sits inside a restricted advertising category with rules that block the tactics every other local business uses.
This guide covers the three factors that decide whether your mortgage campaigns work: where Google lets you show ads, what copy passes review, and how to track loans that fund sixty days after the click. Setup is table stakes. Compliance and attribution are what separates accounts that scale from accounts that burn budget until you shut them off.
Why normal geo-targeting does not work for mortgage ads
ZIP code targeting gets your mortgage ads disapproved. Google's Financial Services policy requires advertisers in restricted financial categories to target by state, not radius or postal code. If you hold an NMLS license in Texas, your ads show everywhere in Texas. You cannot narrow that to a 20-mile radius around your office.
The policy exists because mortgage lending is state-regulated. You must be licensed in every state where your ads appear. Google catches violations during review or after a competitor reports you. The account gets suspended, and you lose weeks reapplying for certification.
State-level targeting creates waste. If you operate in Austin but target all of Texas, you pay for clicks from Amarillo borrowers you cannot serve. The official workaround is negative location keywords: add `-Amarillo`, `-Lubbock`, and every city outside your service area as negative keywords. That requires maintaining lists of hundreds of exclusions.
The better workaround is landing page routing. Send all Texas traffic to a state-level landing page that asks for ZIP code, then routes visitors to city-specific pages or a "We do not serve your area" message. This keeps you compliant while filtering geography after the click.
Most loan officers run campaigns in states where they hold no license because their agency never verified NMLS coverage. You are responsible for compliance, not your vendor. Pull your NMLS record, confirm every active state license, and cross-check against your campaign location settings. If the lists do not match, you are one audit away from a suspension.
The ad copy disclosure requirements no one follows
Every mortgage ad must include your NMLS number. Google requires it in the description lines or the final URL path. Most advertisers bury it on the landing page. That does not satisfy the rule. The disclosure must appear in the ad unit itself, visible before the click.
Generic CTAs like "Get Pre-Approved in 10 Minutes" trigger disapprovals if your process cannot deliver that promise in all targeted states. Google treats mortgage ads as binding offers. Most loan officers use the same ad copy across all states and hope Google does not notice.
Rate and payment claims require APR disclosure and cannot appear in headlines. "3.5% rates available" in a headline gets disapproved. If you mention a rate, the full APR and representative example must appear in the description, which leaves no room for benefit-driven copy. The easiest solution is to avoid rate claims entirely and focus on speed, service, or loan program expertise.
VA and FHA loan ads must include eligibility disclaimers. "VA loans for veterans" passes review. "Get a VA loan" does not, unless you add "for eligible veterans" in the description. Google interprets the second version as a guarantee available to anyone, which violates truth-in-lending principles.
Responsive search ads auto-generate combinations that violate these rules. You write compliant headlines and descriptions separately, and Google mixes them into combinations you never reviewed. Pin your NMLS disclosure and eligibility language to specific positions so they appear in every combination.
Free PDF · the crash course
AI Agents for Marketing Crash Course
Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.
Search campaigns beat Performance Max for lead quality
Performance Max sends mortgage traffic to YouTube pre-roll and Gmail promotions tab. Those placements convert at lower rates than search because the audience is not actively searching for a loan. Someone watching a home renovation video might click your ad out of curiosity, but they are not filling out an application that day. You pay more for a cold click than you would on social platforms.
Search campaigns let you separate purchase intent from refinance intent. A borrower searching "FHA loan Texas" is buying a home. A borrower searching "lower mortgage payment" is refinancing. These leads have different qualification rates, different loan amounts, and different urgency. Blending them in one Performance Max campaign means your bidding algorithm optimizes for whichever converts faster, usually refinance. You end up with a pipeline full of small refis when you wanted purchase loans.
Phrase match on loan-type queries outperforms broad match on cost per funded loan. Broad match on "mortgage lender" shows your ad for "mortgage news," "mortgage crisis," and "mortgage-backed securities." You burn budget on informational searches with zero buyer intent. Phrase match keeps you aligned to transactional queries.
Single Keyword Ad Groups for high-intent terms deliver lower cost per lead versus blended campaigns. A SKAG for "mortgage lender Austin" gets one keyword, three ads specific to Austin mortgage lending, and a landing page that says "Austin Mortgage Lender" in the headline. The tight alignment between keyword, ad, and landing page lifts Quality Score, lowers CPC, and increases conversion rate.
Dynamic Search Ads pull in unqualified traffic that wastes budget. DSA scans your landing pages and auto-generates ads for any search query Google thinks is relevant. If you have a blog post titled "Understanding Mortgage Rates," DSA shows your ad for "mortgage rate news" and "mortgage rate predictions." These searches have informational intent, not transactional. Disable DSA unless your content team writes exclusively for high-intent landing pages.
Landing pages must route by loan type and state
A generic "Contact Us" page converts at lower rates than a loan-type-specific page with a rate table. Borrowers who click an ad for "FHA loan" expect to land on a page about FHA loans, not a homepage with a generic hero section. The first headline they see should match the keyword they searched.
State routing ensures compliance disclosures match the visitor's location and your licensure. Send Texas traffic to a Texas page with your Texas NMLS license number. Send California traffic to a California page with California-specific rate disclosures and your California license. If you route everyone to one national page, you either show no license number or show the wrong one.
Forms that ask for loan amount and property address before contact info qualify leads better. A name/email form gives you a list of people who might want a mortgage someday. A form that asks "What is your estimated purchase price?" and "What is the property ZIP code?" gives you leads with defined intent and a timeline. Longer forms reduce conversion rate. But the leads that do convert close at higher rates, which means lower cost per funded loan.
Mobile pages with click-to-call as primary CTA convert higher than desktop-style lead forms on mobile. Most mortgage traffic is mobile. The borrower does not want to fill out an eight-field form on a phone keyboard. Put a large "Call Now" button above the fold, route it to a tracking number, and let them talk to a human in 30 seconds.
Including your NMLS number and loan officer photo on the landing page lifts conversion. Mortgage is a high-trust transaction. The borrower is sharing income docs and bank statements with someone they found on Google five minutes ago. A generic landing page with stock photos and no human face signals that you might be a lead aggregator. Your photo and license number signal that you are a real person with a real credential.
Conversion tracking breaks at application, not funding
Google's conversion window maxes out at 90 days, but mortgages close in 30 to 60 days. The attribution lag creates a reporting gap. A lead that clicks your ad on January 1 and funds on February 15 shows up in your January ad report as a conversion, but you do not know the outcome until February. If you pause the January campaign in early February because it "did not work," you kill the campaign that actually generated funded loans.
Tracking only applications makes every lead look profitable until you see the close rate weeks later. Cost per application is a vanity metric. If your CPA is low and your close rate is low, your real cost per funded loan is high. If a competitor has a higher CPA but a higher close rate, their cost per funded loan is lower. They win. Google's algorithm does not know this. It optimizes for application volume unless you tell it otherwise.
Offline conversion imports from your LOS let you feed funded loan data back into Google Ads bidding. When a loan funds, your loan origination system records the borrower's email and phone. You export that data, format it as a CSV, and upload it to Google Ads as an offline conversion. Google matches the email and phone to the original ad click, then adjusts bidding to prioritize keywords and audiences that generate funded loans, not just applications.
Enhanced conversions using email and phone matching recover conversions lost to iOS and consent modes. When a borrower submits a form, send their hashed email and phone to Google alongside the conversion event. Google matches that hashed data to a signed-in Google account, which restores attribution even when the click was not tracked due to ATT or cookie restrictions.
Without LOS integration, you are optimizing for lead volume, not funded deal volume. Google sees applications and assumes your campaign is working. You see applications, funded loans, and a cost per funded loan that may exceed your commission after splits and overhead. The campaign may be burning money, but Google's algorithm thinks it is winning because it does not see the back-end data.
Free PDF · the crash course
AI Agents for Marketing Crash Course
Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.
Budget reality: you need thousands per month minimum
Mortgage keywords cost substantial amounts per click in competitive metros. Competitive markets consistently see high CPCs for mortgage and refinance keywords. At high CPCs and typical conversion rates, you need many clicks to get meaningful lead volume. That requires substantial spend for leads. If your close rate is typical, each funded loan costs a significant amount in ad spend.
Starting below adequate monthly budgets means weeks of learning phase with no statistically significant data. Google's Smart Bidding requires volume to learn. If you spend too little per month at high CPCs, you get few clicks and few leads. You need enough conversions in 30 days for the algorithm to exit learning phase. At low monthly spend, it takes many months to collect enough data.
Google's Smart Bidding needs conversions in 30 days to optimize. If your conversion rate is typical and your CPC is high, you need many clicks to get those conversions, which requires substantial spend. Most loan officers cannot afford that. The workaround is to track micro-conversions: form submissions, phone calls, and chat initiations. These fire more frequently than applications, which gives the algorithm more data to work with. The tradeoff is that you optimize for engagement, not application quality.
Loan officers spending small amounts per month compete against lenders spending large amounts who dominate auction priority. Google Ads is an auction. The advertiser with the highest Ad Rank wins the top position. If you bid lower and a competitor bids higher, they win even if your Quality Score is slightly higher. At low monthly spend, you appear in lower positions.
The ROI math only works if you track to funded loans and know your commission per loan. If your average loan is substantial, your commission is typical, and your split is high, you earn a certain amount per funded loan. If your cost per funded loan is lower than that, you are profitable. You can calculate profit per loan and scale accordingly. But if you stop tracking at application and assume every lead is worth the same, you have no idea whether your spend is profitable or disastrous.
What Graphed tracks that your CRM cannot
Graphed tracks cost per funded loan across all channels, not just cost per lead from Google Ads. Your CRM shows that Google Ads generated leads at a certain CPA. Graphed shows how many of those leads funded, which means your actual cost per funded loan and your ROI. It also shows how other lead sources performed on a cost per funded loan basis, revealing which channels are truly profitable.
Graphed tracks time-to-close by lead source so you know which leads fund faster or slower. Leads that fund faster let you recycle revenue faster than leads that take longer. Faster close cycles mean higher cash velocity. You can see which channels deliver faster closes and prioritize accordingly.
Graphed tracks campaign-level ROI using actual loan revenue, not estimated commission. It pulls funded loan amounts from your LOS, calculates your commission based on your split and basis points, subtracts your ad spend, and shows profit per campaign. You see which campaigns are profitable and which are losing money. You can reallocate budget accordingly.
Graphed tracks cross-channel attribution when a lead clicks one channel, then converts via another. Standard Google Ads reporting credits the last click. If someone clicks your ad, subscribes to your email list, then applies via a link in your drip campaign, Google sees zero conversions and your email platform takes full credit. Graphed uses multi-touch attribution to assign partial credit to both channels. You see that one channel initiated the relationship even though another closed it.
Graphed automates LOS integration and pulls funded loan data without manual CSV uploads or third-party tools. You connect your LOS once, and Graphed syncs funded loan data daily. Every funded loan automatically flows into your reporting dashboard, matched to the original lead source and campaign. You log in, see which campaigns are profitable, and reallocate budget in minutes.
---
See Graphed in action.
More Google Ads guides
Google Ads for Patio & Pergola Contractors: Campaign Structure That Delivers $30K+ Projects
Most patio and pergola contractors waste Google Ads budget on low-intent searches and generic landing pages. Here's the campaign architecture that fills your pipeline with $30,000+ projects.
Google Ads for Pavers: Campaign Structure, Costs, and Lead Quality
Google Ads puts paving companies in front of homeowners searching for driveways and contractors hunting for parking lot work. Structure and targeting determine whether you book jobs or burn budget.
Google Ads for Paving Companies: Campaign Structure That Books Jobs
Most paving companies burn half their Google Ads budget on tire-kickers, wrong service areas, and DIY searchers. The fix starts with campaign structure.
Google Ads for Pediatric Dentists: What Actually Books Appointments
Parents searching for pediatric dentists convert differently than adult patients. Campaign structure, keyword intent, and landing page clarity determine whether you book first visits or burn budget.
Google Ads for Pediatricians: Patient Acquisition Without the Agency Markup
Parents search for pediatricians when they need one, not when your practice is top of mind. Paid search captures that intent in real time.
Google Ads for Periodontists: Which Campaign Structure Actually Books Patients
Periodontists run two businesses in one practice: implant shoppers and periodontal disease patients. Your Google Ads structure must separate them, or you waste budget on leads who never close.