Facebook Ads for Insurance Agents: What $220K in Spend Taught Me About Lead Quality

Insurance leads from Facebook close slower, ghost more often, and cost less per lead than Google leads. Agents who build a follow-up system around that reality end up with a pipeline Google can't match on its own. I've run campaigns for auto, home, life, and Medicare supplement agents. The patterns below held up across all of them.

Most insurance agents try Facebook for 30 days, get 50 leads, close two policies, and quit. "These leads are garbage," they tell me. Then I look at their follow-up: one phone call, maybe a text, then nothing.

The lead was fine. The follow-up killed it.

Google gives you someone typing "cheap car insurance quotes near me." That person wants a policy today. Facebook gives you someone who saw a stat about overpaying and tapped out of curiosity. Different intent, different sales cycle, same revenue once they sign.

Insurance works on Facebook because people forget about coverage until you remind them

Insurance is a grudge purchase. You don't wake up excited about a new homeowners policy. But you've had that moment scrolling your phone at 11pm: am I paying too much? Does my policy cover what I think it covers?

Facebook puts those questions in front of people right when they're receptive.

I ran a campaign for an agent in Tampa that opened with "Florida homeowners: your insurance company raised rates 42% since 2023. Most agents won't tell you about switching mid-term." That ad pulled 847 leads in 90 days at $11.40 each. The agent closed 68 policies within six months. Average annual premium: $2,100. Lifetime value makes the $11 lead cost look like a rounding error.

That agent called leads within five minutes, ran an automated text sequence for no-answers, and followed up for 90 days. Not 3 days. Ninety. Most agents don't have that patience, which is exactly why the ones who do clean up.

CPL benchmarks by insurance line

These come from campaigns I've managed or audited directly. Your market will vary, but the ratios between lines tend to hold regardless of geography.

Medicare's CPL looks expensive until you see the close rates. People turning 65 are confused and want a human to walk them through it. During Annual Enrollment Period leads pour in, but CPLs jump because every Medicare agency bids at once. I watched $18 CPL in July become $55 in October for the same audiences, same creative, same everything. The only variable was competition.

A note on auto: it has the tightest margins. An agent spending $15/lead on auto and closing at 8% pays $187.50 per acquired policy on a $1,200/year premium. First-year commission barely covers acquisition cost. The payoff comes in year 2 when the renewal stacks with the cross-sell. If you're only running auto, the economics are fragile.

Creative that works (forget what your carrier gave you)

Carrier-provided creative offends no one and persuades no one. I learned this the hard way early on - we ran Allstate-provided templates for three weeks and CPL was $34 for auto. Switched to custom creative with a local angle, CPL dropped to $12 within a week.

Facebook rewards specificity and local relevance.

The "are you overpaying" angle. "Most [city] homeowners pay $300-500 more per year than they need to. I review policies for free - no obligation, no sales pitch. Just numbers." This angle keeps working because people believe they're overpaying. They're usually right.

The "did you know" format. "Did you know your homeowners policy probably doesn't cover flood damage? Even outside a flood zone, one storm backup can cost $40K. Check page 3 of your declarations page." You position yourself as the person who explains things, not the person who sells things. That distinction matters more than any targeting trick.

The life event trigger. "Just bought your first home? Congratulations. 3 coverage gaps your lender's minimum policy leaves wide open." Facebook lets you reach people based on recent home purchase behavior, new baby, marriage, retirement. These people need insurance right now and don't know it yet.

Video testimonials from real clients. Not the polished carrier ones. A 45-second selfie video where a client says "I was paying $380/month and [agent] got me the same coverage for $260." Raw, slightly shaky, real. Those videos outperform studio content 3-4x on CTR in my accounts. We tried both in a split test for a home insurance agent in Atlanta - the iPhone selfie drove 2.8% CTR versus 0.9% for the professional video. Same person, same message.

What to avoid:

• Stock photos of smiling families with "Protect What Matters Most" headlines. People scroll right past because they've seen this exact creative from 200 other agents.
• Carrier logos dominating the ad. Your prospects care about saving money, not which carrier you represent.
• Long lists of coverage types. Save that for the consultation call.
• "Get a free quote" with zero context. You're competing with 40 other agents in the same feed. Give them a reason to pick you.

Campaign structure for insurance

I run insurance campaigns on ABO (Ad Set Budget Optimization). Each ad set gets its own budget because a Medicare ad set at $30 CPL and an auto ad set at $12 CPL need independent control. CBO would pool the budget and Meta would spend disproportionately on whichever line gets more volume, which is usually auto - not where your best margins live.

Structure I use for multi-line agents:

For a solo agent in a mid-size market, $1,500-3,000/month is the range where you have enough data to optimize without burning through budget on a bad week. Below $1,500, you don't exit learning phase. Above $5,000, you're saturating your local audience unless you're in a major metro.

Objective: Leads. Instant Form or landing page.

Instant Forms give you 40-60% more leads per dollar. Quality drops though. People tap the form half-interested while scrolling on the couch and then don't pick up when you call.

Landing pages give you fewer leads but better ones. The friction of leaving Facebook and filling out a separate form filters casual clickers. Close rate from landing page leads runs 1.5-2x higher in my accounts.

Start with Instant Forms to build data and train the pixel. After you have 50-100 leads, add a landing page variant. Compare close rates at 90 days, not 7. Insurance isn't coffee - people don't buy on impulse.

Follow-up separates profitable agents from the ones who quit

I've watched the same exact campaign produce a 12% close rate for one agent and 2% for another. Same ads, same market, same CPL. The agent who closed at 12% had a system. The one at 2% had good intentions and no structure.

The system that works:

0-5 minutes: Phone call. If no answer, text: "Hey [name], this is [agent] - you requested an insurance review. When's a good time for 10 minutes? No pressure, just want to make sure you're not overpaying."

Same day: Second call attempt at a different time.

Day 2: Email with something useful. "3 questions to ask your current agent at your next renewal."

Day 3: Third call attempt. This is where most agents stop. Don't.

Day 7: Text with a specific hook. "Quick update - I ran some numbers for [city] homeowners your age, average savings runs about $480/year. Worth a 10-minute call?"

Day 14, 30, 60, 90: Monthly check-in. Not "are you ready to buy." Something the lead can use. "Hurricane season starts next month - here's a checklist to verify your coverage."

Agents running this kind of system close 8-15% of leads within 90 days. Agents who call once and move on close 1-3%.

Compliance: what gets insurance ads rejected on Meta

Meta doesn't classify insurance as Special Ad Category by default. Enforcement is inconsistent - I've had the same ad approved on one account and rejected on another in the same week.

Rejection triggers I've seen:

• Health insurance ads referencing specific conditions. "Save on health insurance" passes. "Coverage for diabetes management" gets flagged under health-related advertising restrictions.
• Medicare ads implying government affiliation. Don't use CMS logos, Medicare.gov screenshots, or language suggesting you represent the government. Even innocent phrasing like "Medicare has approved new benefits" can trigger a review.
• Life insurance with fear-based headlines. "Make sure your family's mortgage is covered no matter what" works. "What happens to your family when you die" gets flagged and converts poorly anyway because it makes people uncomfortable rather than motivated.
• Credit-related products like mortgage protection insurance fall under Special Ad Category. If you sell mortgage protection, use SAC from day one.
• Savings claims. "Save $500 on car insurance" needs backup. "Drivers in [city] save an average of $400-600/year" with a disclaimer is safer. The specific-range-plus-disclaimer approach gives you a concrete number without the compliance risk of an absolute promise.

State DOI regulations add another layer. California and New York have the strictest requirements. Include your license number, don't guarantee savings, don't name-drop competitors. Check your state's Department of Insurance advertising guidelines before launching - it takes 20 minutes and saves you from a cease-and-desist letter.

Medicare requires its own playbook

Medicare is the highest-value insurance lead you can get from Facebook. But CMS marketing guidelines are strict, and the penalties for violations go beyond ad rejections - they can include plan termination and fines.

Key constraints:

• Can't use the Medicare name/logo to imply government endorsement
• Can't provide misleading plan comparisons
• Need specific disclaimers about plan availability by area
• During AEP (October 15 - December 7), CPLs jump 150-300%

The agents who win at Medicare on Facebook start advertising in July and August when CPLs sit at their lowest. They build a list through summer, then work that warm list during AEP instead of fighting in the auction with fresh cold campaigns. $18-25 CPL in summer versus $45-65 in October. Same person, same product, triple the acquisition cost because you waited.

One agency I work with spends $8K from July through September and collects about 400 leads at $20 average CPL. During AEP, they focus 90% of effort on those warm leads instead of cold prospecting. Their AEP close rate from the warm list: 22%. That's 88 policies from $8K in pre-season spend. Try getting that result starting cold in October.

Audience targeting for insurance in 2026

Strong creative with broad targeting beats weak creative with narrow targeting. I see this month after month. The agents who obsess over interest stacking but run generic carrier creative waste their time fine-tuning a targeting setup that a better ad would have rendered unnecessary.

That said, some targeting still moves the needle for insurance:

Homeowners. Meta's "Homeowners" behavior segment works because renters don't need homeowners insurance and homeowners carry higher auto coverage. Simple filter, meaningful improvement.

Life events. Recent movers, newly engaged, new parents. A couple with a new baby needs life insurance and probably hasn't thought about it yet. A recent home buyer needs homeowners insurance and wants auto bundled. These people have an urgent need they don't recognize as urgent.

Age ranges. Medicare supplement: 60-65+. Final expense: 50-75. Term life: 28-45. Wide enough to avoid audience fragmentation but narrow enough to keep your creative relevant.

Lookalikes from your client list. Upload hashed emails from your book of business and build a lookalike. Your policyholders share demographic and behavioral patterns that Meta's algorithm can match. I've seen client-list lookalikes outperform interest targeting by 30-50% on close rate. If you have 500+ clients in your book, this is probably your strongest audience.

Exclusions. Exclude current clients, people who already submitted a lead form, and employees of insurance companies. Open your Ads Manager, go to your ad set, and check that these exclusions are active. I audit accounts where agents are paying $15/lead to reach their own policyholders.

Tracking: fix this or you're optimizing blind

Insurance has a 30-90 day sales cycle. Meta's default attribution window is 7-day click, 1-day view. That means 60-80% of your closed policies won't show as conversions in Ads Manager. You're making optimization decisions based on maybe 30% of your real results.

You need offline conversion tracking. Export closed policies from your CRM, match them back to Facebook leads, upload as offline events. This feeds the algorithm data to optimize toward people who buy policies, not people who fill out forms.

Without offline conversions, Facebook optimizes for form-fillers. Some of your cheapest leads will be people who fill out forms for coupons, free stuff, whatever, with zero intention of buying a policy. Feeding back policy data shifts optimization toward real buyers, even when those leads cost more up front.

One agency I audited saw CPL increase from $14 to $22 after implementing offline conversions. Sounds like a regression. But their close rate went from 6% to 14%. Cost per acquired policy dropped from $233 to $157. The algorithm was spending more per lead to reach people more likely to buy, and the unit economics improved despite the higher CPL.

Set this up through CAPI (Conversions API). Server-side tracking is more reliable than browser pixel, especially when the "conversion" is a phone call three weeks after the click. If you're not sure whether CAPI is running, check Events Manager > Data Sources > your pixel > Settings tab. Look for "Server Events" in the connection overview.

Five mistakes I keep seeing

Selling the policy in the ad. Your ad generates a lead. You close the deal on the phone. I had one agent who put his entire rate table in the ad copy - 600 words of premium tiers. Two leads in three weeks. We cut it to a four-sentence hook about overpaying, leads tripled.

Quitting after 14 days. An agent shut down campaigns after two weeks because leads weren't closing. He had 38 leads in his CRM. Six months later I checked out of curiosity - 9 of those 38 had bought policies from competitors. They were real prospects. He abandoned them before they were ready to decide.

Running the same creative for months. Insurance creative fatigues in 3-4 weeks in a local market. You're hitting the same 50,000 people repeatedly. Swap headlines, swap images, swap the hook. Keep the offer, change the packaging. Set a calendar reminder every three weeks.

No CRM integration. Leads arrive via email notification, agent copies them to a spreadsheet, follows up when they remember. That breaks at 30+ leads a week. You need a CRM that auto-imports from Facebook, triggers follow-up sequences, and tracks pipeline position. HubSpot free tier handles this. So does GoHighLevel or AgencyZoom.

Ignoring cross-sell. An auto lead is a home lead. A home lead is an umbrella lead. A new parent lead is a life lead. Map your cross-sell paths and build them into the follow-up sequence. One lead should create 1.5-2 policy opportunities. I've seen agents double their revenue per lead by adding a bundle question to their second follow-up call.

FAQ

How much should an insurance agent spend on Facebook Ads per month?

$1,500-3,000 for a solo agent in a mid-size market. That generates 75-200 leads depending on insurance line and location. Below $1,000 you won't exit learning phase consistently. Above $5,000 you need either a follow-up team or a larger geographic area to avoid saturating the same audience.

How long before I see ROI from Facebook Ads for insurance?

90 days minimum before you have enough data to judge. The first 30 days build data and train the algorithm. Days 30-60 produce leads that enter your pipeline. Most policies close 30-60 days after first contact. Your first Facebook leads might not become revenue until day 60-90. Judging at day 14 means looking at an incomplete picture and making permanent decisions from temporary data.

Are Facebook leads worse than Google leads for insurance?

Different. Google leads have higher intent - those people searched for insurance. Close rates from Google run 15-25% versus 6-15% from Facebook. But Google CPLs are $35-90 for insurance keywords versus $8-35 on Facebook. Cost per acquired policy often lands in the same range once you factor in volume. Running both gives you the demand generation from Facebook and the high-intent capture from Google.

Do I need a landing page or can I use Instant Forms?

Start with Instant Forms for volume and pixel training. After 50-100 leads, test a landing page alongside. Compare 90-day close rates, not CPL. Landing pages produce 1.5-2x higher close rates but fewer leads per dollar. Most agents do best with both running - Instant Forms for volume, landing page for quality, and the data from both feeding the algorithm.

Can insurance agents run Facebook Ads legally?

Yes. Standard insurance lines (auto, home, life) don't require Special Ad Category. Mortgage protection and credit-related insurance do. Include your license number in ad text, avoid guaranteed savings claims, and follow your state DOI advertising guidelines. Medicare ads need CMS-compliant disclaimers. When in doubt, check your state's Department of Insurance website - the rules are public and specific.

Bottom line

Facebook Ads for insurance work because the volume is high, the cost per lead is low, and the lifetime value of an insurance client absorbs moderate close rates. An agent who closes 10% of leads at $15 CPL and retains those clients for five years makes money. An agent who closes 2% and quits after a month doesn't.

The difference between those two agents is rarely the ads. I've run the same campaign for both types. Same creative, same targeting, same market. The difference is follow-up discipline, CRM hygiene, and the patience to let a 90-day pipeline mature instead of demanding results in 14 days.

In a local market, every lead you capture is a lead 40 other agents didn't get. The agent showing up in someone's feed with useful content before renewal date has a real shot at that policy. The agent who only shows up when someone Googles "insurance agent near me" is competing at the most expensive, most crowded stage of the funnel.

Build the pipeline. Follow up. Give it 90 days before you decide it's working or not.