How to run private health Facebook ads as an agent

I’ve been running Facebook ads for insurance agents for years. Right now my team manages about $120,000 a month in ad spend across hundreds of accounts. Some are top producers in their companies. Some are IMOs. Some are newer agents building something of their own. The principles are the same either way.

We work across every vertical (final expense, ACA, term life, mortgage protection, annuities, the whole list), but private health is where I get the most questions. So this one’s specifically for that. Five things that make or break a private health ad campaign.

First, some context on why an agent would run their own ads instead of just buying leads. If you’ve ever bought leads, you know the problem. You call the lead, the prospect already talked to three other agents that day. Or the lead is two weeks old. Or the phone number’s wrong. Or worse, you’re paying premium for a lead that’s been resold half a dozen times before it hit your screen.

When you run your own ads, that whole problem disappears. The lead is generated in real time. It’s exclusive to you. Your name and your information are on the ad, so when the prospect picks up the phone there’s already recognition. You’re the first call, not the fifth. That’s why most of my top agents have stopped buying leads almost entirely.

1. Qualify on the front end

The most common mistake I see is an agent running an ad that casts too wide a net, then getting frustrated when half the leads don’t qualify. That’s not the lead’s fault. That’s the ad’s fault.

For private health, the leads worth chasing are people earning real income. I anchor around six figures because that’s where private plans start to make sense over a marketplace plan. If income isn’t filtered on the front end (in the ad copy, the form questions, or the landing page), you’ll burn hours on calls with people who can’t afford the product.

One thing to know up front: insurance ads fall under Meta’s financial services category, which means less targeting flexibility. You can’t target by income directly anymore. So that filtering has to happen in your copy and your form. Plan for it.

2. Test more than you think you need to

The ad you think will crush usually doesn’t. The only way to find your winner is to put real money behind a few different angles and let the data sort it out.

I launch every new account with at least three to five creatives. Different hooks. Different angles. Different formats. Some agents do well with video. Some win with a static image and a strong headline. Some land best with a testimonial-style ad. You don’t know which until it’s running.

Offer language matters too. “Private health insurance” hits differently than “an alternative to the marketplace.” Same product, different angle, often very different cost per lead. We keep a rotating library of tested creatives we pull from for every agent, and we’re constantly swapping new ones in. What works in March won’t necessarily work in July.

Quick note on lead capture: you can use a Meta lead form (inside Facebook and Instagram) or a landing page. Both work. Lead forms are faster for the prospect. Landing pages give you more control over the experience and the data. We use both depending on the agent and the offer. Don’t pick a side without testing.

3. Budget like you mean it

I see agents try to run ads on $100 a week and then complain nothing’s happening. That’s not how Meta works. The algorithm needs data to optimize. The more you give it, the smarter it gets. Too little data and it’s just guessing.

The floor I recommend is $500 a week, about $70 a day. If you can do more, do more. Some of my agents spend $3,000 a month. Some spend much more. But $500 is the minimum to give the algorithm something real to work with.

For private health specifically, cost per lead usually lands in the $15 to $40 range, depending on whether you’re using a lead form or landing page, how aggressive your filtering is, and how dialed in the campaign is. That number tends to run higher in the first few weeks because the algorithm is still learning your account.

Heads up if your account is brand new: Meta caps your spend for the first three weeks or so, usually around $350 a week. They want roughly $1,000 in total ad spend before they’ll trust you with more. Plan for that.

4. Don’t choke the algorithm with narrow geography

This one fights what feels intuitive. You’re licensed in three states, so you only want leads from those three states. Makes sense on paper.

The problem: Meta needs reach to find the right audience. Targeting one or two states chokes the algorithm. There aren’t enough qualified people in that small a geography to keep cost per lead competitive.

I recommend at least five states. More is better. If you’re not licensed in five yet, get licensed. The cost of new licensing is way less than the cost of running expensive ads in too tight an area. You’ll pay one way or the other, so might as well pay the licensing side.

5. Be patient (or don’t start)

This is the most important one. Most agents who quit on ads quit too early.

Here’s the pattern I see. Agent launches ads. Day one, leads come in but they’re a little expensive. Day three, things slow down. Day five, the agent panics, kills the campaign, walks away saying “ads don’t work.”

Meta’s algorithm needs about two to three weeks to exit the learning phase. Then another month or two to really find its stride. Plan to commit at least 90 days. Not because of any contract. Because that’s how long it takes for a campaign to truly optimize.

The first two weeks are noisy. Leads might cost more than you want. You might get a bad batch. That’s normal. The agents who win are the ones who don’t flinch in week one. By month two or three, cost per lead drops and close rates climb. The agents we’ve had running the same campaigns longest are the ones doing best.

If 90 days isn’t realistic for you, don’t start. That’s honest advice. You’ll burn money and convince yourself ads don’t work when really you just didn’t give them runway.

So do it yourself or have someone run it?

You can absolutely take these five and go run your own ads. Plenty of agents do, and with AI tools the creative side is easier than it used to be.

But running your own ads is a real skill. It takes time to learn Meta. Time to write copy. Time to build creative. Time to manage daily optimization. Most of the agents I talk to want to be selling insurance, not learning Facebook ads.

That’s where my team comes in. We run the ads, but we run them into your account with your name and information on them. You keep the benefit of self-run ads (exclusive leads, your brand on the creative) without doing the daily work. We handle setup, creative, launch, optimization, testing, even video editing if you want to make your own video creative. If that sounds like what you’re looking for, come check out our Ad Manager program.

No pressure either way. If you’d rather run them yourself, the five above are enough to get you started.