Real estate agents love to ask one question when it comes to online leads: How cheap can I get them?
That question is not wrong, but it is incomplete. Cheap leads only matter if you understand what kind of lead you are buying, where that lead is in the funnel, and what your follow-up system needs to do after the form is submitted.
Right now, Meta ads are interesting again for real estate. In some markets, we are seeing leads come in under $3 each and, in others, under $5. That does not mean every agent will see the same cost. It also does not mean these leads are ready to write an offer tomorrow.
But it does mean the game has changed.
5 things to know about Meta’s new algorithm
Here are five things agents need to understand about Meta’s newer ad algorithm.
1. Forget most of what you knew before 2025
For years, agents tried to win with audience targeting. They picked interests, behaviors, demographics and categories they thought might represent buyers or sellers.
Now the system is much more driven by AI signals and intent-based behavior. Meta’s newer algorithm, often referred to as Andromeda, is trying to understand who is showing signs they may be interested in what your ad offers instead of relying only on the manual audience you build.
People usually start looking long before they talk to an agent. They search on their phone. They click on homes. They look at neighborhoods. They start educating themselves.
That is where Meta can become powerful again. It can help you get in front of people earlier in the process, before they raise their hand somewhere else.
2. These are not right-now leads
This is the part agents need to understand before they get too excited about cheap cost per lead. Most of the Facebook leads we are seeing right now are not immediate business. They are not usually people saying they need to buy this weekend or list tomorrow. They are earlier than that.
That is not a bad thing, but it changes how you measure the campaign. I look at many of these leads as database-building leads. You are buying early attention, adding people to your ecosystem and creating a chance to nurture them before they become a higher-intent lead somewhere else.
If someone starts looking 12 to 18 months before a move, getting them early gives you time to educate them, build trust and shape the conversation before they are deeper in the decision process. It also usually means you can acquire them at a lower cost.
The mistake is expecting a top-of-funnel Meta lead to behave like a Google search lead. Google is often stronger for immediate intent. Meta can be stronger for building future pipeline.
Those are different jobs.
3. The algorithm needs creative variety
The new system rewards agents who give it more creative options.
That means you cannot run one or two ads and expect the algorithm to figure everything out. You need multiple creatives inside each ad group so the system can test different angles, hooks, visuals and offers.
They also need to be meaningfully different. Not the same ad with a slightly different headline. Not the same listing photo with a different caption. You need different hooks, images, video styles, buyer angles, seller angles and emotional entry points.
In many campaigns we are seeing perform well, 15 to 20 creatives per ad group is a stronger starting point. Some markets may need more.
The reason is simple: The algorithm needs information. The more quality inputs you give it, the more chances it has to find the right people. If you only give it one angle, you limit what it can learn.
4. Instant forms are usually the better play
Years ago, a lot of agents ran Facebook ads to an IDX page and tried to force registration there. That can still work in some situations, but it is not usually the best play for low-cost Meta leads right now.
Meta wants to keep people on its platform. When you send users away to another website, you add friction and usually increase cost.
Instant forms remove that friction. The lead clicks, fills out the form inside Facebook or Instagram and can be sent directly into your CRM through the Facebook Lead Center and integrations many CRMs already support.
The other improvement is form quality. With dual opt-in options, you can require an extra verification step, such as a code. That may reduce total form submissions, but it can improve quality because fake information becomes harder to submit.
Cheap leads are not useful if the contact information is garbage. A slightly higher quality lead at a slightly higher cost can still be the better business decision.
5. Your market still matters
The last thing agents need to understand is that results are market-driven.
Just because one market gets leads under $3 does not mean every market will. Competition, audience size, local demand, affordability, offer quality, creative quality and how many other advertisers are running campaigns in the same area all matter.
This is becoming more granular. It is not just large market versus small market anymore. It can vary by city, niche, price point and offer.
That is why testing matters.
Do not assume the campaign failed because your cost per lead is not exactly what someone else posted online. Look at your market, your offer, your creative mix, your follow-up and your cost per real conversation.
Cost per lead is only the first number. Cost per appointment and cost per closing are the numbers that decide whether the campaign is working.
Cheap leads still need a real system
Meta ads can be a powerful database builder right now, but only if agents understand the role those leads play.
They are often early. They need nurture. They need education. They need a follow-up system that does more than call once and complain that nobody answered.
The opportunity is not just getting leads under $3. The opportunity is getting in front of consumers earlier, for less money, and building the relationship before they are ready to move.
That is where Meta can still be valuable.
Not because every lead is ready now, but because future clients usually start paying attention long before they ever raise their hand.
