Facebook Lead Ads Bringing Bad Leads? How to Fix Lead Quality

It’s Monday morning. The weekend brought in 31 Facebook leads, and you feel good about that for about forty minutes. Then you start calling. Three numbers are disconnected. One is 713-555-1234. A woman in Cypress tells you she never filled anything out. Eleven go to voicemail, and the voicemail box is full on two of them.

If your Facebook lead ads are bringing bad leads, look hard at the setup before you blame the platform. Meta optimizes for whatever you count as a lead. Count every tap on “Submit” and it will go find the people most likely to tap Submit, who are rarely the people most likely to buy. Fixing it means a slightly harder form and questions that actually filter. Usually a different offer, too. Then somebody has to call within minutes, and once you have the volume, Meta needs your CRM results so it can learn what a good lead looks like.

What follows is the order we work through it for local businesses around The Woodlands and Houston, actual settings included. Fair warning, your cost per lead will go up. Your cost per customer should come down, and that’s the number that pays the bills.

Why are my Facebook leads so bad?

Start with how an instant form works. Someone scrolling Instagram in a car line at a Spring ISD school taps your ad. A form opens inside the app, already filled in with the name, email and phone number from their profile. Two taps later, you have a “lead.” They never left the app, never saw your website, and may have been thinking about something else the whole time.

That convenience is what Meta is selling, and it cuts both ways. Cheap leads, lots of junk. What the junk looks like once you start dialing:

• “I don’t remember filling that out.” Autofill did the work and the whole thing took eight seconds.
• Fake or dead numbers. Some people want the freebie but not the phone call, so they make one up. Others never updated the number on their profile.
• No answer at all. Part of that is Houston reality (plenty of people here won’t pick up an unknown number). Part of it is you calling two days later.
• A lead from Beaumont, for a pool builder in Magnolia.

The algorithm then makes it worse. With the campaign goal set to “maximize leads,” Meta’s delivery system learns who submits forms cheaply and goes looking for more people like them. Junk trains it to find more junk, and it’s very good at its job.

More volume vs Higher intent: change the form type first

When you build an instant form, Meta asks you to pick a form type. The default is More volume, which is built to be quick to fill out and submit. The alternative is Higher intent, which adds a review screen where the person has to look at their information and confirm it before the form goes through, per Jon Loomer’s instant forms guide.

One extra screen sounds trivial. It catches accidental submits, though, and because the phone number is sitting right there in front of them, people with an old number on their profile tend to fix it. Social Media Examiner recommends Higher intent for exactly this reason, unless you have another efficient way to qualify.

Two more switches live nearby.

SMS verification. Under Quality filters, some accounts show “Require SMS verification to submit form.” The person gets a one-time code by text and has to enter it, or the submission doesn’t count as a lead. Jon Loomer’s roundup of newer lead ad features found it in three places (as a quality filter, a separate form type, and an option inside Higher intent) and says they all work the same way. If fake numbers are your main headache, this is the strongest filter Meta offers.

Some accounts also have an autofill checkbox on the email and phone fields. Turn it off and people have to type their real, current details. Fewer finish. The ones who do actually wanted to.

Meta rolls these out unevenly, so if one is missing, don’t lose a day hunting for it. Use what you have.

Expect your cost per lead to climb when you do this, sometimes a lot. We’ve made peace with that. A $28 lead you can reach beats a $9 lead that was never real, and the math usually works out in favor of the filtered form within a few weeks.

How to improve Facebook lead quality with qualifying questions

Contact fields alone tell you nothing about whether someone is a buyer. Custom questions do. Meta’s instant forms support several question types: multiple choice, short answer, conditional questions (built from a CSV so later answers depend on earlier ones), appointment requests and sliders.

The part most people skip is conditional logic. Turn it on and each question gets a Logic column with three routes, as Jon Loomer’s instant forms guide lays out: Go to a question, Submit form, or Close form. Close form sends a non-fit person to a separate ending page without submitting their contact info. The renter in Atascocita, or the guy who wants a $200 fence, gets turned away politely and never becomes a lead you pay for or call.

Say you run a remodeling company in The Woodlands. A form that filters might look like this:

1. “Which area is the home in?” Multiple choice: The Woodlands, Spring, Conroe, Magnolia, Tomball, Other. “Other” routes to Close form.
2. “Do you own the home?” Yes or No. No routes to Close form.
3. “What’s the project?” Kitchen, bathroom, whole home, other.
4. “When do you want to start?” In the next 3 months, 3 to 6 months, just researching.
5. “What budget range are you planning around?” Give real ranges that start where your jobs start.
6. One short-answer question, like “Anything we should know about the project?” Having to type a sentence filters out a surprising number of people.

Keep it to four to six questions. Past that, even good prospects bail on a phone screen. And don’t make every answer a trap; “just researching” people can go into a nurture list instead of your call queue.

Social Media Examiner notes Meta itself says conditional logic may raise cost per lead. It will. Let it.

Offer framing that attracts freebie hunters (and what to say instead)

Plenty of lead problems are really offer problems. The ad does a lot of the filtering before anyone sees a form.

“Enter to win a free roof.” “FREE estimate, no obligation, claim yours today!” “Get a $500 gift card when you book.” Every one of these is a magnet for people who like free things. Some are buyers. Most are not, and Meta will happily find you thousands of them in Harris County alone.

What tends to work better for local service businesses:

• Show a price signal. “Kitchen remodels in The Woodlands typically start around $45,000” scares off the people who were never going to hire you.
• Tell them what happens next. “A project manager will call you within an hour to schedule a walkthrough.” Anyone who doesn’t want a call drops out on their own, and the rest are expecting you when the phone rings.
• Match the ask to the size of the job. For a $15,000 pool remodel, “Book a design consult” brings fewer, better leads than “Get a free quote.”
• Drop the giveaways unless you’re building an email list. If you are, those people are subscribers. Keep them away from your sales team.

Put the same language on the form’s intro screen. The intro is where you can be blunt: “This form is for homeowners in Montgomery County planning a project in the next six months.”

Facebook leads not answering? Fix the first five minutes

A lot of “bad leads” are good leads called too late. A Facebook lead has a short memory. They submitted while scrolling, then went back to scrolling, and by Wednesday your call is just another unknown Houston number.

Speed starts with plumbing. If leads sit inside Meta until someone remembers to download a CSV, you lose. Options, from simplest:

• Email. Jon Loomer’s roundup of newer lead ad features describes a newer lead delivery option right in the form setup that emails new leads to whoever you choose. Better than nothing, barely.
• A direct CRM integration. Meta offers direct integrations with many CRMs, and tools like HubSpot or Zapier can push each lead into your CRM and fire a text alert to whoever is on call.
• An automatic text. The moment the lead lands: “Hi Maria, this is Dave with Lakeside Pools. Got your request. I’ll call you from this number in the next few minutes.” Now your number isn’t a stranger.
• Messenger. The form’s ending screen has a “Connect with leads on Messenger” option that opens a chat after they submit.

Then the calls. Most of our clients run some version of this:

1. Call within five minutes during business hours (the auto-text buys you some goodwill if it’s ten).
2. No answer: leave a short voicemail and send a text.
3. Call again that evening, around 6:30, when people are home.
4. Call and text the next morning.
5. Day 3 and day 7: one text each, then move them to email.

Log every outcome in the CRM: bad number, no answer, not interested, qualified, booked. You need those labels for the next section, and for the scorecard at the end.

If after-hours leads are a big share (they usually are for home services here, because people browse at night), an answering service or a part-time evening setter often pays for itself faster than any ad change.

Tell Meta which leads were good: Conversions API for CRM

Everything so far filters on the way in. This step changes what Meta optimizes for.

With the Conversions API for CRM, you send your lead stages back to Meta: this lead was qualified, this one booked, this one bought. Meta can then use the “maximize number of conversion leads” goal, which Social Media Examiner describes as built to find people more likely to convert after you follow up, even if total lead count drops.

The requirements rule out a lot of small businesses. Per Meta’s Conversions API for CRM documentation, you need to:

• Generate at least 200 leads a month.
• Upload data at least once a day.
• Pick a target stage that happens within 28 days of the lead.
• Have that stage convert at between 1% and 40% of leads.

A one-location HVAC company spending $2,000 a month probably isn’t at 200 leads, and that’s fine. Set up the CRM stages anyway, because clean stage data is what makes the scorecard below work, and you’ll be ready if volume grows. Many CRMs (HubSpot included) have a native Meta connection for this, so it may not need a developer.

Below that volume, we’d send the traffic to a landing page with your pixel and server-side events, and count only the actions you actually want (a booked appointment, a completed quote request) as the conversion. Meta then optimizes toward something closer to a sale.

Facebook lead form vs landing page: which brings better leads?

Instant forms win on cost per lead. Landing pages usually win on quality. Which one you want depends mostly on how fast your team follows up and what a job is worth to you.

There’s also a hybrid. Jon Loomer’s roundup of newer lead ad features describes a “Website and Instant Forms” conversion location where Meta can send each person to whichever one it predicts will work. We’d only try it once both versions work well on their own.

If you go the landing page route, build a page for this ad. (Send it to your homepage and you’re paying for people to read your About page and leave.) One offer, one form, a phone number, real photos of your work in Houston-area homes, and reviews. Our landing pages guide covers the structure.

For most home service and medical clients, we start with a filtered Higher intent form because it’s fast to test, then add a landing page version beside it once there’s budget for two ad sets. Compare them on cost per booked appointment. Cost per lead will mislead you here.

Audience and placement settings that quietly hurt quality

Some junk is a placement problem.

Audience Network places your ad inside other apps and sites, and it has a long reputation for accidental taps. Social Media Examiner suggests excluding it if it produces low-quality leads. How you do that changed in 2026. Per Relevant Audience’s write-up of Meta’s placement change, Meta is rolling out a change where Leads campaigns (along with Sales, Engagement and App promotion) no longer let you pick placements manually at the ad set level. Exclusions for Audience Network, Facebook Marketplace and the right column now live in account-level controls, and those can be set per objective. If your Ads Manager still shows the old placement picker, the rollout hasn’t reached you yet.

A few other settings worth a look:

• Location. A 25-mile radius around The Woodlands reaches well into north Houston. If you don’t want work in Aldine or Humble, don’t pay for it. List the zip codes or cities you actually serve.
• In Ads Manager, break results down by placement and by age. When one placement or one age band produces most of your no-answers, you’ve found a leak.
• Special Ad Categories. Housing, employment, credit and financial services ads run with limited targeting by design. Mortgage and real estate advertisers should lean even harder on form questions, because the audience settings can’t do the filtering.

For the broader picture of building Meta campaigns for local businesses, see our Facebook and Instagram ads guide.

A lead-quality scorecard you can run on last month’s leads

Pull last month’s leads into a spreadsheet, add an outcome column, and fill it in from your call notes. Then score them against this. The thresholds come from our own local service accounts, so treat them as rules of thumb and adjust to your business.

The last row is the one that decides whether the campaign lives. Our cost per lead guide shows how to work backward from job value to what you can afford.

Run the scorecard monthly. When you change something, change one thing at a time and give it two to three weeks, or you’ll never know which change helped.

When to fix it yourself and when to get help

Most of this you can do in an afternoon: switch to Higher intent, turn on SMS verification if you have it, add four questions with Close form logic, rewrite the offer, set up an auto-text. Do those first.

The CRM integration and Conversions API take more skill, and so does a landing page that actually converts on a phone. Outside help tends to pay off there, especially if your team is already stretched answering the leads you have.

If you want a second set of eyes before you change anything, our free Ad Spend Leak Check is a 10-minute video showing where your Meta budget is leaking, in your inbox within 48 hours. Or see how our paid social team runs lead campaigns for businesses from Conroe to Sugar Land.

Frequently asked questions

Why are my Facebook leads fake?

Instant forms autofill a person’s name, email and phone from their profile, so submitting takes seconds and many people do it without real intent or with an outdated number. Freebie offers attract people who type fake numbers on purpose. Switch to the Higher intent form type, turn on SMS verification if your account has it, and ask a few qualifying questions.

How do I improve Facebook lead quality?

Use the Higher intent form type, add four to six qualifying questions with conditional logic that closes the form for non-fits, rewrite the offer so it shows price signals instead of giveaways, call leads within minutes, and send lead outcomes from your CRM back to Meta so it optimizes for qualified leads rather than form fills.

Is a Facebook lead form or a landing page better?

Instant forms produce cheaper leads and are fast to launch, but quality is mixed. Landing pages cost more per lead and usually bring more serious prospects because people chose to leave the app and see your prices and reviews. For big-ticket local services, test both and compare cost per booked appointment, not cost per lead.

Why do Facebook leads not answer the phone?

Most people submit a form while scrolling and forget about it quickly, and many won’t answer an unknown number. Call within five minutes, send an automatic text first so your number is recognized, call again in the early evening, and follow a set call and text pattern for a week before giving up on a lead.

Does the Higher intent form cost more per lead?

Usually yes. The review screen and any extra questions mean fewer people finish the form, so cost per lead rises. What matters is cost per qualified lead and cost per booked job. In local service accounts, filtering out fake and accidental submissions often lowers the cost of each real customer even when the lead price goes up.

Written by Sean Melton

Sean Melton is the founder of First Byte, a digital marketing agency in The Woodlands, TX. A former global CMO of a $250M software company, he has spent his career inside the businesses he markets. His team runs paid social, search, CTV and creative for brands like Equinox, Kroger and Viceroy, and shares what works here.