Fake Leads from Facebook Ads in India: Why They Happen and How to Filter Them

A fake or junk lead is a form submission that cannot become a conversation: the person did not mean to submit, the number is wrong, or a script filled the form. That is a different problem from a genuine but low-quality lead, who answers the phone and simply is not ready to buy. The fixes differ, so it helps to separate the two before you change anything.

The biggest single cause is the form itself. With the default More Volume type, the form opens already filled from the person's profile. Submitting takes one tap. On a phone, in a scrolling feed, that is easy to hit without meaning to, especially on a video ad that people tap to pause.

Then there is the data. Many profiles carry an old number, a typo, or a number that belongs to a family member. The form faithfully passes it along. Add bots and some low-intent placements, and a chunk of your list was never going to pick up.

Meta documents the controls for all this on its lead ads help page, which is where the first two fixes below come from. A common mistake is to blame the audience or the creative first, when the form settings are the cheaper thing to fix.

A Higher Intent form is an instant form type that adds a review screen. After the person fills the form, they see their details and must confirm them before submitting. Meta lists it for Android, iOS and Instagram, and not for desktop.

That one extra screen does two things. It stops the accidental tap, because a second deliberate action is needed. It also lets people correct a wrong number or email that was prefilled from an old profile. Both cut junk leads at the source.

Expect fewer leads. That is the point. If your form produced 100 leads with 40 reachable, a form producing 60 with 45 reachable is better for the sales team even though the dashboard cost per lead goes up.

The switch is simple. Duplicate your current form, change the type to Higher Intent, run both against the same audience for a fixed window, and compare the share of leads your team actually reached. Judge by what happens after the form, not on it.

The second fix is SMS verification. Under quality filters, some accounts show a Require SMS verification option. When it is on, the lead must enter a one-time passcode sent to the number they gave. That removes wrong numbers and most scripted submissions in one move. The catch is that it is not in every account, so open your form settings and see whether you have it. Do not plan a campaign around a feature you have not seen.

The third fix is the cheapest: qualifying questions. Add one or two multiple-choice questions, such as a budget band, a purchase timeline or a preferred location. Meta says more multiple-choice questions usually means fewer, higher-quality leads, so every question trades volume for intent.

Keep it to one or two. A form with six questions on a phone will starve the campaign, and you will end up fighting the delivery system instead of the junk. A budget band and a timeline do most of the work for a property enquiry, and the answers also give your caller something real to open the conversation with.

Avoid free-text questions where you can. Typed answers are slower, get more abandonment and are harder to filter later. Multiple-choice answers can be sorted, scored and routed.

Placement is the fourth fix, and it needs a lighter touch than most blogs suggest. Practitioner reports, including from Jon Loomer and ClickCease, link Audience Network to accidental clicks. That is a report, not a rule for your account. Open the placement breakdown and compare cost per contacted lead by placement before you exclude anything. A placement with cheap leads and a 15% contact rate is worse than one with pricier leads that answer.

The sixth fix is a format check on phone numbers. Indian mobile numbers are 10 digits starting with 6, 7, 8 or 9. A simple regex can reject a number with nine digits or one starting with 2. It will catch typos, and some junk.

It will not tell you the number is live. A perfectly formatted number can be disconnected or belong to someone else. Treat the check as hygiene, not verification, and let the first call or WhatsApp message be the proof.

Every fix has a price, and the price is almost always fewer leads. The table shows what each one filters, what it costs you and how to tell whether it worked. Use the same two measures for all of them: the share of leads contacted and the share that booked a site visit.

The numbers below are round, made-up figures to show how to compare. They are not measured results from any client or campaign.

Say your old More Volume form brings 100 leads on Rs 30,000 of spend. Cost per lead is Rs 300. Your team contacts 45 of them, and 10 turn out qualified, meaning real budget and a real timeline. Cost per qualified lead is Rs 3,000.

Now you switch to a Higher Intent form with a budget question and a timeline question. Same Rs 30,000 spend, but only 60 leads come in. Cost per lead jumps to Rs 500, which looks like a failure on the dashboard. Your team contacts 42 of them (70%), and 15 are qualified. Cost per qualified lead falls to Rs 2,000.

That is the whole argument for judging by the lower funnel. If you only watched cost per lead, you would have reverted the change. Track three columns for each version: contacted %, qualified % and cost per qualified lead. Run them over the same window, and do not stop the test after two days of noise.

The fifth fix is the one that compounds. Instead of letting Meta optimise for anyone who submits a form, you send back what happened next, such as contacted, qualified or site visit booked. Meta's Conversions API is the channel for this, and a CRM can push those stages automatically. Our page on conversion tracking covers the setup side.

There are conditions. Meta suggests about 200 or more leads a month, a stage that most leads reach within 28 days, and a conversion rate for that stage between 1% and 40%. Expect around 3 to 4 weeks for the system to learn. If you generate 60 leads a month, this will be slow, so fix the form first and add this later.

If you sell property, a CRM is where this measurement lives. ArthaLeads imports Facebook and Instagram lead ads automatically, scores each lead from 0 to 100 and flags duplicates before anyone calls. That gives your caller a reason to dial the likely buyer first instead of the oldest row.

Its WhatsApp AI agent can ask a buyer about budget and timeline in English, Hinglish, Hindi or Marathi and save the answers on the lead. That works as a second qualifier after the form, for the leads that survived it.

Be clear about the division of work. ArthaLeads works on the leads that arrive: it helps you organise, prioritise and measure them. The filtering at source still happens in your Meta form settings.

If you want a second pair of eyes on your forms and placements, our performance advertising team reviews exactly this. The honest way to treat fake leads from facebook ads is as a stack: a tighter form, a qualifier, a feedback loop, then a CRM that shows what worked.