Why Your Lead Form Ads Bring Fake Numbers and Unreachable Leads

The campaign report says 300 leads this month at a cost per lead nobody complains about. The sales team says something else: a third of the numbers do not answer, some are switched off or belong to someone else, and the people who do pick up say they do not remember filling in anything. By the third week, the salespeople have quietly stopped calling form leads and are working referrals instead.

This is one of the most common complaints about lead form ads on Facebook and Instagram, and it is usually treated as fraud, bad luck or a “bad audience”. Sometimes it is one of those. Far more often, it is the predictable result of how the form was built and what the campaign was asked to find. This diagnosis covers why instant forms produce unreachable leads, how to check which cause applies to you, and when to stop using the form altogether.

The symptoms: how the problem shows up

Lead form problems look different from a general lead-quality problem. The leads are not just the wrong kind of buyer; many of them never become a conversation at all. If three or more of these describe your account, the form itself is probably part of the cause.

• A high share of numbers never answer, are switched off or are wrong. The sales team marks them “no answer” after two or three attempts.
• People say they did not submit anything. Or they remember tapping an ad but not what it offered.
• The same few names, emails or numbers repeat. Or the email field holds addresses nobody uses any more.
• Answers to your questions are nonsense or all identical. Every lead picked the first option in every drop-down.
• Leads arrive in a spreadsheet days later. Someone downloads the file once or twice a week and forwards it.
• The cost per lead looks cheap, but nobody can say what a customer from this campaign cost.

The last symptom is the one that keeps the problem alive. If the report ends at the form submission, a form that produces many unreachable leads looks like a success.

Why instant forms produce leads you cannot reach

An instant form is designed to make submitting easy. Meta’s lead ads guide describes the form as using information people already shared with Facebook. In practice, the contact fields are often already filled in when the form opens, so a submission can take a couple of taps without leaving the app.

That convenience is exactly what creates the symptoms above. The phone number on someone’s Facebook account may be one they changed years ago, a second SIM they rarely use, or a number they never use for business. Someone scrolling quickly can submit without registering what they agreed to. Nobody has to be dishonest for the lead to be unreachable.

The second force is optimisation. In Meta’s lead form reference, the flag that optimises a form for quality is off by default, and the separate goal that optimises toward a later CRM stage has to be set up deliberately. If the only signal the campaign ever receives is “a form was submitted”, the delivery system learns to find more people who submit forms easily. In our experience, that tends to mean more quick, low-effort submissions over time and fewer deliberate ones.

Put together, an easy form and a volume goal reward each other. The fix is to add friction where it filters, and to give the platform a better signal than the submission.

Five likely causes, and how to check each

1. Submit your own form

Open the ad on a phone, tap through and submit. Note how many fields were already filled in, whether you had to type anything, and whether you were shown what you were agreeing to before the final tap. If you can complete it in a few seconds without reading, so can everyone else.

Then check the form settings. The same lead form reference lists the quality option mentioned above and an option to require phone SMS verification when the person submits. In Ads Manager, look for the equivalent settings when you create or edit the form; their names, placement and availability vary by account and market and change over time, so check what your own account currently offers.

2. Read the answers, not just the count

Export 50 recent leads and read every custom answer. Questions that qualify have answers that cost the respondent a moment’s thought and that you would actually act on differently. “Are you interested?” qualifies nobody. “Which city is the project in?”, “Roughly how many units do you need?” or “When do you plan to buy?” with honest options, including “just researching”, do.

Meta documents several question types, including custom text-input questions, drop-downs, appointment scheduling, store locators and conditional questions, which are set up from an uploaded file of valid answer combinations (Meta for Developers). A typed answer is harder to give carelessly than a pre-selected one. If every lead chose the first option in every list, the questions are decoration.

3. Check what Meta is told after the submission

Ask a simple question: does Meta ever learn which leads answered the phone, qualified or bought? If not, the campaign can only optimise for submissions.

Meta’s lead ads guide says leads can be updated into your CRM as they arrive and that CRM data about those leads can be shared back with Meta to enable quality lead optimisation. Its conversion leads integration guide sets out the conditions for optimising toward a later CRM stage: leads from instant forms, at least 200 leads a month, data uploaded at least once a day, the target stage reached within 28 days of the lead, and a conversion rate to that stage of between 1% and 40%. It recommends storing the 15 to 17 digit Meta lead ID in your CRM, and says this goal “may yield higher quality leads”, which is a possibility, not a guarantee.

If your volume is below those thresholds, the CRM stages are still worth recording. They tell you the real cost per reachable lead and per customer, even if the platform cannot yet optimise on them.

4. Measure the delay between form and first call

A person who submitted a form a few minutes ago remembers it. The same person three days later often does not, and in our experience the “I never filled anything” reply is often a memory problem rather than a fake lead. Pull the submission time and first call attempt for your last 30 leads. If you sell in the UAE, check the UAE rules on calling leads before you speed up the calls.

Meta’s guide to retrieving leads describes two routes: webhooks, which send a real-time update for each new lead, and bulk reads of leads by ad or by form. If your leads arrive by weekly download, the delay is a choice, and an easy one to remove through a CRM integration or a connected tool.

5. Break contact rate down by delivery

Add a “reached: yes or no” column to your lead sheet, then group it by placement, region, age band and device. If one placement or region produces most of the unreachable leads, you have a targeting problem, not a form problem. A B2B supplier that only serves Riyadh and Jeddah, for example, gains nothing from cheap leads elsewhere in the country. Meta’s lead form reference also includes a tracking parameters field, which lets you pass your own campaign labels into each lead’s data so this grouping stays reliable.

What to fix first

Fix in this order. Each step changes the data the next one depends on.

1. Speed of follow-up. Connect the form to your CRM or inbox (our checklist covers what your CRM must record) so every lead is called or messaged the same day, ideally within the hour during working time. In our experience, this alone can recover a good share of the “unreachable” pile with no change to the ads.
2. Friction in the form. Turn on the quality-focused form option and, if your account and market offer it, phone verification. Add one or two typed or meaningful questions. Expect fewer leads; that is the intended result.
3. Honest qualifying options. Give people a respectable way to say “just researching” or “outside your area”, so they do not submit false answers to get through.
4. A clear first message. Write the opening line of the call or WhatsApp follow-up so it reminds the person what they asked for: the ad, the offer and the date. Our sales enablement work covers these scripts, and our guide to a WhatsApp flow from first message to quotation sets out the opening and qualifying messages. For GCS, a new fire-safety brand in Saudi Arabia, we prepared WhatsApp replies for the most common inquiries and a message flow from first contact to quotation request, so no lead gets lost (GCS case study).
5. Outcome feedback. Record contacted, qualified and won in the CRM against each lead ID, then send those stages back to Meta once your volume supports it.
6. Targeting and placements. Exclude the regions and placements that your contact-rate breakdown showed were producing unreachable leads.

Report three numbers side by side from now on: cost per lead, cost per reached lead and cost per customer. A rising cost per lead next to a falling cost per customer is progress; if management asks why the first number went up, our diagnosis of why cost per lead keeps rising helps explain it. If your issue is wider than form leads, our earlier diagnosis on leads that never turn into customers follows the problem through to the quotation stage.

Hypothetical example: the same spend, read three ways

The figures below are invented to show the arithmetic; they are not a benchmark or a client result. The spend is the same in both months, in any currency. In the second month, the form has the quality option on, one typed question and same-day follow-up.

Read only the leads line and the fix looks like a failure: cost per lead doubled. Read the reached line and nothing changed. Read the customers line and each customer cost a quarter less, because the people sales reached had filled in the form deliberately. The quotation and customer figures in the second month are the assumption to test in your own account; the arithmetic is the reason to report all three costs.

When a WhatsApp or landing-page route is the better choice

Instant forms suit offers where a quick expression of interest is enough and your team can follow up fast. They suit high-consideration B2B or high-value purchases less well, because the ease that drives volume also removes the moment where a serious buyer explains what they need.

A WhatsApp route removes the wrong-number problem by design: the person writes from the number they actually use. Meta’s Conversions API for business messaging covers conversations started from click-to-Messenger, click-to-WhatsApp and click-to-Instagram Direct ads, with events including LeadSubmitted, QualifiedLead and Purchase, so chat outcomes can also be sent back to Meta. The setup and qualifying questions are covered in our guide to click-to-WhatsApp ads for B2B inquiries.

A landing page adds the most friction and the most context. It is usually the right call when the sales team needs the buyer to have read something before the first conversation.

When an outside view helps

Most of the checks above need only a lead export and an honest sales meeting. Outside help earns its place when the form, the ad account, the CRM and the sales follow-up are owned by different people who each blame the others, or when you are about to raise the budget on a campaign the sales team has stopped trusting.

For lead form campaigns, our performance marketing work sets up the form, the quality settings and the CRM feedback, and the same team writes the first call and WhatsApp messages, so the form and the follow-up are judged on one number: cost per customer.

Your next step: export your last 100 form leads with submission time, first call time and outcome. If the cause is not obvious from that file, bring it to a business diagnosis session; we will read it with you, separate form, targeting and follow-up causes, and agree which fix to make first.