The short version: Most Facebook lead ads fail because the form is too easy, the follow-up is too slow, and the targeting is too broad, not because the creative wasn’t clever enough. Fix the qualifying questions in your lead form, get sales onto new leads within five minutes, and put your retargeting budget ahead of your cold prospecting budget. Do those three things before you touch a single headline.
Worth reading next: What Makes a Facebook Ad Campaign Convert Leads Into Customers.
The problem isn’t your ad, it’s what happens after someone clicks it
I’ve sat in enough client review calls to know the pattern by heart. Cost per lead looks brilliant on the ads manager dashboard, £6, £9, sometimes £4 in a good week, and the marketing team is thrilled. Then three weeks later sales tells you 70% of those leads never answered the phone, half gave a fake surname, and the two who did book a call weren’t remotely ready to buy.
That’s not a Facebook problem. That’s a “we optimised for the wrong thing” problem.
Facebook’s algorithm is very good at getting you cheap leads. It is not good at getting you good leads unless you tell it, clearly and repeatedly through the signals you feed it, what a good lead looks like. Most businesses never do that part, so they get exactly what they asked for: volume.
Start with the objective, not the creative
If you’re running the Leads objective with an instant form, Meta optimises for people who are likely to submit that form. Full stop. It has no idea whether your sales team can close them. If you’re running traffic to a landing page and hoping people fill in a form there, you’re optimising for clicks, which is a different, weaker signal entirely.
For most B2B and high-ticket B2C offers I now recommend a hybrid: run the Leads objective but send traffic to a proper landing page instead of an instant form whenever the offer is worth more than about £500 in lifetime value. The extra friction of leaving Facebook and loading a page costs you some volume, but the people who land on the page and still fill it in are self-selecting for intent. I’ve watched cost per lead go up by 30 to 40% doing this and cost per qualified appointment drop by half at the same time.
If you’re running instant forms because you need the volume (think local services, quote-based businesses, anything with a low average order value), that’s fine, but the form itself needs work, and that’s where almost everyone leaves money on the table.
Your lead form is quietly ruining your conversion rate
Here’s the bit most agencies won’t say out loud because it sounds like they’re arguing against their own cost-per-lead numbers: making your Facebook lead form easier to fill in makes your business worse off.
An instant form with three fields, name, email, phone, will get you the lowest cost per lead you’ll ever see. It will also get you the highest percentage of tyre kickers, wrong numbers, and people who tapped the form by accident while scrolling, which happens far more than people admit. Meta’s own instant forms are designed to reduce friction, and reduced friction means reduced commitment.
Add two qualifying questions and your cost per lead will rise, usually by 15 to 25%. Your show-up rate and close rate will rise more than that. I worked with a home improvement client running kitchen renovation ads where the original form (name, phone, postcode) was generating leads at £11 each and a sales team that was burning out chasing ghosts. We added two multiple-choice questions, budget range and timeframe, and cost per lead climbed to £16. Appointments kept went from 22% to 51%. The client stopped complaining about “expensive leads” within a month because the leads that came through were people who’d already told you they had £8,000 to spend and wanted the work done this quarter.
That’s the uncomfortable truth nobody wants to put in a case study: the metric everyone screenshots (cost per lead) is often the least useful number in the whole campaign.
Speed to contact matters more than most creative decisions
You can have the best-performing ad on the platform and still waste it if nobody calls the lead for six hours. Response time on Facebook leads decays fast, far faster than on Google leads, because the person didn’t go searching for you. They were scrolling, saw something interesting, tapped, filled in a form in eleven seconds, and moved on with their day. If you call them two hours later they’ve forgotten they did it.
Aim for contact within five minutes during business hours. If that’s not realistic with your team, at minimum set up an instant SMS or WhatsApp auto-reply the second the form is submitted, then a human call within the hour. Businesses that connect their lead ads directly into a CRM so a task or call is triggered automatically see far better contact rates than businesses relying on someone logging into Ads Manager once a day to export a CSV. I’ve written more on this in Best CRM for Facebook Leads: CAPI Without the Headache, which covers exactly this connection point and why so many businesses lose leads simply because nobody was told one had come in.
Get the data flowing back to Facebook, not just to you
This is the part that gets skipped constantly and it’s costing people money on every single campaign. Facebook’s optimisation only gets smarter if you send back information about what happens after the lead lands. If you never tell Meta which leads turned into paying customers, the algorithm keeps chasing more of the same low-intent submissions that got you the first cheap batch.
Set up offline conversions or the Conversions API so that when a lead becomes a qualified appointment or a sale, that event gets pushed back to Meta, ideally within a day or two so it’s still inside the attribution window. Tools like LeadsBridge exist specifically to automate that connection between your lead ads and your CRM without needing a developer, and I’ve broken down how that pipeline works in Unlock the power of automated lead generation with Meta lead ads and LeadsBridge. Once you do this the algorithm starts finding you more people who look like your actual buyers, not just more people who look like anyone who’ll tap a form.
Retargeting deserves more budget than cold prospecting, not less
Businesses new to Facebook lead gen almost always over-invest in cold audiences and under-invest in warm ones. A common split I see is 80% cold, 20% retargeting. Flip that. Someone who has visited your pricing page, watched 50% of your video ad, or engaged with your Instagram post is a wildly better prospect than someone who has never heard of you, and converting them costs a fraction as much.
For most of my clients I now run something closer to 40% cold prospecting for reach, and 60% split across warm retargeting: website visitors from the last 30 and 60 days, video viewers, page engagers, and a lookalike built from actual customers rather than just form-fillers. That last distinction matters. A lookalike audience built from your customer list will outperform one built from your leads list almost every time, because leads include everyone who filled in the form out of mild curiosity, and customers are the ones who paid.
Creative that earns a real answer, not just a swipe
Facebook rewards ads that get genuine engagement, comments, shares, video watch time, over ads that just get impressions. For lead gen specifically, the creative that performs best in my experience isn’t the polished, agency-shot video. It’s the founder or a real team member talking straight to camera for 30 to 45 seconds explaining who the offer is for and, more importantly, who it isn’t for.
That second part is the piece people skip. Telling your audience “this isn’t for you if X” does two things: it filters out low-intent clicks before they even fill in the form, and it makes the people it is for trust you more because you’re not chasing everyone. I’ve seen this single line added to ad copy drop form completions by around 10% and raise appointment show-up rate by 20% or more, because the people left in the pool match the offer.
Test three or four creative angles at once (problem-focused, result-focused, social proof, and the “who this isn’t for” style), let each run for at least three to four days before judging it, and kill anything with a click-through rate under about 1% for cold traffic. If you want a wider menu of what to test and how to structure it, Top 5 Business Marketing Strategies for Facebook covers the broader strategy layer this sits inside.
Budget, bidding, and the patience most people don’t have
Facebook’s algorithm needs data to work, and most campaigns get switched off or edited before they’ve collected enough. As a rough rule, you want roughly 50 conversion events per ad set per week before you judge performance and start reallocating budget. If your lead cost budget is £500 a week and your leads cost £15 each, that’s about 33 leads, not quite enough for the system to fully leave the learning phase, so either raise budget or consolidate ad sets rather than spreading thin across ten variations.
Use cost cap or bid cap once you know roughly what a qualified lead is worth to you, rather than leaving everything on lowest cost indefinitely. Lowest cost will always chase volume; a cost cap forces Meta to hold quality at a level you’ve decided matters, even if it means fewer total leads. For a business where the customer lifetime value is high, that trade is almost always worth making.
A real example from a campaign I ran last year
A financial coaching client came to me spending £3,000 a month on Facebook leads through an instant form, getting leads at £9 each, roughly 330 a month, and converting maybe 1.5% into paying clients. Sales was drowning and morale was low because nobody wanted to call another unqualified lead.
We made three changes over six weeks. First, we added a qualifying question about current income bracket and made the offer specific rather than generic (“free strategy call” became “free 20-minute plan for people earning over £60k who want to retire five years earlier”). Second, we set up automated SMS the moment a form was submitted and pushed the CRM data on booked calls back into Meta as a custom conversion. Third, we rebuilt the audience split to 50% retargeting on video viewers and page engagers, 50% cold with a lookalike sourced from paying clients rather than leads.
Cost per lead rose to £19. Total leads dropped to around 155 a month. Conversion to paying client rose to 6.8%. Monthly revenue from the same ad spend roughly tripled, and the sales team stopped hating their job. Nobody at the start of that project would have signed off on “let’s make cost per lead go up,” and that’s exactly the point.
The tools that make the operational side manageable
None of this needs to be done manually inside Ads Manager with a spreadsheet on the side, though plenty of businesses still run it that way. There are proper management platforms that handle bulk creative testing, automated rules, and reporting across accounts, and I’ve reviewed the ones worth paying for in The Best 7 Facebook Ads Management Tools. If you’re managing lead gen at any real scale, pairing that with a wider ROI-focused toolset like the ones covered in 10 Facebook Ad Tools that Improve your ROI will save hours a week that are better spent on creative and follow-up quality.
And if you’ve concluded that lead gen isn’t a job you want to build in-house at all, that’s a fair call for a lot of businesses. Working with a specialist agency rather than trying to run this as a side project can be the more profitable route, and I’ve put together a rundown of options in 15 Reliable Lead Generation Companies to Work With.
What I’d tell you to do this week if you only had an hour
- Add one or two qualifying questions to your lead form even if cost per lead rises.
- Set up an instant automated reply (SMS or WhatsApp) triggered the second a form is submitted.
- Check your ad account’s audience split and move at least 20 percentage points from cold to warm retargeting if you’re currently over 70% cold.
- Push at least one offline conversion event (booked call, sale) back into Meta so the algorithm learns from real outcomes, not just form fills.
- Write one piece of ad copy that says who the offer isn’t for, and test it against your current best performer.
Frequently asked questions
What’s a good cost per lead for Facebook ads?
It depends entirely on the industry and offer value, so there’s no single benchmark worth chasing. UK service businesses typically see £5 to £20 per lead for consumer offers and £20 to £80 for higher-ticket B2B, but the number that matters is cost per qualified appointment or cost per sale, not the raw lead cost.
Should I use Facebook’s instant lead forms or send traffic to a landing page?
Use instant forms for lower-value, high-volume offers where speed and low friction matter most. Use a landing page for anything over roughly £500 in lifetime value, since the extra effort of leaving Facebook filters out low-intent clicks and improves lead quality, even though it raises your cost per lead.
How quickly should I follow up with a Facebook lead?
Aim for contact within five minutes during business hours wherever possible. Facebook leads decay faster than search-based leads because the person wasn’t actively looking for you when they filled in the form, so an automated instant reply followed by a human call within the hour is the minimum worth setting up.
Why is my cost per lead low but my sales still bad?
This usually means Facebook is optimising purely for form completions rather than customers, which happens when the form has no qualifying questions and no data is being fed back about which leads convert. Adding qualifying questions and connecting your CRM outcomes back to Meta through conversions or the Conversions API fixes this in most cases within a few weeks.
Related reading: How to Use Facebook Hashtags to Get More Reach (What They Do in 2026) and How to Tell Who Unfollowed You on Facebook (2026 Guide).
If you want the full breakdown, here is everything I know about social media marketing.
