If your ads get clicks but no leads, the problem is almost always after the click, not the ad platform. Check five things in order: tracking, traffic intent, message match, landing page friction, and lead follow-up. Most small businesses lose customers on the landing page or in slow follow-up, not in targeting.
- Check tracking first: many "zero conversion" accounts are getting leads that are simply not reported.
- Benchmarks: the median landing page converts 6.6% of visitors (Unbounce, 2024). If you are far below, the page is the problem.
- Simple copy wins: pages written at a 5th–7th grade level converted at 11.1% vs 5.3% for professional-level copy.
- The leak nobody talks about is follow-up. A lead nobody calls back is a lead you paid for twice.
Clicks, leads, customers: what "good" looks like
Before you change anything, know what normal looks like. These are the benchmarks we use to judge whether an account has a traffic problem or a conversion problem.
| Metric | Benchmark | Source |
|---|---|---|
| Landing page conversion rate (median, all industries) | 6.6% | Unbounce Conversion Benchmark Report, 2024 |
| Landing page conversion range by industry (median) | 3.8% – 12.3% | Unbounce, 2024 |
| Facebook lead campaigns, average conversion rate | ≈ 7.7% | WordStream / LocaliQ benchmarks, 2025 |
| Facebook lead campaigns, average cost per lead | ≈ $27.66 | WordStream / LocaliQ, 2025 |
| Google Search ads, average conversion rate | ≈ 7.5% | WordStream / LocaliQ, 2025 |
| Google Search ads, average cost per lead | ≈ $70 | WordStream / LocaliQ, 2025 |
Step 1: Make sure it is not a tracking problem
The fastest win is often finding out you were getting leads all along. Compare what Meta or Google report with what actually reached you: form emails, CRM records, booked calls, phone logs. If the numbers do not match, fix tracking before you spend another dollar.
What to check
- The Meta Pixel fires a Lead event on the thank-you page, not on the form page.
- The Meta Conversions API sends the same Lead from your server, with the same event ID (so it is counted once).
- Google Ads has a conversion action for leads, and enhanced conversions are on.
- Every ad link carries UTM parameters, so GA4 knows which campaign sent the lead.
Without this, the ad platforms optimize for the only thing they can see: clicks. Which is exactly the problem you have.
Step 2: Are the right people clicking?
Google Ads: read the search terms report
Open the search terms report and read every query that cost you money in the last 30 days. Job seekers, people looking for free advice, and searches for a different service drain budget fast. Add them as negative keywords.
Facebook Ads: optimize for leads, not clicks
A campaign optimized for traffic finds people who click, not people who buy. Use a Leads or Sales objective, and give each ad set enough budget to collect data: Meta says an ad set needs around 50 optimization events in a week to leave the learning phase.
Step 3: Message match: does the page continue the ad?
If the ad says "Free roof inspection in 48 hours" and the click lands on a homepage that says "Welcome to Smith Roofing", the visitor has to work to find the offer. Most will not. The headline of the landing page should repeat the promise of the ad, in almost the same words.
| Ad says | Weak landing headline | Matching landing headline |
|---|---|---|
| Free roof inspection in 48 hours | Welcome to Smith Roofing | Book your free roof inspection, within 48 hours |
| $99 new patient exam | Family dentistry since 1998 | Your $99 new patient exam, X-rays included |
Step 4: Remove friction from the landing page
Speed and mobile
Most ad traffic is mobile: Unbounce found 83% of landing page visits come from phones. Test your page on a real phone, on cellular data, not on office Wi-Fi.
Form fields
Every field is a reason to leave. Ask only for what you need to call the person back. In one widely cited test, cutting a form from 11 fields to 4 raised conversions by 120%. Your mileage will vary, but the direction almost never does.
Reading level and one clear action
Short sentences, plain words, one button. Add trust where the decision happens: reviews, real photos, a guarantee next to the form.
Step 5: The leak nobody talks about: follow-up
This is where small businesses lose the most, and none of the ad dashboards show it. In a study of 2,241 US companies, Harvard Business Review found 23% never responded to a web lead at all. A lead nobody calls back looks like an ad that does not work.
Set up an instant text or WhatsApp reply and a notification to whoever calls back. We wrote a full guide on this: speed to lead and the first five minutes.
How long to test before you change things
- 1Give a new campaign at least 7 days before judging it, unless it is clearly broken (no clicks, tracking errors).
- 2Judge on leads, not clicks, once you have 30–50 conversion events.
- 3Change one thing at a time: page, offer, audience. Change everything and you learn nothing.
Facebook vs Google: different intent, different fixes
| Google Search | Facebook / Instagram | |
|---|---|---|
| Intent | High: they are searching for you now | Low to medium: you interrupt their scroll |
| Typical fix | Negative keywords, landing page match | Stronger offer, better creative, warmer audience |
| What kills results | Broad match without negatives | Optimizing for clicks or traffic |
Frequently asked questions
Usually one of three things: the campaign is optimized for traffic instead of leads, the landing page does not match the ad, or the Lead event is not tracked so Meta cannot learn who converts. Check tracking first.
Read the search terms report. Irrelevant searches, broad match keywords without negatives, and sending traffic to a homepage instead of a dedicated landing page are the most common causes.
The median across industries is 6.6% according to Unbounce (2024), with industry medians between 3.8% and 12.3%. Above 10% is strong for most local service businesses.
No. More budget buys more of the same clicks. Fix tracking, the page and follow-up first, then scale what converts.
Give a campaign about a week and enough budget to collect data. On Meta, an ad set needs around 50 optimization events in 7 days to exit the learning phase.
- Unbounce, Conversion Benchmark Report (2024)
- WordStream / LocaliQ, Facebook Ads Benchmarks (2025)
- WordStream / LocaliQ, Google Ads Benchmarks (2025)
- Google, mobile page speed and abandonment (via Marketing Dive)
- CXL, Reducing form fields
- Harvard Business Review, The Short Life of Online Sales Leads (2011)
