Short answer
- Start by checking tracking. If conversions aren't measured correctly, you can't tell what's working.
- Then check, in order: offer → creative → landing page → audience → budget and learning.
- Judge ads by cost per lead or customer, not likes, reach or clicks.
- Give changes enough time. Constant edits can keep an ad set from ever settling.
Match the symptom to the cause
| What you see | Most likely cause | Look at |
|---|---|---|
| Lots of reach, few clicks | Creative or offer doesn't stop the scroll | Steps 2 and 3 |
| Lots of clicks, few leads | Landing page, or the ad promises something the page doesn't deliver | Step 4 |
| Leads, but poor quality | Offer too easy, form too short, or wrong optimization goal | Steps 2 and 6 |
| Leads in Ads Manager don't match your inbox | Tracking problems | Step 1 |
| Results swing wildly week to week | Too little data, too many edits | Step 6 |
| Ads won't spend at all | Audience too small, bid limits, or ad rejected | Steps 5 and 6, plus account quality |
Step 1: Tracking
Before you change anything, make sure you're measuring the right thing.
- Is the Meta Pixel installed on your site, and is the Conversions API set up if your platform supports it? Meta explains both in its Business Help Center.
- Are you tracking the action you care about (a booking, a form, a purchase), not just page views?
- Does the number of leads in Ads Manager roughly match what actually reached you?
- Is the campaign optimizing for that same action? A campaign optimized for clicks will find clickers, not buyers.
Step 2: The offer
The offer is the reason to act now. "We do great cleaning" isn't an offer. "First deep clean $99, booking 12 homes this month" is.
- Is it specific: a service, a price or value, and a reason to act soon?
- Is it the right size? Too weak and nobody moves; too generous and you attract people who'll never come back.
- Does it fit cold audiences? Strangers need lower-risk first steps (a trial, a quote, a consultation) than past customers.
Patterns for local offers are in Marketing campaign examples.
Step 3: Creative
For local businesses, real usually beats polished.
- Real photos and short videos of your work, your team, your space.
- The first second matters on video: show the result or the problem immediately.
- Readable text on the image, if any. Phone screens are small.
- Several variations, so Meta can find what resonates. One image for three months tires quickly.
Short vertical videos are often the strongest format. If you have clips on your phone, Video reels turns them into ads.
Step 4: Landing page
If people click but don't contact you, the page is the first suspect.
- Does it load fast on a phone? Slow pages lose people before they see anything.
- Does the headline repeat the ad's offer? A mismatch feels like bait and switch.
- Is there one clear action, like call, book or a short form, visible without scrolling?
- Is there proof near the button: reviews, photos, licenses?
Sending ad traffic to your homepage usually underperforms a page built for that offer. A dedicated landing page is $149 with me.
Step 5: Audience
Only now look at targeting.
- Location: are you reaching people who can actually come to you or whom you serve?
- Too narrow: stacking many interests can make audiences tiny and expensive. Meta's broader, automated audience options often work well when creative and tracking are strong.
- Overlap: several ad sets targeting the same people can compete with each other.
- Special Ad Categories: housing, employment, credit/financial and some other ads have targeting restrictions. If your business falls into one, it must be declared.
More detail in Meta ads targeting & bidding.
Step 6: Budget and the learning phase
When an ad set launches or changes significantly, Meta goes through a learning phase while it figures out who to show ads to. Meta's guidance is that ad sets need a certain volume of optimization events to exit learning; widely cited guidance puts it around 50 events in a week, and ad sets that don't reach it may show "Learning limited". Check Meta's current help pages for the exact wording.
- Too many ad sets splits a small budget so none gets enough data. Consolidate.
- Too many edits can restart learning. Batch your changes and wait.
- Optimizing for a rare event with a small budget means very little data. Sometimes optimizing for a more frequent, still meaningful action works better at first.
- Budget too low for your goal: if a lead costs more than your daily budget, the system has very little to learn from each day.
If you've checked everything
Sometimes the honest answer is that Facebook isn't the best channel for your business right now, for example when people only look for your service in an emergency and go straight to Google. Best advertising for small business compares options. If you'd like a second pair of eyes on your account, the free audit covers it, and Facebook & Instagram ads management is $250/month plus ad budget.