Why Meta lead quality drops mid-campaign (and which part of your setup is responsible)

The first two weeks look good. Leads are coming in, the cost per lead feels manageable, and for a moment it seems like paid ads might actually work. Then something shifts. The leads get worse. More no-answers. More people who never heard of you, or who can't explain why they filled out the form. The cost per booked job climbs while the cost per lead stays flat, which is the more confusing version of this problem. By week five or six, the math that made sense at the start no longer does.

This is one of the most common patterns in Meta ads for home service businesses, and it almost never gets explained clearly by the agencies running the campaigns.

Why the first few weeks look better than they are

Meta's ad delivery algorithm starts every new campaign in what's called a learning phase. During this period, the platform is testing your ad against different segments of your audience, looking for signal on who actually responds. Early in that process, you tend to get the most motivated fraction of the available audience first. These are people who've been thinking about the problem your business solves, who are already close to a decision, who respond quickly.

The early results aren't fake, but they're also not typical. You're skimming the top of the pool. Once those people have seen and responded to your ad, Meta has to go broader or deeper into the audience to keep delivering at the same volume. The leads that come in after that shift are colder, less decided, and less likely to convert even if they fill out the same form.

An agency that doesn't tell you this isn't lying to you, but they're not doing their job either. The first two weeks of results are a baseline, not a benchmark.

The part of the setup that actually drives the drop

There are three places where mid-campaign lead quality degrades, and they're not equally responsible.

The first is audience targeting. If the audience is too broad from the start, the algorithm has too much room to find cheap clicks that don't mean anything. A roofing campaign targeting everyone within 40 miles aged 25 to 65 will produce a lot of form fills from people who saw the ad on their lunch break and half-remembered filling it out. Tighter audiences built around homeowner signals, specific income brackets, or behavioral data cost more per lead initially but degrade more slowly over time because there's less junk in the pool to begin with.

The second is the lead capture format. Instant Forms inside Meta are convenient, but they're frictionless by design. Someone can submit their name, phone number, and email in three taps without ever fully reading what they're signing up for. Some agencies use Instant Forms because they're fast to set up and produce high form-fill volume that looks good in a report. Landing pages outside of Meta require the lead to leave the platform, load a page, and make a more deliberate decision. That extra step cuts volume, but it also cuts the fraction of people who had no real intention of buying.

The third is follow-up speed, and this one doesn't affect lead quality directly. It affects conversion rate in a way that gets misread as a lead quality problem. A lead that comes in at 2pm on a Tuesday and gets called back at 10am the next morning is almost certainly going to be a harder conversation than one called back in under five minutes. The person has moved on mentally. They may have already called someone else. When enough of this happens, the leads start to look like low-quality leads, when the real issue is that real leads are going cold before anyone gets to them. One contractor on an ASN sales call put it plainly: he was getting people who seemed interested on the form but were completely unresponsive by the time he called. That's a follow-up problem wearing a lead quality mask.

What the math looks like when you separate the variables

The Safe Step campaign that ASN ran produced 247 leads at $11 per lead on $2,800 in total ad spend. That number is only meaningful if you also know the conversion rate downstream. A $11 CPL with a 5% close rate gives you a cost per booked job of $220. The same $11 CPL with a 1% close rate gives you $1,100 per booked job, which changes whether the campaign works entirely.

The agencies that report CPL and stop there are giving you one number when the actual problem is the relationship between two of them. If your CPL is holding steady but your close rate is dropping week over week, the issue is almost certainly follow-up speed or audience drift, not the ads themselves. If your CPL is climbing and your close rate is holding, that's an audience exhaustion problem and you need either a new creative or a refreshed audience.

These are diagnosable problems. But diagnosing them requires someone who is actually watching both numbers and is willing to tell you when the campaign is drifting before you've already lost four weeks of budget to it.

What to look for before you run another campaign

If you've been burned by an agency before and you're evaluating whether to try Meta ads again, the question to ask isn't "do you have a guarantee?" It's "what do you do when lead quality starts dropping in week four, and how will you tell me before I figure it out myself?"

An agency that can't answer that specifically, with reference to actual campaign mechanics rather than vague reassurances, is going to run the same pattern you've already lived through. Early results that look promising, a slow degradation, and a conversation at the end of the contract where they explain why the market was tough.

The right setup involves a landing page rather than an Instant Form for any trade where the job value is over a few hundred dollars, a follow-up system that responds to leads in minutes rather than the next morning, and someone actively monitoring the ratio between CPL and close rate rather than just reporting on the number that looks good. Those three things don't guarantee a campaign works forever. But they're what determines whether the drop-off you experience mid-campaign is a small correction or a full collapse.

Thinking about running Meta ads again

If this describes a campaign you've already lived through, or a situation you're trying to avoid the second time around, ASN builds and manages Meta campaigns for home service contractors with no setup fee and no contract. The structure is designed so you can see whether the leads are converting before you've committed to anything. The contact page is the right place to start if you want to see what a campaign built around your specific trade would actually look like.