Why Reddit gives you 10 different answers about Meta ad performance (and what actually matters for a local service business)

Post a question about Meta ad performance on any contractor subreddit and you'll get ten answers that contradict each other. Someone running e-commerce will tell you $5 CPL is realistic. A solar company will say expect $80. A roofer will say anything under $150 is good. Someone else will say they shut off Meta entirely and went back to Google. Nobody is lying. They're just answering a different question than the one you asked.

This is the problem with pulling benchmarks from Reddit when you run a local service business. The numbers that matter for your situation are nothing like the numbers that matter for someone selling a $2,000 online course or a $30,000 solar installation. Budget, trade, geography, and job value all pull in different directions at once. Until you understand which variables actually control the outcome at sub-$100 daily spend, the Reddit thread isn't useful. It's noise.

Why the same ad platform produces such different results

Meta's ad auction isn't one marketplace. It's a local auction. What you pay per impression, per click, and per lead in Calgary is priced separately from what someone in Phoenix pays. A painter in a suburb of 80,000 people is competing in a completely different pool than a painting franchise running national brand campaigns.

That local pricing reality means any CPL benchmark pulled from a national or global sample is already suspect. A $20 CPL is common in some markets and nearly impossible in others, not because of campaign quality but because of the raw cost of reaching people in that ZIP code.

The second variable is trade specificity. Detailing leads cost less to generate than roofing leads because the urgency and purchase cycle are different. Someone booking a detail does it casually, fast, often on impulse. A roofing job involves multiple quotes, a lot of anxiety, and a much longer decision timeline. The ad can generate the lead either way, but the conversion process after the click is completely different, which affects what a good CPL actually means in terms of revenue.

A real number from ASN's client data shows what niche-specific targeting can produce when it's built around a specific trade: Safe Step, a rubber resurfacing company, ran $2,800 in total ad spend and generated 247 leads at $11 CPL. That result is not portable to every trade or geography. It's the product of a campaign built for that trade, in those markets, at that budget level. Which is exactly why posting it on Reddit would generate three replies telling you that's not realistic and two more asking what industry that's from.

What actually controls performance below $100 per day

At $100 per day or less, you're working with somewhere between $2,800 and $3,000 per month in ad spend. That's a real budget, but it's a budget that punishes certain structural mistakes immediately.

The first is geographic over-targeting. When a previous agency gave Sabro, an auto detailer, leads that were 1.5 to 1.75 hours away, those leads weren't useless because of bad creative. They were useless because the targeting radius was calibrated to maximize lead volume rather than serviceable radius. At low budgets, a wide radius drains spend generating contacts you can't convert without subcontracting, which erodes both margin and your ability to measure whether the campaign actually works.

The second is follow-up speed. A lead that sits untouched for four hours on a local service job is a lead that has already called two other contractors. This isn't theoretical. It's the mechanic behind why Meta leads have a reputation for being low quality in contractor communities. The leads are real. The follow-up gap is where they die. At low budgets especially, you can't afford to lose 30% of leads to slow response and still make the math work.

Third is job value relative to CPL. A $100 CPL is a terrible result for a $200 detail. It's a fine result for a $1,400 epoxy floor install, and it's a genuinely good result for a $6,000 exterior paint job. Any Reddit thread that says "$100 CPL is too high" is implicitly comparing your job value to someone else's, which may have nothing to do with your actual business.

The math Harjap, an auto detailer, ran live on a sales call was: "$100 cost-per-booking against a 3-4 hour job." That's a specific calculation. If your average ticket covers that CPL with a reasonable margin, the campaign works. If it doesn't, you need either lower CPL or higher ticket, not necessarily a different platform.

Why benchmarks from agencies are often just as misleading

The problem isn't only Reddit. Agencies often cite ROAS figures that have the same portability problem. A 10x ROAS result from a painter in one market tells you almost nothing about what you'll get as a roofer in a different city. The number is real, but the context that produced it is invisible in the headline.

What makes a benchmark actually useful is specificity on three things: trade, geography, and budget range. Without all three, you're comparing your situation to someone else's in ways that will either inflate your expectations or talk you out of a campaign that would work fine for your business.

The Safe Step result (247 leads, $11 CPL, $2,800 spend) is specific enough to be useful if you're in resurfacing, epoxy, or a trade with a similar purchase profile. It's not useful for a roofer in a rural market where the addressable audience is a fraction of the size.

Nabil, a painter, achieved 10x ROAS. Yerim, in lighting, achieved 22x ROAS. Those are different trades with different cost structures and different sales cycles. The common thread isn't the platform. It's that the campaigns were built around the specific trade and tested without a multi-month lock-in that made it impossible to course-correct early.

What to actually look for before running your next campaign

If you've been burned by a previous agency and you're evaluating whether to try Meta again, the benchmarks matter less than the structure of the test. A campaign with a tight geographic radius matched to your actual service area, follow-up built to respond within minutes rather than hours, and a CPL target calculated from your real job value and close rate will tell you something real within 30 days. A campaign structured around what some Reddit thread said a good CPL looks like will tell you nothing useful.

The question worth asking any agency before you start is not "what CPL can you guarantee" but "how did you set the geographic radius, and how does lead follow-up work the moment someone opts in." Those two answers will tell you more about whether the campaign is built for your business or built to generate volume numbers that look good on a report.

ASN builds campaigns with that structure, including an AI follow-up tool that responds to new leads within seconds and handles the back-and-forth until the job is booked. If you want to see what that looks like for your trade specifically before committing to anything, the right next step is to reach out at americanservicenetwork.com and ask about results in your niche.