What to actually look for in a Meta ads specialist (not what agencies want you to ask)

Post a question on Reddit about how to vet a Meta ads agency, and the top answers will come from people who run Meta ads agencies. That's not a conspiracy. It's just how forums work. The people most motivated to answer are the ones who benefit from shaping how you ask the question. So you walk away with a checklist built by the people being evaluated, which is roughly as useful as asking a car dealer what questions to ask at a car dealership.

The real problem is that most contractors who've already been burned aren't starting from zero. They're not trying to learn Meta ads from scratch. They're trying to figure out whether the next agency is actually different from the last one, or just better at sounding different. Those are not the same problem, and the standard list of interview questions doesn't address the second one at all.

Why the standard vetting advice fails burned contractors

The typical Reddit answer goes something like: ask for case studies, ask about their targeting strategy, ask how they report results, make sure they have experience in your niche. All reasonable things. None of them will protect you.

Here's why. An agency that's good at sales has case studies ready. They have a targeting answer ready. They have a reporting answer ready. The entire intake process for a sophisticated agency is optimized to answer the questions a suspicious prospect is likely to ask. If the interview questions are predictable, the answers will be too.

The tell isn't what they say. The tell is what they ask you to do before they've proven anything.

A contractor who got burned described it this way: they wanted him to commit to three months after spending $1,000, and in that first month he didn't see anything. By the time it was clear it wasn't working, the money was gone and a refund wasn't happening. That's not bad luck. That's the model working exactly as designed, because the contract exists specifically so the agency gets paid regardless of whether results show up.

The standard intake questions don't surface this, because they're focused on capability claims. Capability claims are easy. Risk structure is harder to fake.

The one structural question that filters out most bad agencies

Before you ask about targeting or creative or reporting cadence, ask this: what do you require from me before you've shown me a single lead?

A setup fee plus a three-month contract is one answer. That means the agency collects before they've demonstrated anything in your market, with your trade, for your type of customer. They are being paid to try, not to produce.

No setup fee and no contract is a different answer. It means the agency's incentive to produce results is continuous, because you can leave the moment it stops working. There's no lock-in to fall back on if the ads underperform.

This isn't just a pricing distinction. It's an incentive structure distinction. When an agency can only keep your business by continuing to generate leads, they have a reason to stay sharp that a locked-in retainer model doesn't create.

One contractor who eventually signed with ASN said it plainly: "Why I picked you guys is, A, there's no setup fee. If it's not working, I want to back out and not be at a loss." Another said: "That's why I chose you, because you said you could end any time, right?" Neither of them led with that question in the vetting stage. Both of them named it as the reason they actually moved forward. The structure told them something the pitch couldn't.

What proof actually looks like in this market

Case studies are standard. Specific, niche-matched case studies with real numbers are not.

There's a meaningful difference between an agency showing you a 10x ROAS slide from a client in an unrelated industry versus showing you results from a business that does what yours does. One of ASN's clients in rubber resurfacing (Safe Step) generated 247 leads at $11 per lead on $2,800 in total spend. That number means something to an epoxy contractor evaluating whether Meta ads will work for a similar floor coating service. It means almost nothing to a roofing company, and showing it to one would be a waste of both people's time.

The right question isn't "do you have case studies?" It's "do you have results from a business that does what I do, and can you show me the actual numbers?" If the answer is a vague reference to past clients or a percentage improvement without a baseline, that's a gap worth noting.

Generic proof doesn't just fail to persuade. On real sales calls, it actively backfires. A contractor evaluating ASN who ran a water delivery business rejected a lighting company result as irrelevant and walked away from the conversation faster than if no proof had been shown at all. Mismatched proof reads as either carelessness or an attempt to paper over an absent track record in the right niche.

What the follow-up setup tells you about operational quality

Most agencies stop at the lead. They run the ad, the lead comes in through a form, and then it's your problem. This is where a large percentage of otherwise workable campaigns fall apart, because a home service lead that doesn't get a response within a few minutes is usually a dead lead. The person filled out three forms, not just yours.

An agency that runs the ad and walks away is doing half the job. Before you sign anything, ask specifically what happens to a lead after it comes in through the form. Who contacts them? How fast? What does that message say?

This matters because the gap between a lead and a booked job is almost always a follow-up gap, not a targeting gap. Fast, relevant contact after a form fill is what converts a lead into an estimate request. Slow contact, or no contact until you check your phone an hour later, converts nothing.

ASN built an AI follow-up tool called Remi specifically to close this gap. It responds to incoming leads within seconds, holds a text conversation, handles basic objections, and books directly to a calendar. That's not a feature that exists to sound impressive in a pitch. It exists because slow follow-up is a documented, consistent reason that Meta ad leads don't turn into jobs, and building it into the model structurally is the only way to guarantee it doesn't get skipped when a contractor is busy on a job site.

How to actually use this when evaluating your next agency

Run the four-part check before you get into questions about targeting strategies or creative process:

First, ask what they require from you before they've produced a single lead. Setup fee plus contract means you're funding their chance to try. No setup fee, no contract means they're carrying the risk until results show up.

Second, ask for proof from a business in your specific trade, with actual numbers. CPL, total spend, lead count. Not ROAS alone, not vague improvement language.

Third, ask what happens to a lead after it submits a form. Specifically. If the answer is "we send you a notification," that's on you from that point. If there's a real follow-up sequence built in, ask to see it.

Fourth, ask what happens if it's not working after 30 days. The answer to that question tells you more about the agency's actual confidence in their own work than any capability claim they'll make in the first 20 minutes of the call.

What to do if you're ready to test again

ASN runs done-for-you Meta ad campaigns for home service contractors with no setup fee and no contract. The 30-lead guarantee is built into the standard offer, and the Remi AI follow-up system is included to make sure leads actually get contacted before they go cold. If you want to see what results look like for your specific trade before committing to anything, the right move is to reach out at americanservicenetwork.com and ask about current availability for your market.