The most common reason local businesses conclude Meta ads do not work is that they measured them like search ads. Those two channels do different jobs and the same yardstick makes one of them look broken.
Search captures demand, Meta creates it
Someone searching for an emergency plumber has already decided they need one. The ad is competing for an existing intent that will be satisfied within the hour.
Nobody on Instagram was looking for a plumber. Meta's job is to put an offer in front of people who were not thinking about it, which means the intent has to be manufactured and the timeline is longer.
Consequences that follow directly:
Cost per lead is usually lower, close rate is usually lower. Both are expected. A blended comparison against search flatters Meta on cost and punishes it on quality, and neither number means much on its own.
The window is longer. A Meta lead frequently closes weeks later. A 30-day report understates it.
Follow-up matters more. A search lead sometimes books itself. A Meta lead almost never does.
Where Meta genuinely works for local services
Planned, discretionary, and seasonal work. Anything the customer could do this year or next. Roof replacement, remodels, landscaping, maintenance plans, pool openings. Search only reaches these people at the exact moment they decide; Meta reaches them before.
Visually demonstrable results. Before and after content, where the transformation is the argument. This performs in a way that a text ad cannot.
Offers with a reason to act. Seasonal pricing, capacity-limited slots, a genuine promotion. Not a fake countdown — an actual reason.
Retargeting. People who visited your site and did not convert are the cheapest and highest-converting audience on the platform, and a surprising number of businesses never turn it on.
Reaching your existing list. Uploading your customer list to build audiences from it is one of the more reliable uses of the platform and one of the least used.
Where it usually does not
True emergency work. Nobody sees an ad and decides their heating is broken. Search owns that.
Very narrow service areas. Below a certain audience size the delivery becomes inefficient and costs climb.
Highly regulated categories. Health and medical adjacent advertising runs into policy on personal attributes, and enforcement takes accounts down rather than just posts.
Creative is the variable that matters
On search, the keyword does most of the targeting work. On Meta, the creative does. The platform will find people who respond to what you showed them, which means the creative is the targeting.
Practical implications: test more creative than you think is necessary, expect fatigue on a timescale of weeks rather than months, and treat video and native-looking content as the default rather than polished brand assets. A phone-shot job walkthrough routinely outperforms a produced ad in this category.
How to measure it fairly
Track Meta leads as their own cohort, with their own close window, and report them separately from search. Look at contribution over sixty to ninety days rather than thirty.
Then compare cost per booked job — not cost per lead — across both channels. Frequently Meta looks worse on lead cost and quality, better on volume and reach, and roughly comparable on cost per booked job once the follow-up is doing its work. That is the comparison that should drive the budget split.