Why Meta Ads Fail For Home Service Businesses (And How To Fix It)
Why do Facebook leads seem so low quality?
Usually because they were contacted too late and too few times. A Meta lead did not go looking for you — they responded to something that caught their eye mid-scroll. That interest is real but fragile. Call within minutes and the lead behaves normally; call the next day and the same person will not remember filling anything in. This is the single most common finding when 9xBooked audits a home service ad account.
It is worth being precise about what "bad lead" usually means in practice. When we audit accounts, the complaint almost always resolves into one of four things, and only the last is genuinely a targeting problem.
- 1Nobody called for several hours because the team was on jobs.
- 2One call was attempted, it went to voicemail, and the lead was written off.
- 3The lead was real but not ready to buy this month, and there was no follow-up sequence to catch them later.
- 4The lead was genuinely wrong — outside the service area, or for work you do not do.
An automatic text that fires within seconds of the form submission, naming your company and saying someone will call shortly. It costs almost nothing, it keeps you in the homeowner's mind while your ad is still fresh, and it consistently moves contact rates more than any targeting change.
How many times should you follow up with a lead?
More than once, which is where most home service businesses stop. Five or six attempts spread over a week or two is a reasonable floor, mixing calls and texts. A large share of leads that eventually book are reached on an attempt after the first, so a one-and-done process throws away leads you have already paid for.
The arithmetic here is unforgiving. If you pay $40 a lead and contact 30% of them because you only call once, your effective cost per contacted lead is $133. Lift contact rate to 60% with a proper sequence and the same $40 lead now costs $67 to reach. You did not change the ad account at all.
It is also worth knowing how unusual fast follow-up actually is. When Harvard Business Review audited 2,241 US companies, only 37% responded to a lead within an hour, 24% took more than a day, and 23% never responded at all — the average response time among those who did reply within a month was 42 hours[1]. Simply being reliably quick puts you ahead of most of your competitors.
A follow-up sequence that works
- 1
Immediately
Automated text confirming the enquiry and setting the expectation that a human will call.
- 2
Within five minutes
First call attempt. If nobody answers, leave a voicemail that names the specific service they enquired about.
- 3
Same day
Second call at a different time of day, plus a short text offering to book a time that suits them.
- 4
Days 2 to 7
Two or three further attempts, varying the time. Then a final text making it easy to reply later without pressure.
Is your offer the actual problem?
Frequently, yes. "Contact us for a quote" asks a stranger to volunteer for a sales conversation with no idea what happens next. A specific, low-commitment first step — a free inspection, a same-day written quote — converts better because it tells the homeowner exactly what they are agreeing to and how long it takes.
| Weak | Stronger | Why it works |
|---|---|---|
| Contact us for a quote | Free 20-minute roof inspection with photos, same day | Concrete, time-bounded, and the homeowner gets something even if they do not buy |
| Best prices in town | Written quote within 24 hours or we tell you why not | Specific and verifiable rather than a claim everyone makes |
| Family owned since 1998 | See what we did for a house on your street | Proof beats positioning when the reader has never heard of you |
A useful test: could a competitor put their logo on your ad and run it unchanged? If yes, the ad is not really about you and there is no reason for the reader to pick you.
Why is my cost per lead so high?
Often because the budget is split across too many ad sets. Meta's delivery system needs a meaningful volume of conversions in one place before it can find people who convert cheaply. Six ad sets sharing a small budget means six campaigns all stuck in the learning phase, each too starved to ever get efficient.
- Consolidate into fewer ad sets so conversion signal concentrates rather than fragmenting.
- Run one broad audience rather than several narrow ones that overlap and bid against each other.
- Make sure the campaign is optimising for a conversion event that actually fires — a broken event means the system is optimising blind.
- Leave changes alone long enough to gather data. Editing daily restarts learning repeatedly.
Pausing campaigns when a week looks bad and restarting when cash allows is the single most expensive habit we see. Each restart discards accumulated learning and pays for it again. A smaller budget running continuously beats a larger one running intermittently.
Are you judging the campaign too early?
Probably, if you are looking at week one. The first two weeks are the delivery system testing who responds, and early numbers swing wildly in both directions. Cost per booked job only becomes a trustworthy figure around day 60. Owners who kill campaigns at day ten routinely kill ones that were about to work.
There is a related trap in the other direction: an unusually good first week convinces an owner to triple the budget immediately, delivery is pushed back into learning at the higher spend, performance drops, and the campaign gets blamed. Scale in steps and let each one stabilise.
Can your business actually handle more work?
This gets skipped and it should not. Generating more leads than your crews can quote, schedule, and complete does not produce revenue — it produces slow responses, missed appointments, and bad reviews. If you are already at capacity, the higher-return move is usually raising prices or close rates rather than buying more leads.
- How many additional jobs per month can you genuinely deliver without quality slipping?
- Who answers the phone when everyone is on site?
- Who sends quotes, and how quickly, when volume doubles?
- What breaks first if lead volume triples — scheduling, quoting, or delivery?
What if the numbers just do not work?
Sometimes they genuinely do not, and the honest answer is to stop. If your average job value is low, your margin is thin, and customers rarely return, there may be no cost per lead low enough to make paid acquisition profitable. Recognising that early saves months of spend and is a legitimate outcome of an audit.
When that is the case, the fix is usually upstream of marketing: raise prices, add a higher-value service line, or build a repeat revenue stream so the first job does not have to carry the whole acquisition cost on its own. Those changes make paid ads viable later.
Get an honest read on your campaigns
Book a free 30-minute audit call. We will look at what you are running, where it is leaking, and whether the underlying economics support paid ads at all.
Book a free audit callFrequently asked questions
How long should I run Facebook ads before deciding they don't work?
At least 60 days with a consistent budget and reasonable follow-up. The first two weeks are delivery learning, and the next few are cutting losing creative. Judging before day 30 usually means judging the learning phase rather than the campaign itself.
Why do my Facebook leads never answer the phone?
Almost always a speed and persistence problem rather than a lead quality problem. Interest decays within minutes of someone submitting a form. Send an immediate automated text, call within five minutes, and attempt contact five or six times over a week before writing anyone off.
Should I pause my ads during a slow cash month?
Reduce the daily budget rather than pausing. Pausing resets the delivery learning phase, so when you restart you pay again for data you already bought. Running a smaller amount continuously preserves the account's history and is cheaper over time.
Is it better to run ads myself or is the problem my agency?
Check the fundamentals before changing anything. Look at your average time to first contact, how many follow-up attempts you make, and whether your offer is specific. If those are weak, switching agencies will not fix the outcome — the leads will be handled the same way by the same team.
Sources
Every figure above links back to where it was published. Dates show when we last confirmed the source was live and still said what we quote.
- [1]The Short Life of Online Sales Leads — Harvard Business Review — Oldroyd, McElheran & Elkington. Verified August 9, 2026.Published 2011, analysing 1.25 million leads across 42 companies. One co-author was then chief executive of a company selling response-speed software, which is worth weighing. Frequently misquoted online as a five-minute rule with a 21x effect, and as an MIT study — it is none of those things.
About 9xBooked
9xBooked is a Meta ads agency for US home service businesses. We build and run Facebook and Instagram ad campaigns for roofing, HVAC, plumbing, landscaping, cleaning, and remodeling companies, delivering exclusive leads that are never resold or shared with competitors. Every engagement carries a written revenue guarantee, sized to your business on a free audit call before you commit to anything.
We publish these guides because most marketing advice aimed at contractors is written to sell rather than to inform. Every statistic here links back to where it was published, and we say so plainly when the data does not exist.
Want this run for your business?
Book a free 30-minute audit call. We’ll look at your service area, job values, and current lead flow, then tell you honestly whether Meta ads will work for you — and put a revenue guarantee in writing if they will.
Book your free audit call