Google Ads for a cleaning company in Honolulu: 202 leads at a $21 blended CPL

Blended cost per lead, before → after — from the ad accounts.

Key facts

• Client: residential and commercial cleaning company in Honolulu, Hawaii (house cleaning, deep cleaning, move-out cleaning, commercial janitorial)
• Channels: Google Ads search, Performance Max, Google Local Services Ads
• Result: 202 leads in 10 working weeks on $4,272 total spend — a $21.15 blended cost per lead
• Trend: blended CPL fell from $30.71 in the first month to $16.98 by month three
• Performance Max delivered the cheapest leads in the account: $5.93–$8.36 per lead in its best weeks
• Starting point: unit economics were negative (−$15.75 per one-time job) — the strategy reversed that before scaling anything

The uncomfortable math we started with

The first thing we did was run the unit economics — and they were negative. At a $171 average ticket with a 30% margin, the business could afford at most $51.30 per customer. The actual cost at the time: $67 — every new one-time job lost $15.75. The business was surviving on repeat customers, and the worst channel (marketplace leads at ~$83, half of them junk) was quietly draining the budget.

We said that out loud instead of taking the budget and hoping. The strategy that followed came from the math, not from a channel preference:

• Cut the junk-lead marketplace spend to a minimal presence
• Refocus on recurring customers — selling a “first deep clean” as the entry product, because a recurring client breaks the negative one-time math
• Rebuild the paid search system so the cost per lead could physically fit under the ceiling

The trajectory below is what happened after: blended CPL fell from $31 to $17 — under the $51 ceiling with room to spare, turning a money-losing funnel into a profitable one.

The setup

Cleaning is a price-sensitive, high-competition niche in paid search. In Honolulu the competition includes both local operators and national franchises, so a small-budget account lives or dies on structure: exact service-plus-location queries, disciplined negative keywords, and conversion tracking wired to calls and forms — not clicks.

What we did

• Built the unit economics model first: average ticket, repeat-customer rate, and the maximum cost per lead the business could afford for a one-time deep clean versus a recurring contract.
• Ran one tight Google Ads search campaign for high-intent queries — “house cleaning Honolulu”, “move out cleaning Oahu” — instead of spreading budget across a dozen ad groups.
• Added Performance Max, which became the cheapest lead source in the account once it had conversion data to learn from.
• Layered in Google Local Services Ads in month three: pay-per-lead, Google Guaranteed badge, no wasted clicks.
• Pruned the search-terms report weekly. In cleaning, this is where accounts quietly bleed — one unfiltered “free cleaning samples” query family can eat a week’s budget.

The trajectory

Period	Spend	Leads	Blended CPL
Month 1 (baseline)	$891	29	$30.71
Month 2, week 1	$871	49	$17.77
Month 3, week 2	$400	17	$23.55
Month 3, week 3	$380	24	$15.81
Month 3, week 4	$459	27	$16.98

Ten working weeks in: 202 leads on $4,272 — and the cost per lead nearly halved while weekly volume held.

Why it matters

Most cleaning companies are told Google Ads “doesn’t work” at small budgets. It works fine — if someone actually reads the query report every week, feeds the algorithm real conversions instead of clicks, and adds channels (Performance Max, Local Services Ads) only after the core search economics hold. Discipline beats budget.

That discipline is the core of how we run Google Ads and Local Services Ads for small-budget home-service accounts. Request a growth audit to see the numbers for your business.