How to Actually Measure ROI on Meta Ads

Q: Should I use Meta's reported ROAS or calculate my own?

Use your own. Meta's reported ROAS only counts events it can attribute (Pixel + CAPI). You'll see 2–3× higher numbers with your own calculation because you're counting phone calls, walk-ins, and referred customers that Meta never sees. Your own math is more honest — and usually more flattering.

Q: How long before I know if my ads are profitable?

Weekly spend + monthly ROAS review is the cadence. First full month gives you a directional read; month 2–3 gives you a stable one. If you're trying to judge profitability after 2 weeks, stop — the data isn't there yet.

Q: What if my jobs have long sales cycles?

Track with a 30-day / 60-day / 90-day booked window separately. Commercial work, full kitchen remodels, and anything $15K+ has a 45–90 day sales cycle. Attribute back to the lead date, not the booking date, so you're comparing ad spend from the right month.

Q: Can I just look at 'leads per dollar'?

Only if your close rate is constant and your job values are constant. It rarely is. Contractors who optimize purely on leads-per-dollar end up maximizing low-quality leads at the expense of real revenue.

If the only metric you watch is cost per lead, you'll make decisions that wreck your business. Here's the full ROI framework we use with 200+ contractors — and why staring at CPL misses the point.

The 4 Metrics That Actually Matter

Metric

What It Tells You

Target

Cost Per Lead (CPL)

How efficiently you're generating form submits

$10–$35 for home services

Lead → Booked Rate

How well your team converts leads to jobs

10–22% (varies by trade)

Cost Per Booked Job (CPBJ)

True acquisition cost

< 15% of avg. job value

Return On Ad Spend (ROAS)

Revenue ÷ ad spend

3–5× minimum

CPL alone is meaningless. A $15 lead that never closes is infinitely more expensive than a $40 lead that books a $10K job. Always pair CPL with close rate and job value.

The 'Cost Per Booked Job' Formula

This is the single most important metric. Calculation:

Cost Per Booked Job = (Monthly Ad Spend + Monthly Management Fees) ÷ Jobs Actually Booked That Month

If you spent $3,000 on ads + $1,500 on management this month and booked 30 jobs, your cost per booked job is $150. If those jobs averaged $3,000 revenue, you made $90,000 in revenue from $4,500 spent. Your ROI is 20x on the management + ad combo, or 30x on ad spend alone.

The Payback Period Question

For contractors with recurring revenue (maintenance plans, service agreements), customers are worth 2–5× the first job's value over their lifetime. Your 'real' cost per customer should account for this.

Trade

First Job

LTV (3-yr)

Multiplier

Roofing

$12,000

$12,500

1.04×

HVAC

$6,500

$18,000

2.77×

Pest Control

$300

$1,800

6.0×

Landscaping

$2,500

$8,000

3.2×

Plumbing

$800

$3,600

4.5×

For trades with recurring revenue, your break-even on a $200 cost-per-booked-job is way below the first invoice. You can afford to spend more aggressively on acquisition because lifetime value is higher.

What to Track Every Week

Total ad spend (Meta + any other platform)
Total leads generated (form submits only — not pageviews)
Total leads your team contacted within 1 hour
Total appointments / estimates scheduled
Total jobs actually booked (signed or deposit collected)
Total revenue from this week's booked jobs

How to Set Up Reporting Without a CRM

Google Sheet with 6 columns: Lead Date, Source (Meta / Google / LSA / Referral), Name, First-Contact Time, Booked (Y/N), Job Value. Fill it every day. At the end of the month, pivot by source. Simplest reporting system that tells you the truth.

When ROI Looks Bad — Diagnosis Sequence

If ROAS is under 3×, work through this checklist:

Is CPL too high? → Creative / audience / landing page problem
Is CPL fine but close rate < 10%? → Team / follow-up / qualification problem
Is close rate fine but job values dropping? → Offer / pricing problem
Is everything fine but ROAS still bad? → You have a math problem (look at margins, not revenue)

Most accounts don't need more leads — they need better attribution. 70% of contractors we audit can't tell you their real cost per booked job. Once they can, they stop making random budget-shift decisions.

Share

8 min read · Updated 2026-04-23

Frequent Questions. Short Answers.

Should I use Meta's reported ROAS or calculate my own?

Use your own. Meta's reported ROAS only counts events it can attribute (Pixel + CAPI). You'll see 2–3× higher numbers with your own calculation because you're counting phone calls, walk-ins, and referred customers that Meta never sees. Your own math is more honest — and usually more flattering.

How long before I know if my ads are profitable?

Weekly spend + monthly ROAS review is the cadence. First full month gives you a directional read; month 2–3 gives you a stable one. If you're trying to judge profitability after 2 weeks, stop — the data isn't there yet.

What if my jobs have long sales cycles?

Track with a 30-day / 60-day / 90-day booked window separately. Commercial work, full kitchen remodels, and anything $15K+ has a 45–90 day sales cycle. Attribute back to the lead date, not the booking date, so you're comparing ad spend from the right month.

Can I just look at 'leads per dollar'?

Only if your close rate is constant and your job values are constant. It rarely is. Contractors who optimize purely on leads-per-dollar end up maximizing low-quality leads at the expense of real revenue.