What is a lead worth to you?

Ad platforms report cost per lead because a lead is the last thing they can see. What you pay for is the customer at the end, and that is cost per lead divided by the share of leads that become paying jobs.

The gap usually opens at the first step. Spam calls, wrong services and people outside your area all count as leads, so a campaign that brings more of them reports a lower cost per lead while sending you fewer jobs. Put the share of leads that are real enquiries in as its own step and you can see it happen.

The last step is worth one job. Every earlier step is worth the step after it, times the share of people who get there. With the example numbers:

Estimate given = $2,400 × 45% = $1,080

Real enquiry = $1,080 × 60% = $648

Lead = $648 × 75% = $486

That $486 is revenue. What you can pay for a lead comes out of the profit on the job, not its price, so the ceiling uses your gross margin: a 40% margin leaves $960 on a $2,400 job, and $960 × 20.25% of leads that become jobs is $194.40. Pay more than that per lead and the ads cost more than the profit they bring in.

Breaking it into steps is not just tidier than one close rate. When cost per customer rises, the steps show where: fewer real enquiries points at the ads and keywords, fewer estimates at how calls are answered, fewer wins at price or follow-up.

If you upload what happens to leads after the click, Google Ads can bid toward the leads that become jobs instead of the ones that just fill in a form. These are the details that decide whether the values you send help it or mislead it.

Send the difference, not the full value. When every step counts as a conversion, a lead that reaches the third step has been reported three times. Sending each step as the increase over the one before keeps the running total equal to that step's value, and a won job adds up to exactly one job. The table under the calculator shows the amounts.

Lean on the second step. Google only accepts an offline conversion within 90 days of the click it belongs to. A job that takes months to close lands too late to teach the bidding anything, while a qualified enquiry or a booked estimate usually happens within days. A funnel with only two steps has no middle step, and reporting a won job's value for a lead that merely looked promising overstates it.

Send revenue, judge on margin. Google bids on whatever value you give it. If those values are revenue, the return to aim for is set by your margin: with a 40% margin you need back $2.50 of value for every $1 of ad spend to break even. The break-even ROAS calculator turns your margin into that target.

This funnel is the same one our client portal uses. A client names their steps and rates there, and when a lead is marked qualified or won on their calls and leads page, that step's value is uploaded to their Google Ads account as the increase over what was already sent. A won job carries the job's actual amount once it is entered.

Everything here is an average. Measure each rate over a stretch of time that has finished: a month that ended last week still has estimates waiting on an answer, which makes the win rate look worse than it is. If you sell a few very different jobs, work each one out on its own rather than blending them.

Gross margin here means what is left after the costs of doing the job itself: materials, the crew's time, subcontractors, fees. Rent, office staff and your own salary are overhead and stay out, or every lead will look unaffordable.

Repeat work and referrals are not counted. A customer who comes back every year is worth more than one job, which can justify paying above the break-even line on the first one. See customer lifetime value before you lean on that, because it only pays if the repeat work really happens.