Organic versus paid roofing leads.
Close rates of 50 to 70 percent against 8 to 15 percent, and what that does to cost per booked job.
The gap is not small.
Organic and referral leads close at 50 to 70 percent. Cold paid-ad leads close at 8 to 15 percent. Shared platform leads from Angi or HomeAdvisor cost $200 to $300, carry no exclusivity, and close at 8 to 20 percent. Lead costs across the industry run $60 to $220. Median job value is around $13,400. Every argument on this page is built from those five published figures and nothing else.
Why the spread is that wide.
Two variables explain almost all of it: intent and exclusivity. An organic enquiry has already chosen you, having read a page that answered the question they typed. A shared platform lead was sold simultaneously to three or four competitors, so you are one quote among several before the conversation begins. A cold paid click sits between them on intent and has no recommendation behind it at all.
Cost per lead is the wrong metric.
Cost per lead flatters whichever channel is quoting it. Divide by close rate and the order reverses: interrupted traffic converts at a fraction of the rate of an enquiry that arrived already convinced, so the cheap lead turns into the expensive job. The only figure that reaches your accounts is cost per booked job, and the guides below work it out channel by channel.
What that looks like against a median job.
At $13,400, a $367 acquisition cost is under 3 percent of revenue. A shared platform lead at $250 and a 14 percent close rate costs $1,786 per booked job, over 13 percent, which is most of the margin on a competitively priced replacement. Nothing about the work changed between those two figures. Only where the enquiry came from.
Volume is decided by mix, not by effort.
Work backwards: revenue target divided by job value gives jobs needed, divided by close rate gives enquiries needed. A $2 million target is roughly 150 jobs. At a 60 percent organic close rate that is 21 enquiries a month. At a 12 percent paid close rate the same target needs 104 a month. Two companies booking identical revenue, one of them running a call centre to do it.
Close rate is close to fixed. Cost is not.
A channel's close rate is a property of how the buyer arrived, and it barely moves with better sales training. What does move is the mix, and what the buyer saw before they made contact. A homeowner who read your claims page, checked your licence details, and looked at three of your projects in their suburb arrives at a different conversation than one who clicked an ad. That is the mechanism a rebuild is buying.
Where paid channels still earn their place.
Volume on a timetable. After a hail event, or when a crew comes free, nothing organic matches the speed of Local Services Ads or a shared-lead account. Used deliberately and measured at cost per booked job, they are a tool. The mistake is not running them. It is running only the channel that never gets cheaper and calling the monthly invoice a strategy.
The three-year comparison.
Run any channel out three years and the monthly figure stops mattering. Bought leads at trade-average rates cost a mid-size roofer six figures across that window and leave nothing owned at the end of it. A one-time rebuild with a small care plan costs a fraction of the same total, and the asset stays with you. That comparison is the whole commercial case, and the cost cluster works it scenario by scenario.
What to do with all of this.
Measure your own numbers per channel and per metro rather than blended, because a blended figure hides the one channel losing money. Then decide what proportion of your pipeline you want to own versus rent. The audit answers the first half of that question for your market at $2,500, and the pricing page carries the second half.
How to measure your own numbers honestly.
Four rules that stop the figures flattering you.
What belongs in the cost, and usually does not.
Platform and ad spend, obviously, but also management fees on that spend, staff time answering shared leads inside the minutes that decide them, software the channel requires, refunded or duplicate leads you paid for, and discounts given to win a job you were quoting against three other contractors on the same enquiry. Leave those out and every paid channel looks cheaper than it is.
The speed-to-lead trap.
Because a shared lead is sold to several contractors at once, the winner is usually whoever calls within minutes. That turns your office into a dispatch desk and rewards staffing rather than skill. Contractors closing at the top of the 8 to 20 percent band almost always have someone paid to answer instantly, which is a real cost that never appears in the lead price and never appears in the platform's case studies either.
Capacity is the constraint nobody models.
More enquiries than your crews can serve is not growth. It is a queue that becomes cancelled jobs and bad reviews. Before raising a lead target, check how many jobs a week your crews complete, how far out you are booked, and what happens to close rate when the earliest start date is six weeks away. Companies that fix scheduling first frequently discover they need fewer leads than they were about to buy.
Better enquiries beat more enquiries.
A page that publishes real price ranges lowers enquiry volume and raises close rate, because the people who were never going to sign filter themselves out before they call. Companies that publish nothing report higher volume and worse close rates, then mistake the volume for success. That trade, less volume for better fit, is the practical thing a rebuild buys and the reason the prices on this site are public.
The channel that closes highest and costs least.
Referrals. They sit in the same 50 to 70 percent band as organic, they cost almost nothing, and they improve with the quality of your work rather than the size of your budget. They also leak more than any other channel, because a referred homeowner checks you online before calling, and a site that does not confirm what the referrer said loses the enquiry before it is ever counted.
The three figures the argument rests on.
Published benchmarks from the sources listed under each figure.
Organic and referral close rate
Enquiries that arrive because the company was found or recommended.
Cold paid-ad close rate
Paid clicks with no prior relationship, across the same trade.
Shared platform lead
Angi and HomeAdvisor pricing, no exclusivity, 8 to 20 percent close.
SOURCES: GHOSTREP, GETBIDDABLE, WEBSITE AND SEO AGENCY. RANGES AS PUBLISHED SEPT 2026.
Every guide in this pillar.
Organic and referral at 50 to 70 percent, cold paid at 8 to 15, shared platforms at 8 to 20, and why the spread is that wide.
$200 to $300 per shared lead with no exclusivity, against enquiries that arrive only to you.
Where each genuinely wins, and why most established roofers end up running both.
Working backwards from a revenue target using median job value and a real close rate.
The only channel comparison that reaches your P and L, worked out with published figures.
Referral and organic enquiries close at 50 to 70 percent. Your site decides whether they convert.
Questions about this pillar.
Should we stop running ads?+
No. Ads buy volume on demand, which matters after a storm and when a crew comes free. The argument is against ads being the only channel, because their close rate never improves with time and their cost per job never falls.
How long before organic enquiries arrive?+
Weeks to months, depending on how competitive the market is and how much authority the existing site has. That is the honest sourced range, and anyone promising a date is guessing.
Are referrals really a channel?+
The highest-closing one in the published data and the cheapest. It is also the one most improved by a site that confirms what the referrer said about you, which is where most roofing companies lose free leads.
Do these close rates apply to commercial work?+
The direction holds, the numbers are residential. Commercial decisions take longer and involve more people, so close rates are lower across every channel and job values are higher.
How do I measure this without a CRM?+
A spreadsheet with source, date, and outcome per enquiry, kept for a year, is enough. Attribute by first contact rather than last click, or referrals disappear into direct traffic.
Run the numbers on your market.
$2,500, and you see cost per booked job for every channel you use.