CoachmarkSales and GTM
Lead source return on investment: which channel earns its keep and which one drains you
Most agents fund four or five lead sources and can name the return on exactly zero of them. One afternoon of honest math usually frees up half a budget.
September 14, 2026 · 5 min read
Count the things you currently pay for. A portal lead subscription. Facebook and Instagram ads. Maybe a listing website, a customer relationship management tool with a lead add-on, a farming mailer, a paid networking group. Now answer this: which one produced your last three closings?
Most agents cannot answer. Not because they are careless, but because nobody ever handed them a simple way to trace a commission check back to the thing that started it. So the subscriptions renew, the budget creeps, and the honest feeling underneath is that some of this money is being burned but you are not sure which part.
Why this is worth one afternoon
Every agent who does this exercise finds the same thing. One channel is quietly carrying the business. One or two are producing volume that never converts. And at least one is a line item that has produced nothing at all for a year and renews automatically every month.
That last one is the cheapest raise you will ever give yourself. It is money you already have, going somewhere it should not, and the only reason it keeps going there is that nobody has looked. One afternoon of arithmetic usually pays for itself several times over before the year ends.
You do not need a bigger marketing budget. You need to stop funding the half of it that has never sent you a closing.
The one number that matters: cost per closing
Forget cost per lead for this exercise. Cost per lead is how channels sell themselves, and it is exactly the number a bad channel looks best on. Cheap leads that never close are the most expensive product in real estate.
The number you want is what you spent on a channel over a period divided by the number of closings that came from it in that period. Use twelve months, not one, because a lead-to-close cycle is long enough that a short window will lie to you in both directions.
Then compare that against your average commission. If a channel costs you eight hundred dollars per closing and your average commission is eight thousand, that channel is a machine and you should feed it. If it costs six thousand per closing, you are working for the channel, not for yourself.
Count the time, not just the money
This is the step everyone skips, and it changes the answer more than any other. Some channels cost little cash and enormous hours. A referral group with a two hundred dollar annual fee and a weekly ninety-minute breakfast is not a two hundred dollar channel. It is two hundred dollars plus seventy-eight hours.
Pick a rough hourly value for your time. Take last year's gross income, divide by two thousand, round it. That is your number. Now add hours times that number to every channel's spend before you divide. Portal leads that need instant callbacks and thirty touches suddenly look very different from past client referrals that need one lunch.
- List every source you pay for, including free-but-time-expensive ones like open houses and networking.
- Twelve months of hard cost per source. Subscription fees, ad spend, printing, postage, dues.
- Estimate hours per month per source, then multiply by twelve and by your hourly number.
- Count closings from each source. Go through last year's closings one by one and assign each to where it truly originated.
- Divide total cost by closings. That is cost per closing. Compare it to your average commission.
Attribution honesty
When you assign closings to sources, you will hit a hard case. A past client referred someone, but that person had also seen your Facebook ad twice and read a neighborhood page you wrote. Who gets credit?
Give it to the source that produced the first real conversation. Not the last click, the first human contact. This is imperfect and it is fine. You are not building an attribution model for a corporation, you are making a decision about whether to renew a subscription. Consistency across the twelve months matters more than precision on any one deal.
One thing to watch for. Your sphere and past clients will almost always come out on top, because the cost is near zero. That is a real finding, not a rounding error. It is why a 12-month sphere touch plan tends to be the highest return activity on the whole sheet.
What to do with the answer
Do not blow up your whole lead generation setup in one week. Make three decisions and hold them for a quarter.
- Feed the winner. Take the channel with the lowest cost per closing and increase it by a meaningful amount, not a token one. Most agents underfund the thing that is already working.
- Fix or kill the middle. For a channel with volume but poor conversion, give it exactly one quarter with one specific fix, usually speed of follow up. If it does not move, cut it.
- Kill the zero. Any channel with no closings in twelve months gets cancelled today. Not evaluated, not optimized. Cancelled.
If the middle-of-the-pack channel is your Facebook and Instagram advertising, resist the urge to cut it before you check the funnel underneath it. Often the ad is fine and the follow up is the leak. Cost per appointment tells you which one you are dealing with, and your cost per lead is lying to you covers the same trap one layer down.
The worksheet
One page, one afternoon, once a year. Do it in the fall so the decisions land before you set next year's budget.
MY HOURLY NUMBER: last year gross / 2000 = $______
PERIOD: last 12 months AVG COMMISSION: $______
SOURCE CASH HRS/MO TIME COST CLOSINGS $/CLOSE
----------------------------------------------------------------
sphere/referrals ______ ______ _________ ________ _______
facebook + insta ______ ______ _________ ________ _______
portal leads ______ ______ _________ ________ _______
open houses ______ ______ _________ ________ _______
farming mailers ______ ______ _________ ________ _______
networking group ______ ______ _________ ________ _______
website / content ______ ______ _________ ________ _______
time cost = hrs/mo x 12 x hourly number
$/close = (cash + time cost) / closings
credit the FIRST real conversation, not the last click
DECIDE (hold for one quarter)
FEED lowest $/close -> ______________ (+____%)
FIX volume, no conversion -> ______________
one fix: ______________ judge on ____/____
KILL zero closings in 12 months -> ______________The reason this exercise stings a little is that it usually confirms something you already suspected and had been avoiding. That is the point. A budget built on hope renews itself forever. A budget built on cost per closing gets smaller and produces more.
Once the decisions are made, the work is holding them for the full quarter instead of drifting back. That is where Coachmark helps, keeping the fix you committed to in front of you daily and checking your real ad numbers against what you said you would do. Block two hours this week. Cancel the zero before you close the laptop.