Most solar retailers I talk to can tell me their cost per lead down to the dollar. Ask them their cost per settled sale, split by channel, and the room goes quiet. They know the blended number, the one that averages the outsourced call centre, the doorknockers, and the two in-house reps into a single figure that looks fine on a spreadsheet and tells you almost nothing useful.
That blended average is where money quietly leaks. It hides the channel that is losing you money behind the channel that is carrying you. And if you can’t see which is which, you end up doing the worst possible thing: pouring more spend into the channel that feels busy rather than the one that actually pays.
This is a business problem before it is a software problem. But you can’t fix it without getting the data into one place. So let me walk through what cost per sale actually means for solar, why outsourced and in-house channels need separate maths, and what you need in your systems to see the truth.
What cost per sale actually means for solar
Cost per sale is not cost per lead, and it is not cost per quote. It is your total channel cost divided by the number of jobs that actually settled and got paid for. A lead that never books is not a sale. A quote that sits in the pipeline is not a sale. A signed contract that cancels in the cooling-off window is not a sale either.
For a solar channel, the total cost side of that equation has more moving parts than most operators account for. At minimum it includes:
- Lead or appointment cost (what you pay per lead, per booked appointment, or per marketing dollar spent).
- Commission paid to the setter, the closer, or both.
- Overhead allocated to that channel (salary, superannuation, tools, management time).
- Chargebacks and clawbacks on jobs that fell over after the commission was paid.
Divide the honest version of that total by settled sales, not signed contracts, and you get a number you can actually make decisions with. The gap between “signed” and “settled” is exactly where solar burns operators who don’t track it, because a solar sale can die for reasons that have nothing to do with the rep: finance falls through, the roof fails inspection, the customer changes their mind inside the statutory cooling-off period.
Why outsourced and in-house channels can’t share one formula
Here is the core of it. An outsourced channel and an in-house team carry cost in structurally different shapes, and if you use the same formula for both you will misread which one is profitable.
An outsourced call centre or doorknocking agency typically charges you a per-lead or per-appointment fee, often with a commission or success fee stacked on top. The cost is variable and it lands close to the sale. When you buy an appointment and it converts, you can trace the spend to the outcome fairly cleanly.
An in-house rep is the opposite. You are paying salary, super, vehicle, phone, and a share of your office and management overhead whether they close three jobs this month or thirteen. That cost is largely fixed and it does not map neatly onto any single sale. To get a real cost per sale for an in-house channel, you have to allocate that overhead across their settled jobs over a sensible period, not pretend it is a per-lead fee.
Try to force both into one blended line and the outsourced channel will usually look expensive per lead while the in-house team looks cheap, right up until you notice the in-house rep only settled four jobs and their fully loaded cost per sale is double the agency’s. You cannot see that unless you attribute cost separately by channel. This is the same discipline behind proper job costing for installers: a number is only useful once you know exactly what went into it.
Commission and clawbacks are a cost input, not an afterthought
Commission is where the maths gets slippery, because solar commission structures are rarely flat. You have setter and closer splits, per-watt (PPW) structures, tiered payouts, self-generated versus company-lead rates, and overrides for team leaders. Each of those changes the true cost of a sale, and a blended average buries all of it.
There is a compliance layer here too. If any of your reps are on commission-only arrangements, the payment structure still has to satisfy minimum-wage and piece-rate obligations, and how you classify a worker as an employee or contractor carries real legal weight (Fair Work Ombudsman, n.d.). Commission-only sales are legal in Australia, but they come with conditions that most operators underestimate, and getting the classification wrong is an expensive mistake (Zepys, 2026a). If you are structuring rep pay, it is worth understanding the legal considerations before you sign anyone up (Zepys, 2026b).
Clawbacks are the other half of the picture. When a job cancels after you have already paid commission, you are meant to claw that money back. In practice, if you are not tracking clawbacks at the rep and channel level, you are carrying the cost of dead deals as though they were sales. That inflates your settled-sales count, understates your real cost per sale, and hides which channel produces the most cancellations.
Chargebacks and cancellations behave differently by channel
Cancellations are not evenly distributed across your channels, and they are not handled the same way either. An outsourced agency working under a service-level agreement may carry some of the cancellation risk contractually, or it may not, depending entirely on how the deal is written. An in-house team’s cancellations land squarely on you.
High-pressure outsourced doorknocking, in particular, tends to generate more cooling-off cancellations, because a customer who felt pushed into signing on the doorstep is the same customer who reads the contract the next morning and rings to pull out. If your outsourced channel looks cheap per settled sale but generates a wall of cancellations, the cheap number is a mirage. You paid for appointments, paid commission, and then lost the jobs, and unless your SLA claws that back, you wore the lot.
The only way to see this is to track cancellation rate as a channel-level metric alongside cost per sale. A channel with a slightly higher cost per sale and a much lower cancellation rate is very often the better channel to scale.
A worked example (illustrative figures)
Let me put rough, illustrative numbers on it. These are made up to show the method, not real figures from any business.
Say an outsourced call-centre channel charges you $120 per booked appointment. It takes five appointments to settle one sale, so that is $600 in appointment cost per settled job. Add a $400 success commission and you are at $1,000 per settled sale, before you account for cancellations. If one in five of those settled jobs then cancels in cooling-off and the SLA does not refund you, your real cost per sale climbs closer to $1,250.
Now the in-house channel. A rep on $70,000 plus super, plus roughly $20,000 in vehicle, phone, and allocated overhead, costs you around $90,000 a year fully loaded. If that rep settles 90 jobs in the year, the overhead-and-salary cost per sale is about $1,000. Add their commission on top and you might land at $1,400 per settled sale, but with a lower cancellation rate because they are working warmer, referral-led leads.
On a blended average, both channels might report somewhere around $1,200 per sale and look interchangeable. Split them out and the picture changes: the outsourced channel is cheaper per job but leaks on cancellations, while the in-house channel costs more but converts more reliably. That is a real decision you can only make with the numbers separated.
The software categories that make this possible
None of this works if the data lives in three different places, which is exactly the problem the generic “best solar software” listicles never solve. The top-ranking guides for solar sales software mostly cover CRM, quoting, and design tools in broad strokes without ever defining cost-per-sale attribution (WorldMetrics, 2026a; SurgePV, 2026). To actually track cost per sale by channel, you need three capabilities working together:
- A CRM with channel-level attribution, so every lead is tagged by source and channel from the moment it lands, and that tag survives all the way to the settled sale.
- Commission tracking that handles setter or closer splits, PPW, tiers, and clawbacks at the rep level, not a spreadsheet someone rebuilds each pay run.
- A reporting or BI layer that can segment settled sales, cost, and cancellations by lead source and channel, so you get true cost per sale rather than a blended figure.
The reason most operators can’t produce this is not that they are lazy. It is that the data is genuinely scattered: the agency has its own reporting, commission lives in a spreadsheet, and the CRM tracks sales but not the full cost stack (WorldMetrics, 2026b). Nobody joins them up, so nobody sees channel-level cost per sale. This is the same fragmentation behind the three-tool problem that eats so many install businesses.
This is the gap CurrentFlow is being built to close. The idea is to bring lead source, commission, and sale outcome into one system, so a retailer running mixed outsourced and in-house teams can see true cost per sale by channel instead of a blended average that hides the leak. It is the tool I wanted when I was watching operators make scaling decisions on numbers that lied to them.
The operational risk of not tracking this
Here is what happens when you don’t split cost per sale by channel. Your outsourced channel feels busy, the appointments keep coming, so you decide to scale it. You put more spend in. Six months later your revenue is up and your margin is down, and you cannot work out why, because the extra jobs came with extra commission, extra lead cost, and a cancellation rate you were never watching.
You scaled the channel that felt productive instead of the one that was profitable. That is the single most common way I see mixed-model solar retailers erode their own margin, and it is entirely avoidable. Track cost per sale by channel, watch cancellations alongside it, and you will make the opposite decision: feed the channel that quietly settles jobs and holds them.
Get the number honest first. The software is just what lets you keep it honest at scale. If you want to be told when CurrentFlow is ready to pull your lead-source, commission, and sale-outcome data into one place, join the waitlist.
References
Fair Work Ombudsman. (n.d.). Piece rates and commission payments. https://www.fairwork.gov.au/pay-and-wages/minimum-wages/piece-rates-and-commission-payments
SurgePV. (2026). Solar CRM & sales software buyer’s guide (2026). https://www.surgepv.com/hub/solar-sales/sales-software
WorldMetrics. (2026a). Top 10 best solar sales software (2026 review). https://worldmetrics.org/best/solar-sales-software/
WorldMetrics. (2026b). Best solar business software: 2026 expert picks. https://worldmetrics.org/best/solar-business-software/
Zepys. (2026a). Is commission only sales legal in Australia? https://zepys.com/learn/is-commission-only-sales-legal-in-australia
Zepys. (2026b). Legal considerations for commission only sales in Australia. https://zepys.com/learn/legal-considerations-commission-only-sales-australia
FAQ
What is a good cost per sale for a solar business?
There is no single benchmark, because it depends on your system size, margin, and channel mix. The more useful question is not “what is a good number” but “is this channel’s cost per sale lower than the margin it produces, after cancellations”. A channel is only worth scaling if its true cost per settled sale, including clawbacks, leaves you enough margin. Comparing your channels against each other tells you far more than comparing against an industry average.
Why can’t I just use one blended cost per sale figure?
Because a blended average hides which channel is actually profitable. Outsourced channels carry variable per-lead and commission costs that sit close to the sale, while in-house teams carry mostly fixed salary and overhead that has to be allocated across their settled jobs. Blend them and you will usually misjudge which one to scale, which is how retailers end up feeding an underperforming channel.
How do chargebacks and cancellations affect cost per sale?
They inflate it, because you have spent money on a job that never settled. If you count signed contracts rather than settled sales, cancellations make your cost per sale look artificially low. Cancellations also vary by channel, with high-pressure outsourced doorknocking often generating more cooling-off cancellations, so tracking cancellation rate alongside cost per sale is essential to seeing the real picture.
What software do I need to track cost per sale by channel?
Three things working together: a CRM that tags every lead by source and channel and holds that tag through to the settled sale, commission tracking that handles splits, tiers, and clawbacks at the rep level, and a reporting layer that can segment cost, settled sales, and cancellations by channel. The problem for most operators is that these live in separate systems that nobody joins up.
Are commission-only pay structures legal for solar reps in Australia?
Commission-only arrangements are legal, but they still have to satisfy minimum-wage and piece-rate obligations, and the way you classify a worker as an employee or contractor carries real legal consequences. Before you put reps on commission-only, it is worth reviewing current Fair Work guidance and the legal considerations so you structure the arrangement correctly.
