A truck, a trailer, a logo and a business card. That was the whole company when they came to me. No website to redesign, no content to improve, and no way for anyone searching for junk removal in Southwest Florida to discover they existed at all.
What existed, and what did not
The equipment was real and the owner knew the work. Everything a customer would ever encounter had to be built.
- The website: structure, copy and pages.
- The content: photography, video and writing. None of it existed beforehand.
- The Facebook page, the Instagram account and the Yelp listing, created and filled rather than optimized.
- Tracking, wired in while the site was being built rather than added later.
- A job estimator, built around real disposal facilities, real drive times and a real crew. It is embedded further down this page.
That is a wider brief than most agencies take, and it is the reason I keep the client list short. Work at this depth does not survive being spread across fifteen accounts.
The first problem was pricing, not marketing
A new trade business has no reliable sense of what a job costs it. Accounting software will not answer the question either, because it records what was invoiced rather than what the work consumed. Quoting therefore runs on instinct: some jobs feel profitable, some feel thin, and nothing distinguishes them afterwards.
Fixing that early matters more than fixing it later. The prices set in the first few months become the prices customers expect, and a business that undercharges from day one spends years unwinding it. So the costing came before any spend on demand.
The estimator therefore prices from a target margin, cost divided by (1 − margin), instead of multiplying by a markup. It applies a 1.30× labor burden by default, since a $25/hour employee does not cost $25 an hour once payroll taxes, insurance and unbilled time are included. Results are scored against published industry benchmarks, so a figure that looks acceptable can still be flagged.
The tool itself
What follows is the working estimator, not a demonstration version. Their branding, their formulas, their benchmark panel, and a full methodology write-up with sources at the bottom. Put your own numbers through it.
It is badged Internal for a reason. Two versions exist: a simplified one on their own site that quotes a job without exposing cost or margin, and this one, the operator's back end, where every input is visible and the profit sits on screen. I built both, which is why I am comfortable showing the real one here.
It also keeps notes to itself about which inputs are still missing and which defaults could not be checked against a local source. Those have been left in place. A costing tool that conceals its own assumptions is how businesses end up quoting from numbers nobody ever verified.
None of the figures loaded into it are their pricing. Every default is a published industry benchmark, cited inside the tool.
Three numbers, in sequence
The estimator answers the first question: what a job costs, and what it should therefore be sold for. Two more have to be answered before advertising money is safe to spend, and the order is not optional, because each one depends on the answer above it.
Most of the accounts handed to me for repair were spending without these answers. Usually nobody had told the owner the questions existed. Both tools below are the ones I use.
Two · What a lead is worth
A cost per lead means nothing on its own. Thirty dollars is cheap for one business and ruinous for another, and the difference comes entirely from what a job leaves behind once it is finished.
The calculation is job value × margin × close rate. That produces the break-even: the point at which a lead has paid for itself and returned nothing further.
Question two
What can you afford to pay for a lead?
A lead is worth
$120
Break-even. Below this figure the work makes money, whatever the budget happens to be.
Comfortable target
under $40
Roughly a third of break-even, which is a 3× return on ad spend.
Three · What a budget will buy
Budget questions only become meaningful once the value of a lead is known. This one gives the volume to expect, and how quickly the platform will accumulate enough data to improve.
Question three
What will your budget actually buy?
Leads a month
43
Real volume at a workable cost. That is a business, not a science experiment.
How fast Meta learns
~10 conversions per ad set, per week
Below Meta's 50-a-week ideal, so expect a longer, noisier learning period. That is normal at local budgets.
Raise the ad-set count above and watch the second figure fall. Lead volume stays where it was, but each ad set now sees a fraction of the data, and the rate at which the system learns collapses with it. Splitting a small budget four ways is the most common self-inflicted wound in local advertising.
Where it stands
This is an early-stage account. The site is live, the profiles are populated, the tracking is running, and the estimator is in daily use. Eighteen leads so far, four of them closed.
I am not going to draw a growth curve through a sample that size. What can be said is that the order was right: establish what a job leaves behind, calculate what a lead is therefore worth, and only then decide what one can be bought for. It is unglamorous work, and it is the difference between scaling a business and spending faster.
The tool carries their branding rather than mine because it belongs to them. Software built around a specific business is worth considerably more to that business than a generic version with someone else's logo on it, and it is the clearest evidence I can offer that the work is genuinely built to order.
FAQ
Questions about this build
Why build a costing tool instead of just running ads?
Without it there is no way to judge the ads. A $60 lead is cheap for one business and a slow bleed for another, and the difference comes from what a job leaves behind once it is done. Spending first and measuring afterwards is how accounts get shut down in month three for reasons nobody can identify.
How long did the estimator take to build?
An afternoon. That was only possible because the system I work in already held their pricing, their service area, their disposal facilities and what those charge. I was not reconstructing the business from memory while also trying to get the arithmetic right, which is usually the point at which these projects are abandoned half-finished.
Are the default numbers in it real?
They are published industry benchmarks with sources cited in the tool, not this client's rates. Every default is replaced with the business's own figure as soon as it is available. They exist so the tool is useful on day one, not so that anyone relies on them.
Do I get to keep a tool like this?
Yes. Tools, tracking, dashboards, ad accounts and their history stay in your name. If we stop working together you keep all of it, including everything the account has learned.
Is this only useful for junk removal?
The structure applies to any business that quotes jobs: labor with a real burden on it, a variable material or disposal cost, drive time, and overhead to carry. Landscaping, moving, restoration, cleaning. The inputs change. The fact that markup is not margin does not.
Do you build the social profiles too, or only advise on them?
Build them. Here the Facebook page, the Instagram account and the Yelp listing did not exist, so they were created and filled with content that also did not exist yet. For a business starting from nothing, being told to sort its profiles out is homework, not help.
Something here sound like your situation?
Tell me what is going on and I will tell you honestly where I would look first.
Let's Talk