Is a website worth it for a small business? That usually gets answered like a matter of taste, somewhere between “everybody has one” and “mine never did a thing for me.” It’s neither. It’s arithmetic. The whole article in one paragraph: a $2,400 Launch site plus a $300/month care plan costs $6,000 in year one. If your average job is $6,000 and you keep 35% after materials and labor, the site has to bring in three jobs before it has paid for itself. Three. In a year.
Three jobs might be nothing to you. It might be a stretch. Either way, the number only means something built out of your inputs. So let’s build it.
The formula, in one line
Break-even customers = total first-year cost ÷ (average sale × your margin).
That’s how many new customers the site has to produce in twelve months before it stops being an expense. Cost comes off your quote, average sale off your last twenty invoices. Margin is where people quietly fudge things, so it gets its own section.
The two places owners get the inputs wrong
What you actually keep
A $6,000 install is not $6,000 of value to you. Subtract materials, subcontractors, the crew’s hours, the fuel. What’s left is what pays for the website. The invoice total is the easy number to grab. The margin is the one the formula wants.
How many inquiries turn into jobs
Ten form fills is not ten jobs. Close one of every three inquiries and you need nine of them to land three jobs. Under one a month. Much less intimidating than “we need to generate leads,” and it’s the same sentence.
Four worked examples with real prices
Nobody below is a real client. These are business shapes with round numbers so you can swap yours in. The cost side is real: our published prices, with the rest on the pricing calculator.
The trades business at a $6,000 average install
First-year cost: $2,400 build plus $3,600 of care equals $6,000. Average job $6,000 at a 35% margin is $2,100 of gross profit each.
$6,000 ÷ $2,100 = three jobs in the first year. One every four months, and the fourth is profit. At a one-in-three close rate, nine inquiries over twelve months.
The salon at a $75 ticket
Small tickets change the shape completely. The unit to count is a new client, not the haircut that brought them in.
Say a first visit is $75, 40% of new clients come back, and the ones who do visit three more times that year. A new client is worth 1 + (0.4 × 3) = 2.2 visits, or $165. Keep 50% after product and the stylist’s cut: $82.50 to you.
$6,000 ÷ $82.50 = 73 new clients in year one. About six a month.
Two ways to move that, and they cost different things. Push repeat visits to 50% and the bar drops to 64 clients. Raise the first visit to $85 and you land in the same place. One is a price conversation with every new client. The other is a follow-up text.
The B2B consultant at $4,000 an engagement
First-year cost $6,000 again. Engagement $4,000, and after tools and subcontracted help you keep 60%, so $2,400 each.
$6,000 ÷ $2,400 = 2.5, so call it three engagements. One wrinkle: your product is your own hours, so the constraint eventually becomes your calendar. The site’s job shifts from “find me work” to “find me better work,” and you watch average engagement value.
The online store at a $60 average order
This one surprises people. An online store starts at $8,500, not $2,400, so year one at the entry care plan is $12,100. Average order $60, and after product cost, processing and shipping you keep 30%: $18 an order.
$12,100 ÷ $18 = 673 orders in the first year. Roughly 56 a month. About two a day.
Two a day sounds modest until you notice nothing guarantees them. Store break-even is harder, because $18 has to be earned hundreds of times instead of three. It’s also why a store costs more to build: hundreds of small conversions need product pages, checkout and speed all working.
All four, side by side
| Business | Value per new customer | Margin kept | Gross profit each | Year-one cost | Customers needed |
|---|---|---|---|---|---|
| Trades install | $6,000 | 35% | $2,100 | $6,000 | 3 (one every four months) |
| Salon | $165 (2.2 visits) | 50% | $82.50 | $6,000 | 73 (about 6 per month) |
| B2B consultant | $4,000 | 60% | $2,400 | $6,000 | 3 per year |
| Online store | $60 order | 30% | $18 | $12,100 | 673 orders (about 2 a day) |
Same formula, four wildly different bars. Anyone quoting one universal “websites deliver X% ROI” figure is averaging these into a number that describes none of them.
How many months until the site is free?
Payback period = build cost ÷ monthly gross profit from the site.
At a job a quarter, the trades business earns $700 a month in gross profit, so the $2,400 build is paid back in month four and the next two jobs cover the year’s care. Year two resets in your favor, because there’s no build to pay for: trades needs two jobs, the store 200 orders.
The line you don’t get to cut
When money gets tight, the first thing owners reach for is the care plan. Every time. It’s the only line that recurs, so it looks like the flexible one.
It isn’t. A website is software on the open internet, and software rots on a schedule: core updates, plugins, PHP versions, SSL, backups, the bots hammering your login page tonight. Cancel and nothing happens for a while, which is the dangerous part. Then the form quietly stops delivering mail, the site goes down in your busy season, or it gets defaced and Google flags it. Any of those sets your return to zero while the $2,400 stays spent, and recovery costs more than a year of care. Maintenance belongs in the never-cut column, next to insurance.
The four numbers to track so next year isn’t a guess
Everything above is a projection. These four turn it into a measurement, and you need all four:
- Form fills, by form. Not “contact submissions.” Which form, on which page. A quote request and a newsletter signup are not the same event.
- Tracked calls. If the phone is your main channel and it isn’t tracked, you’re guessing on your biggest input. At minimum, ask where people found you.
- Close rate. Of the inquiries the site sent, how many became customers. Analytics never sees that one; it lives in your invoicing software.
- Average sale and margin. Recalculate yearly. Both drift.
Set those up before launch, not eleven months later when you’re deciding whether to renew. Without them the site’s contribution is invisible, which is how good websites get blamed for underperforming. Wiring tracking during the build costs a fraction of retrofitting it, and it’s part of what we do rather than something you’re Googling at 11pm.
The line nobody puts in the spreadsheet
Your hours.
A website isn’t one task. It’s forty small ones: copy that sounds like you, photos that don’t look stock, a structure a stranger can navigate in four seconds, forms that reach the right inbox, speed on a phone with two bars, accessibility, redirects, tracking, the launch checklist. Each one is an evening. At $150 an hour, thirty of your own hours is $4,500 of billable capacity spent dodging a $2,400 invoice, before the month-three rework when you find the quote form was landing in spam.
No customers yet, still testing whether the idea has legs? A rough placeholder is fine for a minute. That minute ends the day somebody pays you. After that it’s your storefront, and every visitor it fails to convert is a job that went to whoever looked like a real company. That’s the moment to get us involved.
What a website can’t do on its own
A website converts demand. It does not conjure it.
The math above assumes somebody is already looking: search, a referral, a truck wrap, the sign out front. If nobody knows your name yet, the site can’t be the only thing you do this year. It still has to exist first, because spending on attention with nowhere to send it is the real waste. Tell us where your work comes from and we build for that channel.
The flip side goes uncounted. Referred customers look you up before they call, size you up in seconds, and never fill out a form. Keep those out of the break-even math, but know the site is working. We publish our project work for that reason: the “do these people look real” test gets run on you constantly.
What if you’re already booked solid?
Then capacity changes what you’re buying. Not whether to buy.
A new site won’t invent hours you don’t have. True. It changes the mix instead: trade $2,000 jobs for $6,000 ones, hold your price because you’ve stopped looking like the cheap option, own an asset a buyer will pay for instead of a Facebook page. So run the formula on the upgrade. If a better small business site lifts your average job from $4,000 to $5,500, one job a quarter covers the whole year.
Common questions
How many leads should a website generate for a small business?
Enough to cover your break-even number at your close rate. Three jobs at one-in-three means nine inquiries a year. A salon needing 73 clients needs a much bigger stream, which is why those two sites get built differently.
What’s a reasonable payback period?
For a service business with jobs in the thousands, the build often pays back inside the first few closed jobs. For a store selling $60 orders, realistically most of year one. Longer than two years and the question becomes whether you have a website problem or a traffic problem. That’s a diagnosis, not a verdict. Send the numbers and we’ll tell you which.
Does a cheaper website change the math?
It lowers the top of the fraction. It usually lowers the bottom too, and that’s the part nobody prices in. Cheap builds skip the things that produce inquiries: a clear offer above the fold, forms that actually deliver mail, a tappable phone number, proof you’ve done the work, load times that don’t shed half your mobile visitors. A $500 site that converts nobody has an infinite payback period.
What about the AI website builders?
Worth answering straight. If you’ve never touched one: there are tools that take a description of your business and produce a whole site layout in an hour. They’re real, they work, and we use that class of tool inside our own process. The honest limit is that what comes out is a starting point, not a finished business asset. It doesn’t know your close rate, can’t write the offer you say out loud on estimates, doesn’t wire forms to your inbox, doesn’t set up the four numbers above, and doesn’t answer the phone in October when something breaks. And if you already tried one and it looked fine in preview but wrong the second it went live – spacing off, copy that sounded like everyone else’s, a form you weren’t sure actually sent – you found the gap exactly where it is. That gap isn’t a prompt you haven’t cracked. It’s the job.
Run your numbers, then let’s build the thing that has to hit them
Every version of this question collapses into one division problem. Pull your last twenty invoices, average them, apply your honest margin, divide first-year cost by what’s left. Ten minutes with a calculator and you’ll know your break-even number better than most people quoting you.
Then the half that decides it: that number only gets hit if the site is built to convert. That part is ours. Prices and the calculator sit on the pricing page, so go get your number. Then send it over with what you sell and I’ll tell you what it takes to clear it, before anybody signs anything. Bring the invoices. This is my favorite part.