The 10 Marketing Numbers Every Business Owner Should Know by Heart

Ask most business owners what their marketing costs and you get a monthly figure. Ask what a customer costs and the room goes quiet.

That gap is expensive. Without it, every decision becomes a feeling. You cut a channel that was working. You keep one that never was. You judge a whole year on whether the phone felt busy.

So this guide fixes that. Ten numbers, in plain language, with the arithmetic written out. Real 2026 benchmarks so you can see where you sit. And a calculator further down that takes six figures you already have and produces the rest.

None of it requires software you do not own. Most of it can be done on the back of an envelope.

Why industry averages lie to you

Start here, because it saves you from the most common mistake.

You have probably seen "the average cost per lead is $198" quoted somewhere. It is a comfortable number. It is also traceable to a survey from 2017, and it has been recycled ever since without anyone rechecking it.

The current picture is both more useful and more humbling. Blended cross-industry cost per lead sat around $214 in 2026, up from roughly $198 the year before. But that average conceals an eleven-fold spread.

Now look at what those numbers mean rather than what they say.

A law firm paying $650 per lead is operating normally, because a single client might be worth $50,000. An e-commerce brand paying the same has a serious problem. Same number, opposite verdict.

So the only benchmark worth chasing is your own break-even. We will calculate it shortly, and it takes about ten seconds once you have three figures.

The chain: how the numbers connect

Here is the single most useful idea in this article. These metrics are not a scorecard. They are a chain, and each link changes the meaning of the one before it.

Take a real dataset. The SearchLight Home Services benchmark tracked $6.72 million in Google Local Service Ads spend across 888 contractors and 126,650 leads in February 2026. Follow one lead through it.

Read that left to right and something clicks. The headline number — $53 — is not the number that matters. It becomes $233 once you account for the leads that never booked. That is a 4.4x difference between what most owners quote and what a customer actually costs.

Then keep going. A $1,826 job at a 25% margin produces about $457 in gross profit. Take away the $233 acquisition cost and you have roughly $224 of profit on the first job. That is the real answer, and you cannot get to it without walking the whole chain.

Now change one link. Improve the book rate from 43.9% to 55% — better phone answering, faster response, nothing else — and the cost per customer falls to about $96. Same ads. Same spend. More than double the profit per job.

Run your own numbers

Six figures in, everything else out. Estimate where you have to — a rough number you actually use beats a precise one you never calculate.

The two rows in orange are the ones to memorise. Cost per paying customer is what marketing actually costs you. The LTV to CAC ratio tells you whether the business can afford to grow.

And the break-even cost per lead is your permission slip. Any lead below that number is profitable on the first job alone, before any repeat work. Most owners have never calculated it, which is why they either overpay in panic or underspend out of fear.

The ten numbers, one by one

Each of these gets a definition, the arithmetic, and an honest note on what good looks like.

1. Marketing spend

Everything you spent to get customers this month. Ads, agency fees, software, print, sponsorships, the van wrap amortised across its life.

The common error is counting only ad spend. That flatters every number downstream. If you pay an agency $1,500 and spend $2,000 on ads, your marketing spend is $3,500, and your real cost per lead is nearly double what your ad dashboard shows you.

2. Leads

Every genuine enquiry: calls, forms, texts, chat, walk-ins. Not impressions. Not clicks. People who raised a hand.

Define "genuine" once and stick to it. Wrong numbers, spam form fills and sales calls do not count. Changing the definition mid-year is how businesses convince themselves things improved when they did not.

3. Cost per lead

Marketing spend ÷ leads. Spend $2,000, get 40 enquiries, and your cost per lead is $50.

Useful for comparing channels against each other. Almost useless on its own, for the reason the chain section explained. Watch it as a trend rather than as a verdict.

If it is climbing, four causes are likely: more competitors bidding, a shift toward pricier channels, declining ad or landing page quality, or ordinary seasonal drift. Cross-industry cost per lead rose from about $198 to $214 between 2025 and 2026, so some increase is just the market.

4. Close rate

Customers won ÷ leads. Home services businesses in that 888-contractor dataset averaged 43.9%.

This is the most improvable number on the list and the one most owners ignore. It is not a marketing metric — it is an operations metric — which is exactly why it hides.

Things that move it, in rough order of impact: answering the phone within three rings, calling back within five minutes rather than five hours, quoting in a day rather than a week, and having someone who is actually good at the conversation.

5. Cost per paying customer

Marketing spend ÷ customers won. Or, equivalently, cost per lead ÷ close rate.

This is the number to know by heart if you only learn one. In that February 2026 dataset it was $233 against a $53 cost per lead — a 4.4x gap that most owners never see.

Compare it against gross profit per job, not revenue. A $233 acquisition cost on a $1,826 job sounds trivial until you remember you only keep the margin.

6. Average ticket

Typical revenue from one job. Easy to pull, easy to forget, and one of the few numbers you can raise deliberately.

Raising it is often faster than winning more customers. Bundling, a good-better-best option, a maintenance add-on, or simply quoting the better solution first — each moves this number without touching your marketing at all.

7. Gross margin

What is left after the direct cost of delivering the work — materials, subcontractors, labour on the job — before overhead.

Every marketing decision runs through this figure. A 4x return on ad spend is excellent at a 50% margin and loss-making at 20%. If you know only one thing from your accounts, know this.

8. Customer lifetime value

Average ticket × jobs per customer × gross margin.

Use gross profit, not revenue. A plumber with a $600 ticket, three jobs per customer and a 45% margin has a lifetime value of about $810 — not $1,800. The revenue version makes every downstream ratio look better than it is, which is precisely why so many people use it.

9. Return on ad spend

Revenue ÷ marketing spend. That 888-contractor dataset averaged 7.84x.

Beware this one. It is the most quoted and most misleading number in marketing, because it ignores margin entirely. A 4x return at a 20% margin means you spent $1 to make $0.80 of gross profit. You lost money while celebrating.

Always convert: revenue × margin − spend. That is what you actually made.

10. Marketing spend as a share of revenue

Marketing spend ÷ revenue. The usual rule of thumb is 5–10% for established businesses and 10–20% for aggressive growth.

Treat it as a sanity check, not a target. If every extra dollar returns more than a dollar of gross profit, the percentage barely matters — and businesses that cap marketing at an arbitrary percentage often cap their own growth by accident.

2026 benchmarks by industry

Useful for orientation, not for judgement. Your break-even beats every number in this table.

Two things stand out. Referrals are the cheapest leads available to any business, which is a strong argument for spending some of your effort on reviews and word of mouth rather than all of it on ads. We covered the mechanics in how to get more Google reviews.

And the gap between Local Service Ads at $53 and blended Google Ads at $70 is worth a look if you are in the trades. Different channels, materially different economics.

The numbers to stop tracking

Not because they are meaningless, but because they get confused with results.

The test is simple: if a number can double while your revenue stays flat, it is a diagnostic, not a result.

Rankings deserve a fair word. They matter — they are how the leads arrive in the first place — but a ranking report with no conversion context is a progress update, not a result. If your agency sends only ranking graphs, ask for the lead numbers beside them. We covered what a proper report should contain in what a Denver SEO agency actually does.

Website sessions are the sneakiest of the six. Traffic can climb 40% while enquiries fall, and it usually means the new visitors are the wrong ones. If that is happening to you, why your website gets traffic but no calls covers the diagnosis.

What if you can't track any of this?

Most small businesses cannot, at least at first. That is normal, and it is fixable in about a week. You do not need perfect attribution — you need enough to stop guessing.

Step one: call tracking

For most local businesses the majority of leads arrive by phone, and phone calls are the biggest blind spot in marketing. A tracking number on your website, and a different one on your Google Business Profile, tells you which channel produced the call.

This alone unlocks cost per lead, close rate and cost per customer. It costs a few dollars a month.

Step two: know where forms come from

Add a hidden field to every form recording which page it was submitted from, and make sure notifications include it. Then you can tell a blog enquiry from a service page enquiry.

Two related findings are worth knowing while you are in there. Multi-step forms — breaking one long form into two or three short steps — can roughly double completion compared with a single static form. And adding phone validation at the form level cuts junk submissions by up to half, which improves your cost per qualified lead even when raw cost per lead does not move.

Step three: ask the question

"How did you hear about us?" Asked by whoever answers the phone, written down every time.

It is unfashionably low-tech and it is often more accurate than analytics, because it captures the referral that started six months ago and the van someone saw parked on their street. Attribution software cannot see either.

Those three steps get you seven of the ten numbers. Average ticket and gross margin come from your accounts. Jobs per customer comes from your customer list. That is the full set.

One more thing the numbers will tell you

When you run the calculator, watch what happens if you nudge close rate up by ten points versus nudging cost per lead down by ten dollars. The close rate change almost always wins, and usually by a lot.

That points somewhere specific. If your leads are fine but your close rate is poor, the problem is rarely marketing. It is the speed of your response, the quality of your quote, or the experience someone has when they land on your site and try to work out whether to trust you.

That last one is worth checking honestly. A site that was fine in 2019 can quietly halve your close rate today without anything visibly breaking. If yours is dated or was never really built to sell, a good Denver web designer is usually a faster return than another dollar of ad spend — and our own conversion optimization work exists for exactly this reason.

The 30-minute monthly review

Monthly is the right cadence. Weekly is noise — small samples swing wildly and mean nothing. Quarterly is too slow to catch a problem before it costs you.

Block thirty minutes on the first working day of each month and fill in the same table.

1. Pull six figures. Spend, leads, customers won, revenue, average ticket, gross margin.
2. Calculate four more. Cost per lead, close rate, cost per customer, profit after spend.
3. Compare with the last three months, not just last month. One month is weather; three months is climate.
4. Ask one question per number that moved more than 20%. What changed? Not "is this good" — "what caused this."
5. Write down one action. One. The review is worthless if nothing changes because of it.

Do that twelve times and you will know your business better than most owners know theirs after a decade. Not because the numbers are clever, but because almost nobody bothers to look at them consistently.

Frequently asked questions

Keep reading