The only marketing metric that matters for most Aussie SMBs

ROAS is a ratio that flatters everyone, including channels that aren’t working. CPL tells you what a lead cost but nothing about what it was worth. CAC sounds precise until you realise most businesses calculate it wrong, and even the correct version ignores what you paid to fulfil the work. Lifetime value is real, but quoting it without a retention model is guessing dressed up as analysis.

None of these metrics is useless. Every one of them is routinely used to justify marketing spend that isn’t producing business results.

The metric that actually matters — for most Australian businesses under $100k per month in marketing spend — is marketing-attributable contribution margin from new customers. It’s not a novel concept. But it’s the one that agencies tend not to push because it’s the framework most likely to reveal that the work isn’t earning its fee.

Why the Standard Metrics Mislead

ROAS (return on ad spend) measures revenue attributed to ads divided by what you spent. A 4× ROAS on Google Shopping means $4 of attributed revenue for every $1 spent. The problem: it says nothing about whether that revenue is profitable. If your gross margin is 35% and you’re running at 4× ROAS, you’re making 40 cents of gross profit per dollar of ad spend before platform fees, agency fees, fulfilment, returns, and overhead. That’s not a good business.

The deeper problem with ROAS is that platform attribution is designed to make the number look good. Last-click, 7-day click / 1-day view, view-through — these all inflate attributed revenue beyond what an unbiased model would show. A ROAS of 4× in Meta Ads Manager and 4× in Google Ads Manager simultaneously is mathematically impossible if the same sale is being counted in both platforms. It happens routinely because both platforms claim credit for the same conversion.

CPL (cost per lead) has the same structural problem, compounded by lead quality. A $45 CPL sounds efficient until you realise 60% of those leads are outside your service area, wrong business size, or already working with a competitor. The leads that actually convert might cost $180 each once you filter for quality — and that’s before they close.

CAC (customer acquisition cost) is the cleanest of the three if calculated correctly: total marketing and sales spend divided by new customers acquired in the same period. Most SMBs calculate it as ad spend divided by purchases attributed to ads, which understates the true figure by excluding agency fees, tools, internal time, and a chunk of non-attributed touchpoints.

The Metric That Holds Everything Together

Marketing-attributable contribution margin from new customers is the difference between:

• The value those customers generate in their first 12 months, minus the cost to fulfil that value (gross margin per customer)
• The total marketing investment required to acquire them

Written as a formula:

Marketing contribution margin = (New customers from marketing × Average first-12-month value × Gross margin %) − Total marketing spend

The first-12-month boundary is intentional. Businesses that cite three-year or five-year LTV to justify current acquisition costs are betting on retention that may not materialise. If you can prove retention — if you have actual cohort data showing customers returning at a known rate — then extend the window. If you can’t, 12 months is conservative and honest.

Total marketing spend means everything: ad spend, agency fees, tool subscriptions, internal time, any creative production costs. Not just the platform line item.

What This Looks Like in Practice

Consider a service business doing $80,000 per month in revenue. They’re spending $8,000 per month on marketing across Google Ads, Meta Ads, and a content programme. Their average client contract value in the first year is $12,000. Gross margin — revenue minus direct delivery costs — is 55%.

Each new client contributes $6,600 in gross margin ($12,000 × 55%).

Last month, four new clients came through marketing-influenced channels. That’s $26,400 in marketing-attributable gross margin.

Marketing spend was $8,000.

Marketing contribution margin: $26,400 − $8,000 = $18,400

That’s the number that matters. Not the CPL, not the ROAS. Is $18,400 in contribution margin from an $8,000 investment a good result? That depends on the business’s overhead structure — but it’s a real number you can act on.

Now change one variable: the business discovers that only two of those four new clients came through paid channels. The other two were word-of-mouth and wouldn’t have come through marketing regardless.

Marketing contribution margin: $13,200 − $8,000 = $5,200

Still positive — but now the question is whether $5,200 in monthly contribution margin justifies $8,000 in marketing spend, and whether adjusting the channel mix would improve it. That’s a conversation with real numbers. It’s also the conversation most agencies would rather not have, because it might mean recommending a smaller budget.

Attribution for Businesses With Limited Data

The standard objection to this framework at SMB scale is attribution: how do you know which new customers came from marketing? For businesses under $1M revenue, formal multi-touch attribution with a proper data warehouse is expensive and usually overkill.

A simpler approach works adequately for most situations.

Ask every new client where they heard about you — not via a dropdown in a form, but as an actual question during onboarding or the sales call. Record it. “Google search”, “saw a Meta ad”, “referral from a colleague”, “LinkedIn” — this is rough, but it’s directionally correct and it’s data you own. Most SMBs don’t do this, which is why they default to platform attribution and end up with inflated numbers.

Track the source field on leads — use UTM parameters on all paid ads, make sure your CRM or lead sheet captures them, and match closed deals back to lead source. This takes an hour to set up and produces a running picture of which channels are generating clients, not just clicks.

Run a simple 3-month cohort — for any new channel or significant budget change, track the 3-month post-acquisition value of customers from that channel separately. Do customers acquired through Meta Ads have the same average contract size as customers from Google? If Meta is bringing in project work and Google is bringing in retainer clients, the CPL comparison is meaningless.

None of this is sophisticated. That’s the point. Sophisticated attribution is a solution to a problem most SMBs don’t have. The actual problem is that nobody’s tracking new clients back to marketing activity in any systematic way at all.

Why Most Agencies Don't Frame It This Way

The contribution margin framework is uncomfortable for agencies that sell on vanity metrics because it makes the performance question binary: did the marketing investment produce more contribution margin than it cost? If the answer is no, the investment was wrong, regardless of how good the click-through rates were.

An agency that optimises your Google Ads campaigns and reduces your CPL from $80 to $60 while your close rate drops from 30% to 18% has produced worse results despite better-looking numbers. Under a CPL framing, the campaign improved. Under a contribution margin framing, it got worse.

Our [Marketing Strategy service page](/services/strategy/) covers how we approach measurement and reporting for clients — including what we commit to tracking from the start of an engagement. Our [CRO service page](/services/cro/) covers the conversion side of the equation: what happens after the lead comes in matters as much as what it cost to generate.

This week: get a piece of paper, write down how many new clients or customers came through marketing in the last three months, their average first-year value, your gross margin, and your total marketing spend. Run the formula. You may be surprised by what you find — in either direction.

If you want a structured way to model the numbers, the [CRO Calculator](/cro-calculator/) is a useful starting point for testing conversion and margin improvements at different spend levels.