Google Ads can be worth testing when people already search for what you sell. Your offer and enquiry process have to work too. And a new customer must contribute enough to cover the full cost of winning one. When one of those fails, fix it first or put the money somewhere else.
Bottom line Yes. Fund a limited test with a written review date
Bottom line No, not yet. Name the fix and its cost before you spend
When Google Ads is worth considering
Search ads put you in front of people at the moment they look for what you sell. You choose the searches and the area. You pay when someone clicks. The demand already exists. You're renting access to it.
The limits come from the same place. If few people search for your service where you work, no budget fixes that. Ads also amplify whatever is already there. A muddled offer, a slow callback or a thin margin gets more expensive with every click.
So the useful question isn't which industries Google Ads works for. Anyone who says it always works for lawyers, or never works for online stores, hasn't seen your numbers. Your own margins, close rate, capacity and cash decide whether it's worth it, not an industry verdict. The question is what makes the opportunity plausible for your business.
Urgent work suits search because the need is specific and the timing is now. That holds as long as someone picks up the phone. High contribution work, like a renovation or a legal matter, can afford fewer leads at a higher price.
Repeat business can justify a loss on the first sale. But your own records must show that customers return. You also need to be able to fund the gap until they do.
Weak foundations change the answer to not yet. If the business can't answer, measure or fulfil the enquiries, that isn't a verdict on the channel. It's a list of things to fix before the first click.
The maths that decides whether it is worth it
Whether Google Ads is worth it for your business comes down to one comparison. On one side is what a new customer contributes. On the other is what it costs to win one.
Contribution is what a sale leaves after the direct costs of delivering it. Those costs include materials, subcontractors, stock, delivery, payment fees and expected returns. Contribution is measured before any spending on winning the customer.
That spending is the acquisition cost:
media
management
tracking and other tools
staff time on the campaign
setup work on the page or the tracking
Fixed overheads and tax sit outside the model. So the bottom line is contribution after included acquisition costs. It isn't net profit. Calling it that hides the costs left out.
In the table, CPC is cost per click and CPL is cost per lead. CAC is customer acquisition cost, the cost of winning one new customer.
Measure | Formula | What it tells you |
|---|---|---|
Expected clicks | Media spend divided by CPC | A scenario assumption, not a traffic promise |
Genuine leads | Clicks multiplied by the enquiry rate | Real enquiries, with duplicates and spam removed |
New customers | Genuine leads multiplied by the lead to customer rate | Use the rate for the same lead stage, from a mature batch of leads |
Media CPL | Media spend divided by genuine leads | Not the full cost of a lead |
Included cost CAC | All included acquisition costs divided by new customers | Name the costs included and the period |
Break even all in CAC | Contribution per new customer | Before any allowance for overheads or profit |
Break even all in CPL | Contribution per customer multiplied by the lead to customer rate | A cost per lead, not a cost per customer |
Media CPL ceiling | All in CPL ceiling less other acquisition cost per lead | The other cost per lead changes with volume |
Take a job worth $1,000 that leaves 45% after delivery costs. That means each new customer contributes $450. If 35% of genuine leads become customers, the most a lead can cost in total is $157.50. That's $450 multiplied by 35%. The most a customer can cost is $450.
Those are different units. Mixing them up is how campaigns get judged wrong. Both are break even figures with nothing left over. That's why a sensible target sits below them, leaving room for overheads and profit.
Our guide to a good cost per lead for Google Ads covers how to set that target. It also explains what counts as a genuine lead.
The media ceiling is lower again, because the other acquisition costs come out of the same $157.50. Say a month brings 40 genuine leads and $600 of management, tracking and other costs. That's $15 a lead. Taking it off leaves a media CPL ceiling of $142.50.
Check it from the other end. At that ceiling, 40 leads cost $5,700 in media, or $6,300 with the other costs. Forty leads at 35% is 14 expected customers, who contribute $6,300 at $450 each. Nothing is left over: that's what break even means. So don't deduct the fees a second time from a figure already called all in.
Volume moves the ceiling too. With 20 leads, the same $600 is $30 a lead. That drops the media ceiling to $127.50.
Keep every cost in one ledger:
media
management
tools
staff time
one time setup for the page and tracking
A setup bill spread over six months in the model still leaves the bank in month one. Because of that, record how each line is allocated in the model and when it's actually paid.
A forecast can also show a fraction, such as 8.4 expected customers. That's an average across many months. A report shows the customers you actually won, counted from your own sales records.
A managed account adds a fee line. Elev8d recommends at least $3,000 a month for a managed account. A self managed test in a low CPC niche can start nearer $1,000 to $1,500. Our guide to how much Google Ads costs in Australia covers market prices and every cost category.
The estimator works out media spend and keeps fees separate. It doesn't calculate gross or net profit, break even CPC or ROAS (return on ad spend). You'll need to add contribution and any missing costs to your own ledger.
What to fix before you spend a dollar
A business can pass the maths and still waste the test. That happens because the clicks land somewhere that can't turn them into customers.
These six conditions aren't a score out of six. One critical failure is enough to fix first. That could be a page that breaks on a phone or an offer you can't lawfully advertise. It could also be outcomes you can't measure, or no capacity to do the work.
Condition | What ready looks like | Watch for |
|---|---|---|
Offer and customer | One clear service, the area you actually serve, the customer you want and the next step | A discount, guarantee or free consultation isn't compulsory. Every claim must be supportable |
Destination | A page that matches the search, has one main action and works on a phone | A relevant existing page can work. A new landing page isn't always needed |
Measurement | Forms, calls, bookings or sales recorded and linked to the outcome | A click on a phone number isn't an answered call. GA4 and Tag Manager are options, not requirements |
Trust | Genuine proof: credentials, examples of work and accurate reviews where allowed | Having no reviews doesn't decide your conversion rate or rule advertising out |
Follow up and capacity | Each lead has an owner, a response time and a recorded outcome, and someone can do the work | Missed leads raise the cost per customer, not the media cost per lead |
Remarketing and exclusions | A suitable later audience, and exclusions for unwanted traffic | Depends on eligibility, policy, list size and the goal. Not an entry requirement |
Every claim in the ad and on the page has to be one you can back up. The ACCC's guidance on false or misleading claims explains the risk. Any statement that creates a false impression about goods and services can break the law. A claim about the future also needs reasonable grounds when it's made.
The page itself has to answer the search the ad paid for. Our guide to Google Ads landing page vs homepage covers which page to use.
Measurement needs testing, not trust. We ran 180 Google Ads account audits from August 2021 to September 2026. In those audits, we checked the conversion actions used for bidding or reporting. About 7 in 10 accounts had at least one fault in them.
Calls are a common gap inside that figure. In the same audits, about 2 in 5 accounts that take enquiries by phone had that gap. The goal they used for bidding or reporting counted no calls.
Set up and test your Google Ads conversion tracking before the first click is bought. If the phone matters, do the same for your call tracking for Google Ads.
Follow up decides what the enquiries are worth. Give every lead a named owner and a response time the business can keep.
Then test the route end to end. The same audits found a broken route in about 1 in 5 accounts. A clearly labelled test enquiry showed the success message but never reached anyone who would respond.
Missed leads don't change the media cost per lead, because the enquiries still arrived. They cut the share that become customers. That pushes up the cost per customer.
Remarketing can help later, but it isn't an entry requirement. Health and other sensitive services face tighter rules. Our guide to Google remarketing ads covers when it fits.
The account has a checklist of its own, from billing to conversion actions and campaign settings. It's in our guide to setting up Google Ads for a small business.
We'll check your tracking, your page and your lead handling. Then we'll tell you what we'd fix before you spend.
Four worked scenarios with honest verdicts
Four example businesses, one set of rules. The inputs are chosen teaching figures in AUD, not industry benchmarks, Elev8d prices or client results. Figures exclude GST, assuming the business claims it back. If it can't, the GST is a cost too.
Each scenario counts media, a management fee and tracking or tools. Setup is $900 for the plumber, $600 each for the clinic and removalist and $1,200 for the store. Each counts one sixth of that one time cost, to spread it over six months. That spread is a planning aid, not accounting advice.
Item | Plumber | Clinic | Removalist | Fashion store |
|---|---|---|---|---|
Media | $3,000 | $2,500 | $2,000 | $3,000 |
CPC | $25 | $10 | $10 | $2 |
Clicks | 120 | 250 | 200 | 1,500 |
Enquiry or order rate | 10% | 6% | 8% | 2.5% |
Leads or orders | 12 leads | 15 leads | 16 leads | 37.5 orders |
Lead to customer rate | 70% | 35% | 35% | Not applicable |
Expected customers or orders | 8.4 | 5.25 | 5.6 | 37.5 |
Revenue per first sale | $650 | $450 | $800 | $75 |
Contribution rate | 55% | 60% | 40% | 20% |
Contribution per sale | $357.50 | $270 | $320 | $15 |
First sale revenue | $5,460 | $2,362.50 | $4,480 | $2,812.50 |
Contribution before acquisition | $3,003 | $1,417.50 | $1,792 | $562.50 |
Management | $500 | $500 | $400 | $500 |
Tracking or tools | $50 | $50 | $50 | $100 |
Setup allocation | $150 | $100 | $100 | $200 |
Total included acquisition | $3,700 | $3,150 | $2,550 | $3,800 |
Included cost CAC | $440.48 | $600 | $455.36 | $101.33 |
Contribution after acquisition | -$697 | -$1,732.50 | -$758 | -$3,237.50 |
Swap in your own CPC, enquiry rate, close rate and contribution. The same arithmetic shows where your business stands.
Emergency plumber
The plumber has the friendliest assumptions of the four. It gets urgent searches, a 70% close rate and a 55% contribution rate. It still loses money.
On media and management alone, the loss is $497. That's $3,003 of contribution less $3,000 of media and $500 of management. Add $50 of tracking and $150 of setup allocation and the loss is $697.
SourceExample scenario, AUD excluding GST
The diagram's spend is all in. So its contribution multiple of about 0.81 means the campaign hasn't paid for itself.
The click price is the problem. At $25 a click and a 10% enquiry rate, each lead costs $250 in media. Each lead is worth $357.50 multiplied by 70%, or $250.25. Media alone uses the whole ceiling.
On the same budget and costs, the plumber breaks even only at a CPC of about $20.30. The alternative is an enquiry rate of about 12.3%. Urgent intent and a high close rate don't make the numbers work by themselves.
Clinic with repeat visits
The clinic loses $1,732.50 on first visits: $1,417.50 of contribution against $3,150 of acquisition costs. Its case rests on patients coming back. Say the 12 month contribution from each new patient really reaches $1,080, about four visits' worth. Then the 5.25 expected patients contribute $5,670. That leaves $2,520 after the same $3,150, before any retention, reminder or financing costs the model leaves out.
First visit contribution leaves -$1,732.50 after included acquisition costs
If 12 month contribution reaches $1,080 a patient, $2,520 remains before omitted later costs
That's a conditional result, not cash at launch. The $3,150 leaves in the first month. The $5,670 arrives over a year, if it arrives at all. Use the clinic's own records of return visits rather than a belief that patients usually come back. Our customer lifetime value calculator models value and payback from the repeat purchases you enter.
Health services carry extra advertising rules. The TGA's frequently asked questions on advertising health services and cosmetic injections cover prescription only medicines. Those medicines, including ones used in cosmetic injections, can't be advertised to the public. They can't be referred to in an ad either, even through an acronym, nickname or hashtag. The TGA adds that clinics should focus their advertising on the types of consultations available.
Ahpra's Summary of the advertising requirements says advertising must not use testimonials. It also warns that patient stories may create an unreasonable expectation of beneficial treatment.
This scenario is unit economics, not a template for a treatment ad. Check every ad and landing page against both rule sets before spending. Google Ads for clinics deals with the compliance side.
Removalist
The removalist loses $608 on media and management alone. That's $1,792 of contribution against $2,000 of media and $400 of management. Add $50 of tracking and $100 of setup allocation and the loss is $758.
SourceExample scenario, AUD excluding GST
One fix doesn't rescue it. Raise the enquiry rate from 8% to 10% and the same 200 clicks bring 20 leads. Those leads become 7 expected jobs. That's $2,240 of contribution, still short of the $2,550 of costs. Any proposed fix has to go back through the numbers before anyone assumes it works.
Online fashion store
The store's average order is $75. Its product margin is 35%. Fulfilment and expected returns take another 15% of revenue, leaving 20%, or $15 an order.
Keeping 37.5 expected orders throughout, revenue is $2,812.50 and contribution $562.50. Against $3,800 of acquisition costs, that's a loss of $3,237.50. In a real store, start from net revenue and don't deduct returns twice.
Cheap clicks don't save it. Media alone costs $80 an order against $15 of contribution. Even at half the CPC, 75 orders would contribute $1,125 against the same $3,800. At a 20% contribution rate, each dollar of media needs $5 of revenue just to pay for itself. The order value, the margin or the repeat rate has to change first.
All four lose money on these inputs. That's the honest result. A losing Ads scenario doesn't make SEO or social media profitable by default either. A viable case comes from changing an input the business can actually change, then recalculating.
How long can it take to recover the cost?
Five milestones run on different clocks. They're easy to confuse, so here they are in order:
The first conversion can arrive on day one.
Enough mature data to judge the economics takes longer.
A positive period, when contribution beats that period's costs, comes later again.
Recovering the setup and early losses comes after that.
Cash payback comes last of all.
Reporting lag stretches the early part. Google's page on how long it takes for your customers to convert explains the lag. Conversions can be reported up to 90 days after the click, depending on the conversion window. Recent results can show fewer conversions and a higher cost per conversion than they eventually will. So Google suggests a date range ending at least 30 days ago, or more with a longer window.
Google Ads also records each conversion against the date of the click. Its page on understanding your conversion tracking data explains this. That's why last month's figures keep moving for a while.
Allow 4 to 8 weeks for a first read and about 90 days to judge cost per lead. Allow longer when the sales cycle is long. Our guide to how long Google Ads takes to work explains this. That's when the evidence starts to mean something, not a date when the campaign turns a profit.
Cash runs on its own clock. Here's an example business that pays $1,200 for setup up front. It loses money in its first period and earns a modest contribution after that.
Example period | Contribution after included cash costs | Cumulative position |
|---|---|---|
Setup payment | -$1,200 | -$1,200 |
Period 1 | -$800 | -$2,000 |
Period 2 | $200 | -$1,800 |
Period 3 | $600 | -$1,200 |
Period 4 | $600 | -$600 |
Period 5 | $600 | $0 |
Period 2 is the first positive period. But the business is still $1,800 behind. It only gets back to zero in period 5 if every assumption holds. This isn't a normal ramp or an expected profit date.
Each period's contribution is assumed collected after its costs are paid. The setup appears once, as the payment. It isn't counted again as an allocation. Your real dates come from your own payment and collection terms.
Business.gov.au's guide to choosing between cash and accrual accounting sets out the difference. Cash accounting tracks the actual money coming in and going out. Accrual records sales and expenses when they take place. A job invoiced in March and paid in May is March revenue and May cash. A 12 month lifetime value forecast is neither in month one.
When outcomes become measurable: [date, plus the conversion and sales lag]
First positive operating period: [assumed period, and why]
Cumulative loss to recover: [setup plus early losses, in dollars]
Cash collection and payments: [when costs are paid and when customers pay]
Break even cash date if assumptions hold: [date]
If collection or repeat value is delayed: [what you'll do, and when]
What could change the result?
Three levers move the result. The margin sits behind all of them. Each lever needs its own diagnosis. After every change, the numbers need running again.
Lower the cost per click
Check the search terms you're paying for before chasing a cheaper click. Irrelevant searches inflate the cost of every useful lead. A cheap click from the wrong person is no saving. Our guide to Google Ads cost per click and how to lower it covers the diagnosis.
Improve the enquiry rate
Send a test enquiry. Then check, on a phone, that the page delivers what the ad promised. If people click and don't enquire, our guide to Google Ads clicks but no leads works through why.
Improve the lead to customer rate
Look at what happens after the enquiry. Check how fast someone responds, how many quotes go out and where leads go cold. Better follow up costs no extra media.
Contribution per customer is the fourth lever. Pricing, job mix, delivery costs and returns all move it. The break even CPL moves in proportion.
For the removalist, no single lever is enough. Each of these still leaves a loss of $246 to $310 on its own:
a CPC of $8
an enquiry rate of 10%
a close rate of 45%
a contribution rate of 50%
Applied together, the first three turn a loss of $758 into $1,050 of contribution after acquisition costs.
SourceExample scenario, AUD excluding GST
The steps depend on the order, because each lever works on the volume the one before created. That's 250 clicks instead of 200, then 25 leads, then 11.25 expected jobs. Modest improvements compound. One heroic assumption rarely carries a campaign.
Suppose no realistic combination of CPC, enquiry rate, close rate and margin clears the ceiling at any volume. Then Google Ads isn't the channel for that offer at that price, for now. The honest answer is no.
Compare the next investment in Ads or another channel
Whether Google Ads deserves the next investment depends on what else the same money and time could do. No alternative is free just because it has no media bill. SEO has no charge per click. But the work still costs money and takes time, with no certainty at the end.
Option | Demand and feedback speed | Effort and cash | Unsuitable when |
|---|---|---|---|
Google Ads | Catches existing searches, with a first read in weeks | Media, setup and management, paid from the first click | Searches are thin or the economics fail at any volume |
Organic visibility (SEO) | Builds over months, so feedback is slower | Content and technical work, paid well before any return | You need the money back this quarter |
Paid social | Creates interest before people search, with feedback in weeks | Media plus a steady supply of creative | The offer can't be shown in a picture or a short video |
Fix the enquiry journey | Uses demand you already have, and a test enquiry shows the fault at once | Staff time and page or process work | Too few suitable people are enquiring in the first place |
Retention and referrals | Uses past customers, so the speed depends on the repeat cycle | Staff time, plus permission to contact customers | There are few past customers or little need to buy again |
The best next dollar goes to whatever is holding the business back. That often isn't more traffic.
Bottom line Fund a limited test and judge it on contribution
Bottom line Budget for the work and the uncertainty
Bottom line Fix the measured constraint first
Our guide to Google Ads vs Facebook Ads for small business compares the two on outcomes and skills. The direct channel comparison sits in our guide to SEO vs Google Ads. If organic visibility is the alternative, work out whether SEO is worth it for a small business. Use your own numbers.
Agency help is part of the same decision: a cost and a capability, not a promise. A fee is good value when the work it buys changes the numbers by more than the fee. The change can come from:
cheaper useful clicks
a higher enquiry rate
cleaner measurement
hours the owner gets back for paying work
Running it yourself saves the fee but not the maths. Without the $500 fee, the plumber is still $197 short, before counting the owner's time.
Doing it yourself suits one service in one area and a modest budget. It also suits an owner with time to read the search terms every week. If the economics fail without a fee, no fee fixes them. And no agency, ours included, can promise a profitable campaign.
Fees are compared in our guide to Google Ads management pricing. Choosing a provider is covered in our guide to how to choose a Google Ads agency.
Plan an affordable test and make the next decision
A test is worth funding when you know the question it answers and the most it could lose. Set the budget from the loss the business can carry if every assumption fails. Don't set it from what a plan says it needs. Our guide to how much to spend on Google Ads covers setting a monthly figure.
Make the unknowns visible
Write each input as a range with its source and date. Leave a blank where you have nothing yet. Google's page About Keyword Planner forecasts explains that Google estimates what you might get from historical search data. It also lists reasons forecasts can differ from actual results.
A forecast is planning evidence, not a promise of leads. So are industry figures such as those in our Google Ads cost per lead by industry table. Run conservative, base and favourable scenarios instead of one national average.
One uncertain input can swing the whole result. Take a test with these inputs:
$2,000 of media
a $10 CPC
$500 of other acquisition costs
a 25% close rate
$500 of contribution per customer
Then change only the enquiry rate.
Enquiry rate | Clicks | Leads | Customers | Contribution | After $2,500 of costs |
|---|---|---|---|---|---|
4% | 200 | 8 | 2 | $1,000 | -$1,500 |
8% | 200 | 16 | 4 | $2,000 | -$500 |
12% | 200 | 24 | 6 | $3,000 | $500 |
Tripling the enquiry rate turns a $1,500 loss into a $500 gain with nothing else changed. The favourable case is only as good as the assumption behind it. Losses sit below the line in coral.
So size the loss allowance for the conservative case. If losing $1,500 would hurt, this test is too big for the business.
Judge new customers, not reported conversions
Compare mature batches of new customers against your own sales records. Brand searches and returning customers can flatter a campaign. That's because some of those people would have found you anyway. Attribution assigns credit for a sale. Incrementality asks whether the ad caused it.
Google's Conversion Lift is built to measure the causal impact of ads. But it isn't available for all Google Ads accounts. A small campaign may also lack the conversion volume. Report that limitation rather than claiming proof. Our guide to Google Ads ROI covers the full formulas and reconciling reported sales with your records.
Make the next decision
The review has five possible outcomes:
start
fix a named blocker
continue within a limit
stop
record the test as inconclusive
Stop at once for broken measurement or an operational failure that wastes the spend. Poor mature economics can justify stopping before any three month mark. Sparse data can leave the test inconclusive after the loss limit is reached. Then the right call is to stop and record it as inconclusive, not as proof the channel fails.
Continuing is a separate decision from starting. The setup money is spent either way. So ask whether the next period's contribution will cover its avoidable costs: media, fees and tools.
Don't keep spending to win back the setup. But keep reporting the full history so a run of good months can't hide the total. Growing a campaign that passes is covered in our guide to how to scale Google Ads.
Business question and outcome: [the offer and area, what the test must show and what counts as an outcome]
Known inputs and uncertain assumptions: [demand, CPC, enquiry rate, close rate and contribution, each with its source and date]
Full cost and maximum affordable loss: [ledger total, and the loss the business can carry]
Capacity and cash horizon: [work you can take on, and how long you can wait for cash]
Review conditions after the sales cycle: [review date, and the outcomes needed to judge]
Decision: [start, fix, continue, stop or inconclusive]
Owner and next action: [name, action and date]
What we recommend at Elev8d
We start with the maths, not the ads. Before anyone opens an account, we work out what a customer contributes. We also work out the most a lead can cost once fees and tools are counted. If that ceiling can't be met at any realistic CPC and enquiry rate, we say so. The money goes into the offer, the enquiry process or another channel instead.
When the maths can work, we fix the blockers next. That means a page that answers the search, plus tested conversion and call tracking. It also means a named person who responds fast. Then we fund one service in one area, with a loss allowance the business can live with. We write the review date down before launch.
We take a first read at 4 to 8 weeks and judge cost per lead around day 90. We go by new customers rather than reported conversions. If the next period can pay its way, we continue within a limit. If it can't, we change one lever and recalculate, or we stop. Inconclusive is a result too: we'd rather stop cleanly than keep spending to win back the setup.
Say the numbers support a test and you'd rather not run it yourself. Our Google Ads team can run that test for you.
FAQ: Google Ads for small business
Can a new business test Google Ads?
Yes, with a credible offer, a working enquiry route and a loss it can afford. What it lacks is history. It has no close rate, no enquiry rate and no record of what a customer is worth.
Start with conservative assumptions and a small loss allowance. Treat the first months as evidence gathering. Being new doesn't guarantee any particular conversion rate.
Is there any free advertising on Google?
Not the way most people hope. A normal Google Ads campaign pays for every click. A Google Business Profile is free to create. But it's a listing, not an ad campaign.
Google Ad Grants is only for eligible nonprofits, as Google's Ad Grant requirements and program FAQs set out. Our guide to Google Ad Grants in Australia covers eligibility.
Google Ads promotional offers are for businesses new to Google Ads. They ask you to spend a set amount or complete an action within a set time. And they carry their own terms.
How small can a first test be?
Google's description of Google Ads says there's no minimum spending commitment. It also says you set and control your own budget. The practical floor is a budget that buys enough clicks to learn something within your loss allowance. Our Google Ads minimum budget guide covers what a few hundred dollars a month can tell you.
Do I need a set number of conversions first?
No, not to start a test. Our rule of thumb is 15 to 30 conversions a month before trusting a target. Our guide to Smart Bidding strategies explains this. Google itself sets no minimum for Target CPA.
Fewer conversions don't stop a test. They mean more caution with automated targets and a longer wait for a result that means anything.
Next steps
Work out your break even cost per lead
Take contribution per customer from your own job records. Use the close rate from a mature batch of enquiries. Multiply them, then take off your fees and tools per lead to find the media ceiling.
Fix the first blocker you find
Send a test enquiry through every route. Call your own number from a mobile and read the page on a phone. Fix what fails before you spend.
Write the test plan before you spend
Fill in the test template, including the loss you can carry. Then book the review before the first click is bought.
