Split a budget between search and social, honestly
A social lead almost always costs less than a search lead. It also books at a lower rate, because nobody on Facebook went looking for you. This splits a budget across both and settles the argument on cost per job rather than cost per lead.
Your numbers
The answer
Blended cost per booked job
$314.24
15.9 jobs from $5,000
Google — cost per job
$303.07
11.5 jobs from 38.5 leads
Meta — cost per job
$343.83
4.4 jobs from 36.4 leads
Blended cost per lead
$66.80
74.9 leads in total
Revenue
$19,093
15.9 jobs at $1,200.00
How this was worked out
- Google spend
- $5,000 × 70% = $3,500
- Google leads
- $3,500 ÷ $90.92 = 38.5
- Google jobs
- 38.5 × 30% = 11.5
- Meta leads
- $1,500 ÷ $41.26 = 36.4
- Meta jobs
- 36.4 × 12% = 4.4
- Blended cost per job
- $5,000 ÷ 15.9 = $314.24
What the published benchmarks say, and what they do not say
| Channel | Published cost per lead | Market | Sample |
|---|---|---|---|
| Google search — Home & Home Improvement | US$90.92 | United States | WordStream 2026: 13,474 search campaigns, April 2025 to March 2026 |
| Meta lead campaigns — Home & Home Improvement | US$41.26 | United States | WordStream 2025: 726 lead-objective campaigns, April 2024 to June 2025 |
| Either channel, Australia | No published figure | — | No dataset with a stated method exists |
| Lead-to-job rate, either channel | No published figure | — | Nobody publishes this by channel in any market |
The two published rows are medians in US dollars from two different windows, so even the ratio between them is rough. The two blank rows are the numbers that would actually settle this argument, and neither exists — which is why both lead-to-job rates on this page are inputs rather than defaults we pretend to know.
What to do with this
- At these settings Google search buys the cheaper job — $303.07 against $343.83. Note that the channel with the cheaper LEAD is Meta, which is the comparison most people stop at.
- Meta would have to convert at 13.6% to match Google on cost per job here. You have it at 12%. That gap, not the cost per lead, is the whole argument.
- Before shifting budget on this, check both channels are counting the same thing as a lead. A form fill on a social lead ad and a phone call from a search ad are not comparable objects, and most channel comparisons are wrong at exactly this point.
Change anything above and the link in your address bar changes with it, so you can send the exact result to whoever needs to see it. Nothing you type is sent anywhere — the whole calculation happens in your browser.
What this does, and how the maths works
What it does
- Splits one monthly budget across Google search and Meta at whatever ratio you choose.
- Applies a separate cost per lead and a separate lead-to-job rate to each channel, because those two numbers are what make the comparison honest.
- Reports blended cost per lead and blended cost per job, and shows which channel is actually buying the cheaper job at your settings.
How to use it
- Enter the total monthly budget and the share you want on Google.
- Enter what a lead costs on each channel. If you have run both, use your own figures. If not, the fields carry published US medians and say so.
- Enter the share of leads from each channel that become paid work. This is the input that decides the answer, so use your own records if you have them.
- Read the cost per job for each channel, then adjust the split.
The formula
- Google spend = total budget × Google share
- Leads per channel = channel spend ÷ channel cost per lead
- Jobs per channel = channel leads × channel lead-to-job rate
- Cost per job = channel spend ÷ channel jobs
- Blended cost per job = total budget ÷ total jobs
Reading the answer
- If search buys the cheaper job, the case for social is reach and remarketing rather than direct enquiries — and it should be budgeted as that.
- If social buys the cheaper job at a lead-to-job rate you can actually evidence, shift budget. If the rate is a guess, measure it for a month before you shift anything.
- The break-even point is worth finding: drag the lead-to-job rates until the two cost-per-job figures match. That tells you how much worse social leads can be before search wins outright.
What it cannot tell you
Both default costs per lead are United States medians in US dollars and they aggregate whole industries. There is no Australian equivalent for either channel. Treat them as a starting shape, replace them with your own numbers, and do not compare a US dollar figure against your Australian budget without noticing you have done it.
Where every figure on this page came from
Every default in the calculator above is either your own number or a figure from one of the studies below, with the market and the sample stated. Nothing here is an estimate somebody felt was about right, and no United States figure is wearing an Australian label.
WordStream / LocaliQ, Google Ads Benchmarks 2026
WordStream (LocaliQ) · 2026 · measured in US · high confidence
13,474 United States search campaigns, 1 April 2025 to 31 March 2026. Figures are medians.
United States data. Australian auction prices are widely believed to differ, and no published Australian dataset measures by how much.
WordStream, Facebook Ads Benchmarks 2025
WordStream (LocaliQ) · 2025 · measured in US · high confidence
726 United States lead-objective campaigns, 1 April 2024 to 30 June 2025. Figures are medians.
United States data, lead-objective campaigns only.
Frequently Asked Questions
Everything you need to know about working with SoudCoh
Have more questions? Let's chat!
Book a Free Strategy CallBy cost per lead, yes, in the only datasets with a published method. WordStream's United States figures put Home & Home Improvement at US$41.26 on Meta lead campaigns and US$90.92 on Google search — roughly half. By cost per booked job the answer is usually the other way round, because a search lead went looking for the service and a social lead was interrupted while doing something else. Neither dataset publishes a lead-to-job rate by channel, in any market, so that half of the comparison has to come from your own records.
There is no defensible generic answer, and a percentage split is the wrong first question anyway. Start by funding search to the point where you are not losing enquiries to budget — the impression share calculator shows whether you are — and treat social as the next dollar rather than a fixed share. If you cannot yet evidence a lead-to-job rate for social, run it for a quarter at a size you can afford to learn from before you weight a plan on it.
Because cost per lead flatters whichever channel produces the most leads, regardless of whether any of them turn into work. A channel producing leads at $30 that book 5% of the time costs $600 a job. A channel producing leads at $90 that book 30% of the time costs $300. Cost per lead says the first is three times better. Cost per job says it is twice as expensive.
None with a published methodology that we could find. The WordStream and LocaliQ Facebook datasets are explicitly United States only, and no Australian source publishes a comparable figure with a stated sample and window. We have not relabelled the US number as Australian, and any calculator that shows you an AUD Meta benchmark without naming its dataset has.
Tag the source on every enquiry at the point it arrives, not afterwards from memory, and reconcile it against invoices monthly. Phone calls are where this usually falls apart: without call tracking, every phone enquiry lands in the same bucket and the channel comparison becomes fiction. That plumbing is the least glamorous and most valuable part of running two channels at once.
Run these next
One number rarely settles anything on its own. These three answer the questions this one raises.
Cost per lead calculator
Work out cost per click, conversion rate and cost per lead from three numbers, then compare against a US benchmark and our measured Australian range.
SoudCoh case studies · WordStream 2026
Impression share and missed leads
Convert impression share and the two lost-share figures from your Google Ads account into missed enquiries, and separate the ones budget can buy from the ones it cannot.
Your own numbers only — no benchmark used
True cost per booked job
Add management fees, tools and tracking to your media spend, divide by booked jobs rather than leads, and see what is left of the margin after acquisition.
ATO
We will read
your real ones.
Send us the account rather than the estimate. We will tell you what it is actually costing to win a job, which part of the chain is leaking, and whether it is worth fixing — before anyone asks you to sign anything.