Australia
Migration Agency
Google Ads and Meta Ads planned around each visa category.
- Marketing-attributed revenue
- AboutA$1.22m
- Reported over the latest 30 days.
- Ad spend
- AboutA$36,000
- Reported return on ad spend
- 34×
Quick answer
SoudCoh is a Melbourne Facebook and Instagram ads agency for businesses whose customers are not searching yet. Meta creates demand that Search later catches, so for most trades we run Search first. Our pixel and Conversions API set-up counts only real calls and forms.
Source: SoudCoh research, reported Google Ads data from Australian trade and home-service businesses. An enquiry is a recorded phone call or submitted form, not a booked job.
We create Facebook and Instagram campaigns that introduce your services, explain the offer and make it easy to enquire. The plan covers the audience, creative, landing page and follow-up, with results reviewed against your business goals.
250+ active client engagements · Campaigns running in Australia, the UK, the UAE, Saudi Arabia and New Zealand · Tracking built before the first ad runs
Australia
Google Ads and Meta Ads planned around each visa category.
Explore WELND’s paid social work and the website journeys behind finance and retail enquiries. The stories show the creative, the landing experience and what we delivered.
Showing 3 of 3 client projects
Swipe through the projects or use the arrows.


Australia
Connected
paid social and the finance enquiry journey
Meta advertising and website delivery
Read the case studyWebsite development

Australia
Debt test
nine questions answered privately in the browser
Website delivery
WELND: a finance website with a private debt test and calculators. See the design, service pages and enquiry experience.
WELND is a finance and mortgage brokerage based in Port Melbourne, specialising in tax, trading and unsecured debt for clients across Australia.
Website development

United Kingdom
Measure + choose
product ranges, guides and appointment booking
Website delivery
Impress Blinds: a blinds catalogue with measuring tools and local appointments. See the design, service pages and enquiry experience.
Impress Blinds makes and fits made-to-measure blinds, curtains and shutters, with free home appointments across England and Wales.
A closer look

WELND
Website project
WELND is a finance and mortgage brokerage based in Port Melbourne, specialising in tax, trading and unsecured debt for clients across Australia.
The work for WELND includes:
We built a nine-question debt test with no credit check, no email and nothing submitted. We also added a home page that answers common beliefs about debt before it asks for anything.

For the campaign side, read WELND’s paid social case study. Its website story explains the page and enquiry journey; the campaign story explains the lead-generation approach.
These are common areas to review when a Meta campaign attracts attention but enquiries or sales are below expectations.
Boosting picks the objective for you, and the objective it picks is engagement. You buy likes on a post about your new van, and nobody can tell you what a booked job cost.
When the same creative runs repeatedly, response can change. We review performance and test fresh images, videos and offers when needed.
Browser tracking alone loses a real share of conversions to privacy settings and blockers. Meta then optimises toward the conversions it can still see, which is a partial picture of your customers.
Three interests, two behaviours, an age band and a ten-kilometre radius, all stacked. The ad set never gets enough volume to leave the learning phase, so the cost per result goes up rather than down.
Without exclusions, your cheapest audience — people who already visited — sits inside your most expensive campaign. You pay prospecting prices to reach people you had already reached.
A newsletter signup and a A$4,000 job in one column. Meta then optimises toward whichever is cheapest to get, which is never the job.
A lead form takes two taps, so people fill it in while they are still on the couch. Answered in five minutes it is a customer; answered in two days it is somebody who has already called a competitor.
Real ads from SoudCoh’s own Meta account, and what testing them taught us.
Swipe for more. Text as it ran, trimmed where long. The creative panels are redrawn here, not the original image or video.
We can do this for you.
We test your Facebook and Instagram ads the same way, and keep backing the ones that win.
People move quickly through the feed. A clear offer, relevant image or video and easy next step help them understand why your ad matters.
The first two lines are all most people read. So they carry the offer and the reason to believe it, not the company history.
The offer, in one line a stranger understands.
The offer, in one line a stranger understands.
The proof underneath it — the guarantee, the response time, the licence.
The first two lines carry the offer and the reason to believe it, because that is where the feed cuts to “See more” and most people never tap it. The image has to work at thumbnail size on a phone held at arm's length. The headline is the last thing read before the tap.
Then we go back and read which combination Meta actually served, which is regularly not the one we expected. That check is the difference between making six creatives and knowing which two are doing the work.
125
characters of primary text
Roughly where the feed cuts to “See more”. Most people never tap it, so the offer lives above the fold or nowhere.
40
characters in the headline
The bold line under the image. It is the last thing read before the tap, and the easiest thing in the account to write badly.
30
characters in the description
The quiet line under the headline. Proof, not adjectives — the guarantee, the response time, the licence.
10
cards in one carousel
Ten services, ten suburbs or ten products in a single unit, each with its own image, headline and landing page.
50
results per ad set, per week
Roughly the volume Meta needs before delivery stops guessing. Split your budget across six ad sets and none of them ever get there.
1–10
% lookalike similarity
1% is the closest match to your source list and the smallest reach; 10% is the widest. Which one works is a question the data answers, not the brief.
Meta product limits and delivery mechanics, not performance claims.
SoudCoh · all channels
250+
Active client engagements
Google, Meta, LinkedIn, TikTok and organic, run by one team
National cultural institution
10
Paid channels, one media budget
Always-on, surge and test media for a national cultural institution
Where campaigns run
5
Countries
Australia, the UK, the UAE, Saudi Arabia and New Zealand
Figures trace to live engagements and are dated in the case study or engagement record they come from. Public-sector and cultural clients are identified by sector and jurisdiction only. None of it is a forecast for your account.
Our Meta management service covers measurement, audiences, creative testing, landing pages and reporting.
We review Meta Pixel and, where appropriate, Conversions API so eligible enquiry and sales events are reported consistently, with consent and duplicate handling checked.
Calls over sixty seconds, form submissions, bookings, chats and purchases as separate events with values attached. Meta optimises toward whatever you tell it matters, so it has to be told properly.
Customer lists, website visitors, video viewers, people who opened a lead form and did not finish. Those are your warmest audiences and they cost the least to reach. Interests come last.
A lookalike is only as good as its source. Seeded from all website traffic it finds you more traffic; seeded from people who actually booked, it finds you more of them.
Exclusions on every campaign so the two never overlap. Otherwise your cheapest audience sits inside your most expensive campaign and the reported cost per lead is an average of two very different things.
Fewer ad sets, each with enough spend to reach the volume Meta needs to leave the learning phase. Splitting a budget six ways feels careful and is the most common reason an account never settles.
The offer and the reason to believe it, before the truncation. Vertical video for Reels and Stories, captions burned in because most of it is watched on mute, and copy written for this platform rather than pasted from the website.
We test new concepts against existing ads and give each test enough budget and time to assess the result.
Frequency and cost per result read weekly, creative replaced on a schedule rather than after the numbers have already gone. Waiting for the drop is a month of budget spent finding out.
Product feed, Advantage+ shopping campaigns and dynamic retargeting for browsers and cart abandoners, with server-side tracking so revenue lands back in Ads Manager accurately. Online stores are very much in scope; the build is just different.
Thirty to forty-five minutes on what you sell, who buys it and what a customer is worth. Then we open the ad account, read the last ninety days, and write down every leak.
Pixel, Conversions API, conversion definitions and CRM routing go in first. Building creative on top of broken measurement is how agencies end up arguing about numbers in month three.
Audiences, exclusions, campaign structure, creative concepts, copy and landing pages. A real test enquiry is pushed through every tracking path. Nothing is live and nothing has spent.
Campaigns are created paused so you can read the whole build first. You activate when you are ready, not when we are.
Ad sets need volume before their numbers mean anything, so structure is held steady long enough to get it. Then the creative and audiences that are not carrying their share get cut.
A creative refresh cycle, weekly optimisation, reporting on a fixed cadence whether the news is good or bad, and a proper conversation each quarter about where the next block of budget should go.
We prepare the campaigns for review, check the tracking and agree on the budget before launch.
Meta optimises toward whatever it is told counts, and creative wears out fast. So a Facebook account puts its weight on what it counts, and on holding every ad to your cost-per-lead line.
Rule 01
The number it moves: True cost per lead
The pixel and the Conversions API go in before the first ad runs, so Meta learns from calls, forms and bookings rather than clicks.
Rule 05
The number it moves: Cost per lead
Concepts are tested wide; an ad or audience running above your cost-per-lead line is cut, and its budget goes to the few that book calls.
The case studies identify the channels used for each business. Google Ads results illustrate paid-search work and should not be read as Meta campaign results.
Before you ask us anything
Here is what everybody asks first.
Cost, timing, contracts, who owns the ad account, reporting, and what happens when a campaign is not working.
The first call is free and there is no deck.
Talk to us about your adsWe typically work best with established businesses comfortable investing A$10,000+ a month in marketing. We agree the mix of services, ad spend and fees with you before work begins.
Lead form campaigns can produce enquiries within a day or two of going live, because the form is two taps and never leaves the app. The first two to four weeks are a learning period, though — Meta is working out who to show your ads to, and we are cutting the audiences and creative that are not carrying their share. Campaigns generally settle into their best rhythm around 60 to 90 days. Your industry, your creative and your budget all move that timeline.
No long-term lock-in. We would rather keep the work because it is producing than because a contract says you have to stay. What we do ask for is enough runway to be fair to the account — a Meta campaign judged on its first ten days is being judged during the part where the platform is still learning. The terms are put in front of you in plain English before anything is signed.
We own and run the ad account, because the build is proprietary methodology refined across 250+ engagements — audience architecture, exclusion logic, creative testing frameworks and conversion tracking. Your Facebook page and Instagram account stay yours throughout, and you get full-visibility reporting on performance and spend at all times. On long-term marketing agreements, ownership and handover options are available to discuss.
We tell you, in the report, in plain English, and we change what we are doing. A campaign that is not producing gets rebuilt, repointed at a different audience, given different creative, or stopped. If the honest answer is that Meta is not the right channel for your business right now, we would rather move the budget somewhere that is than keep billing you for it. What we will not do is let it run quietly and hope it comes good.
A fixed cadence, good news or bad. Each report covers what was spent, what it produced, what we changed and why, and what happens next — written in the language you use about your own business rather than platform jargon. You can also see live performance and spend for yourself at any time. If a number moved, you will find out from us before you find out from your bank statement.
Yes, for planned work, in SoudCoh's experience of paid social: Facebook and Instagram ads suit jobs a customer can be prompted to want, such as a hot-water replacement, a switchboard upgrade, solar or air conditioning. Emergency work, like a burst pipe or a power outage, often starts with a search, making Google Ads worth considering. Creative testing can affect enquiry cost: on our own Meta account, our best ad was 2.7× cheaper per booked call than our median ad, across 184 ads tested.
SoudCoh is a Melbourne agency that runs Facebook and Instagram ads for lead generation, and these are the checks we would make of any agency. Ask to see real ads it has run and what testing them showed; who owns the ad account and pixel; whether leads are tracked with the Conversions API, not the pixel alone; and how fast a new lead reaches your phone. This page shows ads from our own Meta account, and our WELND case study covers paid social for a finance business.
Google captures demand that already exists — somebody types what they want and you are either on the page or you are not. Facebook creates it, by reaching people based on who they are rather than what they are typing. That makes Facebook harder work creatively and usually cheaper per person reached, and it makes Google the faster route to somebody who is ready right now. The most effective accounts run both: Facebook builds the awareness, Google catches the searches that follow, and the budget moves between them on evidence.
Yes — Meta is one of the strongest channels an online store has, and we run it end to end. Product catalogue and Advantage+ shopping campaigns, dynamic retargeting for browsers and cart abandoners, prospecting creative built for the feed, and full pixel and Conversions API tracking so revenue lands back in Ads Manager accurately. The approach differs from our service-business work — there we are capturing calls, bookings and quote requests, here we are capturing orders and repeat purchases — but the discipline is identical: track it properly, test it constantly, scale what pays.
With the browser pixel and the Conversions API running together. The pixel loses a meaningful share of conversions to privacy settings, tracking prevention and ad blockers; the Conversions API sends the same events from your website's server instead, so Meta sees conversions it would otherwise miss and can optimise toward them. It is set up before the first ad runs, matched against the pixel so events are not double-counted, and checked in the account rather than assumed.
Yes, and it is part of the job rather than an extra. Static images, carousels, short-form vertical video and the copy that goes with them, written for the feed rather than lifted from your website. Where your own footage or photography exists we will use it — on social, real material from the job site regularly outperforms polished production. We also run a refresh cycle, so creative is replaced before fatigue shows up in the cost per result.
A lookalike asks Meta to find people who resemble a list you already have — your customers, your website converters, your video viewers — at a similarity of 1% to 10%. It works when the source list is good and it is useless when it is not. Seeded from all your website traffic it will find you more traffic; seeded from people who actually booked a job, it finds you more of those. We build them from converters, refresh the source as the list grows, and test the similarity percentage rather than assuming 1% is best.
Yes — by suburb, by postcode, or by a radius around your premises, and set to people who live in or were recently in the area rather than anyone who has merely shown interest in it. It works the same way in any city we run: Sydney, Perth, Brisbane, Adelaide, Canberra, and outside Australia in the UK, the UAE, Saudi Arabia and New Zealand.
Service businesses and trades, finance, real estate, healthcare, professional services, hospitality, B2B, and online retail. The common thread is not the industry — it is whether the decision can be triggered rather than only answered. If somebody has to be searching before they will buy, Google is the better first move and we will tell you so. If the need can be created, or the purchase is visual, Meta is usually the cheapest place to reach people at volume.
Most of our clients use two or three of these together, with the budget moving between them on evidence rather than being split by habit.
Written for the trades who ask us this most often: the paid social formats currently earning their keep. And because a cheap lead that never gets called is not a cheap lead, the follow-up that turns an enquiry into a booked job is the half of it no ad platform will do for you.
We review your account history, creative, targeting and conversion tracking, then recommend the improvements worth making first.
A free Meta account review and clear recommendations.