Many Meta ads problems are not ad problems. Clicks that do not become orders, returns that look fine in the platform but not in the bank, spend that cannot rise without the numbers falling apart. If you are looking for a Meta ads agency for ecommerce, it helps to know whether the ads are the cause or the symptom.
SSO runs Meta advertising for Australian physical-product businesses as part of a wider plan. We work on the creative and the campaigns, and on the offer, website and economics they depend on. This page explains what the work includes, how we judge it and why it always starts with a Validation Sprint.
What our Meta ads work includes
Inside an ongoing programme, our Meta work covers Facebook and Instagram advertising from brief to report.
-
Campaign management.
Structure, budgets, audiences and placements, set up for the objective the plan needs and changed as evidence comes in.
-
Creative.
Hooks, angles, briefs and the ads themselves, tested against each other so the budget follows what works.
-
Tracking and measurement.
Checking that purchase data reaches the platform and matches the store closely enough to make decisions on.
-
Offer and landing page input.
What the ad promises and what the page delivers need to match. Where they do not, that goes into the plan.
-
Reporting in business numbers.
Cost per customer, blended return on media and gross profit after media, read alongside the platform's own metrics.
What we do not bring is a universal targeting recipe. Audience settings that suit one store's product, price and creative can fail for another, and Meta changes its tools and defaults regularly. We choose settings from the evidence in your account and from current Meta guidance, not from a template.
Meta ads rarely work alone. Google Ads can capture people already searching for what you sell, and email can bring first-time buyers back. If you are deciding between the two ad channels, see Google Ads vs Meta Ads for ecommerce.
Creative, offer and purchase experience
An ad can only sell what the offer and the site can close. We work on the three together, because improving one while another leaks just moves the problem.
Creative
Meta shows ads to people who were not searching for you, so the creative has to do the work that search intent does elsewhere. We test angles against each other: the problem the product solves, who it is for, and why someone should buy it from you rather than elsewhere. We follow what moves and stop what does not, quickly.
For Holtsee, a premium eyewear brand with strong margins, the order was to prove the creative first, then switch on retention, then scale the highest-margin product. By month three, blended return on ad spend sat near 6×, with 80% gross margins held.
Offer
The offer decides what you can afford to pay for a customer. If a customer is worth $55 and traffic costs $400, no campaign fixes that.
That was Apothecary by Justin in January 2026: $3,204 of ads, $548 of revenue and 8 orders. Three agencies before us had tried to optimise the ads. We changed the offer first, taking the average order from $69 to $137 by August 2026, and that is what gave the ads room to pay.
Purchase experience
Speed, friction, trust and checkout. A site that leaks is a tax on every campaign you will ever run. When clicks arrive but orders do not, we follow the click through the product page, cart and checkout before blaming the ads. Our guide to Facebook ads getting clicks but no sales walks through the same checks.
What we check before changing a campaign
- Does the landing page deliver what the ad promised, in the first screen?
- Are price, delivery and returns clear before checkout?
- Do the product page and add to cart work properly on a phone?
- Does purchase tracking record each order once, with the right value?
- Can the margin carry the current cost of buying a customer?
How we evaluate campaign performance
The return Meta reports is one view. It is based on Meta's own attribution settings, and other channels may claim some of the same orders. We read it, but we do not run the business on it.
-
Break-even first.
The return on media the business needs just to cover the variable cost of an order. For Apothecary by Justin, break-even was 1.20×. Our guide to what a good ROAS is shows how to work yours out.
-
Blended return across the store.
Total revenue against total media spend, which cannot count the same order twice. Apothecary's reached 2.61× in August 2026, on $27,469 of revenue from $10,525 of media.
-
Cost per customer against what a customer is worth.
If buying a customer costs more than that customer earns you, the campaign loses money on every new customer, however the dashboard looks.
-
Gross profit after media.
What is left once product costs and media are paid, month by month.
-
The next dollar, not the average.
Before spend goes up, we look at what the extra spend is likely to return, not the average so far. More in when to increase your ad budget.
| Measure | What it says | What we ask next |
|---|---|---|
| Platform-reported return 3.0× | Meta attributes $3 of revenue to each $1 of Meta spend | How much of that would have happened anyway, and do other channels claim the same orders? |
| Blended return 1.6× | Total store revenue is 1.6 times total media spend | Is that above what the business needs? |
| Break-even 1.8× | The store needs 1.8× to cover the variable costs of its orders | Blended is below break-even, so more spend at this return widens the loss |
Hypothetical figures for illustration.
We do not promise a ROAS. The return you need comes from your costs. The return you can achieve comes from testing, and it can change as spend, creative and competition change.
Numbers can also get worse before they get better. When an offer changes or campaigns are rebuilt, there is often a dip first. Apothecary took about seven months and $20,000 of capital all up to reach a month that paid him back. We explain why in why numbers dip before they improve.
Why engagement starts with a Validation Sprint
We do not run Meta ads as a standalone retainer. Ads management happens inside the Growth Program, after a Validation Sprint has shown what the ads need to achieve and whether the business can support them.
The reason is practical. Better ads for a business that cannot afford its customers just make the problem more expensive. The Sprint checks the business first, then the site, then the ads, and stops at the first one that does not hold.
The Sprint is four weeks, $4000 + GST, paid upfront. On a new business, week two puts campaigns live to buy data, with test media agreed and capped before anything goes live. That week is not meant to be profitable. Week three tests angles, audiences and offers against what week two showed. If you already trade, we audit your account and run one targeted test instead.
At the end you receive the plan in writing. If we go ahead, Meta work runs inside one of two routes: Launch, where the economics need building or repairing, or Accelerate, where they already work and you need capacity. Roughly six in ten completed Sprints are not followed by an offer from us (September 2026 record, not a promised future rate), and a positive finding does not automatically lead to one.
Want the full picture of what SSO does beyond Meta? Read our overview →
Not ready for a Sprint? The free Growth Bottleneck (opens in a new tab) tells you in under 4 minutes where your business is most likely stuck. If you want to run your own ads, the SSO AI Toolkit includes Ad Creative Briefs built on our Meta framework, which opens once your numbers say ads can pay.