Add up the sales your channels claim credit for and you’ll often find the total exceeds your actual revenue.
Google Ads counts a sale. Meta counts the same sale. Your email platform counts it too. Nobody is lying — each is reporting that it touched the customer somewhere on the way to purchase. The problem is that you’re reading three overlapping claims as if they were separate results.
This is attribution, and it determines where your budget goes. Getting it wrong means cutting the channel that was working and doubling down on the one that took the credit.
Why the Numbers Disagree
A customer sees a Facebook ad on Monday, searches your brand on Wednesday, clicks a Google ad, leaves, then opens an abandoned cart email on Friday and buys.
Which channel earned that sale?
Meta will claim it, because its ad started the process. Google will claim it, because its ad was clicked. Your email platform will claim it, because its email preceded the purchase.
All three are describing something real. The sale is one sale.
The Models, and What Each One Hides
| Model | Credits | Systematically undervalues |
|---|---|---|
| Last click | Final touchpoint | Awareness and discovery channels |
| First click | Initial touchpoint | Closing channels like retargeting and email |
| Linear | Every touchpoint equally | Nothing — but flatters weak touchpoints |
| Time decay | Recent touchpoints more | Long consideration cycles |
| Data-driven | Modelled contribution | Needs volume; opaque reasoning |
Most Australian businesses default to last click, because it’s what the platforms show by default and it feels concrete. It also systematically undervalues everything that happens before the final click — which is most of what builds demand.
The practical consequence: businesses cut brand and awareness spend because it “doesn’t convert”, then wonder why their high-performing search campaigns get more expensive over the following six months. If your Google Ads costs have crept up without an obvious cause, our guide to Google Ads for Australian Shopify stores covers the account-side checks worth running first.
What Changed Recently
Two things have made attribution meaningfully harder, and neither is going away.
Privacy changes. Cookie restrictions, iOS tracking prompts and browser blocking mean a growing share of the customer journey is invisible to your analytics. Reported conversions are increasingly modelled estimates rather than counted events.
Cross-device behaviour. Someone researches on their phone at lunch and buys on a laptop that evening. Unless they’re logged in, those are two different people as far as your tracking is concerned.
The result is that attribution reports are directionally useful and precisely wrong. Treating them as exact is the mistake.
The Question Nobody Asks
There’s a simpler check available to most businesses, and it costs nothing: ask customers how they found you.
Not in a form field — those get skipped or answered carelessly. Ask on the phone, in the follow-up email, or during onboarding.
The answers rarely match your analytics, and the gap is informative. “I’ve been seeing your posts for months and finally looked you up” is a real customer journey that attribution software will record as direct traffic with no source at all.
A Practical Approach
For most Australian businesses, three things beat any single model.
Watch trends, not absolutes. Cost per acquisition moving month on month within one channel is reliable. Comparing two channels’ claimed conversions is not.
Run holdout tests. Turn a channel off for two or three weeks and watch total revenue, not that channel’s reported revenue. If total revenue doesn’t move, the channel was claiming credit for sales that would have happened anyway. This is the single most informative test available, and almost nobody runs it.
Track blended cost per acquisition. Total marketing spend divided by total new customers. It ignores the attribution argument entirely and tells you whether marketing as a whole is getting more or less efficient.
One caveat before you act on any of it: if the problem is that traffic arrives and doesn’t convert, attribution isn’t your issue at all. Our article on doubling enquiries without increasing ad spend deals with that situation.
Where to Start
This week: Add up the conversions each platform claims for last month and compare the total to your actual order count. The gap is the size of your attribution problem.
This month: Calculate blended cost per acquisition for the last six months. That trend line is more useful than any individual channel report.
Next 90 days: Run one holdout test on the channel you’re least sure about. Two weeks off, watching total revenue rather than channel-reported revenue.
Frequently Asked Questions
Which attribution model should I use?
For most businesses with a short purchase cycle, data-driven or time decay is more realistic than last click. But the model matters less than being consistent — changing models mid-year makes your own data incomparable.
Why does Google Ads report more conversions than my store?
Google counts conversions on the date of the ad click, not the date of purchase, and includes modelled conversions where tracking is incomplete. Some difference is expected. A large or growing gap suggests a tracking problem.
Is GA4 attribution reliable?
It’s directionally useful and increasingly modelled. Treat it as an estimate that’s good at showing change over time and poor at settling arguments between channels.
How do I attribute offline sales?
Ask, and record it in your CRM. It’s manual and imperfect, but for businesses where enquiries convert by phone it’s usually more accurate than the analytics.
Should I use an attribution tool?
Only once your spend justifies the cost and complexity. Below a certain volume, blended cost per acquisition plus holdout tests will tell you what you need to know.
Related Reading
- Google Ads + Shopify: How to Drive More Sales to Your Australian Store
- How to Choose a Digital Marketing Agency in Australia
- Landing Page Conversion Rate: How to Double Enquiries Without Increasing Ad Spend
Perfect attribution isn’t available any more, and chasing it wastes time. What’s available is knowing whether marketing overall is getting more efficient, and testing rather than assuming which channels are carrying the load.
HHP Design manages Google Ads, Amazon Ads and SEO for Australian businesses, with reporting built around outcomes rather than channel credit. Get in touch to discuss your setup.