A dead account turned into ten times return, four months running.
In March the account returned $398 on $381 of spend, which is a rounding error away from losing money. From April it returned roughly ten times its spend, and it kept doing it for four months.
- Chelsey Jean
- Ecommerce, health and wellness
- March to July 2024
- Google Ads, Google Shopping, Merchant feed, Conversion tracking, SEO
The background.
Chelsey Jean is a health and wellness brand in Australia specialising in pneumatic lymphatic drainage products, selling through a WooCommerce store. We were brought in to run website management, SEO and Google Ads together rather than as three disconnected jobs.
The budget was small and had to stay small, which rules out the usual answer of spending more until something works. Everything had to come out of structure and selection instead.
The problem.
A return of 104.54% means the account brought back about a dollar for every dollar it spent, before the cost of the product, the shipping and the time. It was not losing money dramatically. It was doing nothing, expensively.
The audit went through the campaigns, the conversion tracking, the keyword structure and the product feed. Everything downstream depended on tracking being right first, because an account optimizing toward a number that is wrong will confidently get worse.
- $398.06
- Conversion value, March 2024
- 104.54%
- Return on ad spend
- $25.99
- Cost per conversion
- $381
- Spend
What we did.
Conversion tracking audited first
- Detailed tracking was implemented so campaign performance could actually be measured. Nothing else on this list is worth doing until this one is right.
Campaigns restructured around categories
- Products were organized into specific categories and the campaign structure was built to match, weighted by the traffic each category was already getting. Each category was then tested on its own so budget could follow the winners.
High value products only
- On a tight budget, spend went to high value items rather than being spread across everything. A small budget spent evenly is a small budget wasted evenly.
Shopping set up properly
- Shopping ads were implemented across the range with product meta details and images optimized to stand out. Wholesale and on sale products were excluded, and separate feeds were built per category so each could be tested weekly.
Keywords grouped by intent
- Phrase and broad match were both used for testing, grouped by what the searcher was trying to do. Branded terms were added as negatives on the non brand campaigns to stop them cannibalising each other, alongside a general negative list.
A branded campaign of its own
- The brand already had traffic, so a responsive search ad was set up on the brand name rather than letting that demand go uncaptured or get absorbed into a generic campaign.
Audiences segmented properly
- Website traffic, other paid platform traffic, organic traffic, email, product page visitors, all visitors and the customer match list were built out as separate audiences, with remarketing segmented by recency and by behavior such as cart and checkout.
Product pages tied back to the keywords
- The terms each product campaign targeted were worked into the matching product page, so the search relevance and the ranking improved together rather than the ads carrying it alone.
Tested continuously
- Headlines, descriptions, URLs, images and landing pages were all A/B tested, and performance was monitored and adjusted rather than set once and reported on monthly.
The results.
| Return on ad spend, March | 104.54% | 1,001.47% in April | 10x |
|---|---|---|---|
| Return on ad spend, May | 104.54% | 924.45% | Held |
| Return on ad spend, June | 104.54% | 1,437.72% | 14x |
| Cost per conversion | $25.99 | $8.35 | Down 68% |
| Conversion value, June | $398.06 | $32,300 | 81x |
| Conversion value, 17 June to 16 July | $398.06 | $27,400 | On $2,570 of spend |
What it added up to.
Four consecutive months at roughly ten times return matters more than any single month of it. One good month is luck. Four is a structure that works.
The brand now averages around nine hundred dollars in daily sales, and about thirty four percent of that is attributed to Google Ads. The rest comes from the search and site work running alongside it, which is the point of doing them together.
Ready to move the numbers?
Let us talk about your current goals, what is working, what is not, and where the biggest opportunities may be.