Doubling revenue was three jobs, not one
Paid social brought the customers. The landing pages, the email program and the merchandising decided what each one was worth. Revenue rose 103%.
The brief
A brand doing well enough to stay flat
The brand sells jewelry and apparel to the same woman, which is an advantage nobody was using. Through the prior year revenue ran inside a narrow band, lifted for the December season, and settled back roughly where it started. Order value did the same. Nothing was failing, which is the harder brief, because there is no single broken thing to point at.
Paid social was running without a funnel behind it. Meta carried spend, but the ads pointed at product pages, and a product page assumes a shopper already knows the brand. Nothing downstream was built to catch a cold visitor who clicked and was not ready.
The email list was the clearest tell. Email revenue doubled inside the first month of the engagement, before any acquisition work had time to reach anyone. A number that moves that fast was not created by the work. It was already sitting there, waiting for someone to send the email.
So this was not a channel problem. Acquisition, retention and order value were three separate jobs, and doing only the first would have bought traffic the rest of the business could not convert or keep.
What we built
Three builds, running together
Paid social first, because nothing downstream matters while the top of the funnel is empty. The email program alongside it, because the list already existed and was the fastest revenue in the business. Then the merchandising work, because the cheapest order a brand will ever get is the second item in an order that was already happening.
- Meta as the primary acquisition channel
- Concepts and angles written to the brand’s audience, not the category
- Static, carousel and video running against each other
- Headline and copy variants A/B tested to find the performers
- A purpose-built landing page behind every ad
- Social proof, shop the look, video and real reviews on those pages
- Twelve to fourteen campaign sends a month
- New launches, seasonal moments and product spotlights
- Every campaign cross-selling between the two categories
- A welcome flow built to convert what paid social was sending
- Cart abandonment capture
- Post-purchase sequences carrying what customers actually ask
- Higher-priced hero products promoted to first-time buyers
- Upsell modules on the product page
- Complete the look bundles in the cart
- Email flows recommending premium add-ons
- Recommendations driven by real browse and purchase behavior
The user experience
What actually changed for a shopper
A media plan is only worth what it changes on the screen. These are the changes carrying the result, and the reason each one should.
The numbers
Aug 2024 to Apr 2025, against the year before
| Metric | Change | Reads as |
|---|---|---|
| New customers | +160% | Result |
| Monthly email revenue | +148% | Result |
| Online store revenue | +103% | Result |
| Orders | +95% | Result |
| Return on ad spend | +650% | Result |
| Average order value | +11% | Guardrail |
| Returning customers | +21% | Watch |
Revenue outgrew orders, so the growth is volume with order value intact rather than discounting. The honest tension in this set is the last row. New customers grew close to eight times faster than returning ones, which is what an acquisition-led first year looks like and also the thing that has to change in the second. A base acquired this quickly has to be worked, or the same growth has to be bought again next year.
The takeaway
The channel is not the build
Paid social takes the credit in a result like this, because it is the visible part and the part with an invoice attached. It is the smallest part of the work. A cold click is worth exactly what is standing behind it: a page written for someone who has never heard of the brand, a welcome sequence that earns a second attempt at the order, a cart that knows what goes with what. Buy traffic without those and all you have bought is traffic.
The three results at the top of this page are not three campaigns. They are one customer, met three times.