They had a real business. They just didn't have this channel.

Where they started

A company that had solved everything except this

Kemp had built a multi-million dollar business in EMS and lifeguard equipment before we ever spoke. Real products, real distribution, real customers. What they didn't have was Amazon — not a neglected account, not a messy one. Nothing.

That's a harder starting point than it sounds. There was no history to read, no baseline to improve, and no way to know how the catalog would behave in a channel where the buyer finds you through a search box instead of a sales rep. Their season also runs hard: pool and rescue season means May through July carries more than half the year's revenue, so a launch that missed the window would cost a full year.

What we did

Built the channel in the order that compounds

• Launch, January 2024. Catalog structure, listings, and pricing built from scratch and set to hold up against the season rather than the month.
• The first ten months ran with zero ad spend. Listings, catalog, and pricing did the work. Growth came from the pages, not from buying it.
• Advertising started in October 2024, once organic performance proved the pages converted and there was something worth putting money behind.
• Two full seasons operated since, with inventory and pricing planned against the May–July peak instead of reacting to it.

The sequencing wasn't caution for its own sake. Spending into pages that don't convert is how an account shows motion without producing profit, and it's most of why the first year's margin nearly doubled before a single advertising dollar was spent.

What changed

Four and a half times the revenue, at nearly double the margin

Measured across the same seven months each year, so the season is held constant: sales went from $311,392 to $790,861 to $1,356,583. Profit after Amazon costs and cost of goods went from $30,527 to $147,556 to $256,902. Trailing twelve months now stands at $1.67M, roughly a third the size of the business Kemp spent years building elsewhere.

The number we care most about is margin. It went from 9.8% to 18.7% in the first year, then held at 18.9% while revenue grew another 72%. Holding margin through that kind of growth is harder than reaching it once, and it's the difference between a channel that's worth something and one that's just busy.

What didn't improve: advertising efficiency was worse this past year than the year before, at 7.75% of sales against 7.12%. That comparison flatters the earlier year, which had months with no advertising running at all. Within this year the trend is the right direction — 9.3% in the first half down to 7.3% in the second. Last November also finished at under 3% margin, the weakest month in the record.

Why it matters if you're reading this

A new channel is not a new business

Kemp already knew how to make and sell their product. What they were missing was one channel, and the temptation in that situation is to treat it like a growth experiment — open it, spend into it, see what happens.

The opposite worked. Ten months with no ad budget, building pages and pricing that could carry a season, then advertising once there was something to amplify. If you've got a business that works everywhere except Amazon, the order you do things in matters more than how much you spend.