How a deals strategy and ad optimisation moved Compostic from buying growth — to compounding it.
Compostic is the New Zealand–owned home-compostable bag brand sold across the United States, including Whole Foods Market in the Pacific Northwest and California. Sandwich bags. Snack bags. Cling wrap. Gallon zip bags. Variety packs. Every product breaks down in 12–24 weeks in a home compost — no industrial facility required.
By 2026, Compostic was the #1 compostable resealable bag brand on Amazon.com by parent-level revenue, ahead of Matter and Cleanomic. The product was working. The category had momentum. The problem was downstream.
In March 2026, Compostic's Amazon US sales grew 41.5% year-on-year. By any conventional measure, the brand was winning. Look one level deeper — at the same period — and the picture inverted.
Advertising costs were climbing faster than revenue. Discounting was holding velocity steady but eroding contribution margin. Net profit had fallen 16.3% despite the headline sales increase. The brand wasn't growing. It was buying growth — and paying for it twice: once at the till, once on the P&L.
The category needed an operating discipline that aligned spend, price, and rank into one compounding loop instead of three competing levers.
The diagnosis was clear: Compostic's three growth levers — deals, ads, and pricing — were running in isolation. Each had been optimised for its own metric. None had been coordinated around contribution margin. The play was to put all three on the same playbook.
Best Deals were sequenced around the TACOS–rank flywheel — each deal sized to lift BSR, not just clear inventory. Cooling periods enforced. Stockout risk audited before nomination. Deal price set at Amazon's max-eligible — never deeper than required.
Bleeders paused below 10% click-conversion. Converting search terms harvested from STRs and promoted to exact-match targets. Bids raised on brand and high-CVR generic terms — capped on category head terms. Outcome: more spend converted, less spend wasted.
Average selling price came down — through deal pricing, not blanket cuts. Each SKU verified against the contribution margin floor before any change went live. Velocity rose. Buy Box held. TACOS stayed in the 25–26% band even as price softened.
Between March and April 2026, the change reversed the trend on every metric that mattered.
The headline numbers are the visible result. The deeper outcome is what compounds:
Sales velocity increased materially. Higher unit-throughput taught Amazon's ranking algorithm to favour Compostic's listings over slower competitors — translating ad-driven traffic into organic placement.
Organic rank improved across the catalog. The deals weren't promotional events. They were rank-building investments. The lift persisted after each deal closed.
The flywheel started turning. Lower price unlocked higher velocity. Higher velocity unlocked better organic rank. Better organic rank unlocked headroom for paid bids. The whole system began funding itself.
It is a borrow against the next quarter, paid with compounded interest. Most Amazon agencies optimise one lever at a time — ads, listings, or deals — because that's how their teams are organised. The result is exactly what March 2026 looked like: sales up, profit down, the brand running fast in the wrong direction.
The Marketplace OS approach treats Amazon as one system with one north star — contribution margin at or above floor — and coordinates every lever in service of that single number. Sometimes the right move is more spend. Sometimes less. Sometimes a deal. Sometimes a price hold. The discipline is in knowing which, and when.
For Compostic, the answer was all three at once — sequenced.