Ecommerce Business Solutions
Case Study · Home & Furniture

Eight years and $80.6M in ad sales on a freight-heavy catalog

A furniture and bedding brand, established offline but doing $100K to $200K a month on Amazon when we took the account on in 2016. Over the next eight years advertising delivered $80.6M in ad-attributed sales at a 10.53% ACoS, on a catalog where freight and returns punish every mistake. The click-through rate stayed at 0.45% throughout, and that was deliberate.

$80.69M
Ad-attributed sales
account lifetime
10.53%
ACoS
held across the run
9.50x
Return on ad spend
lifetime
0.45%
Click-through rate
deliberately low, see below
How to read the numbers. Figures come from the Amazon Ads campaign manager for this account. “Sales” in that report is ad-attributed revenue, not total store revenue, so the $80.69M is what advertising was credited with rather than everything the brand sold. The reported window is the account lifetime; our engagement ran from 2016 until the brand was acquired in October 2024. We do not guarantee outcomes. We show our work.
Background

Where the account started

The client. A furniture and bedding brand: mattresses, foam products and home furnishings. Established off Amazon, but relatively new to the platform and running at roughly $100K to $200K a month when we came in.

What was wrong. No coherent advertising structure, a catalog whose variations had fragmented, and listing health nobody was actively managing. The demand existed. Nothing was set up to capture it efficiently.

Our scope. Full advertising ownership across Sponsored Products, Brands and Display, plus catalog and listing hygiene, variation architecture and fulfilment strategy, from 2016 through to the acquisition in October 2024.

The Problem

Three things made this category punishing

  • Freight and returns eat the margin

    Heavy, bulky items make standard FBA punitive. Oversized storage fees, fulfilment costs and removal orders take a large bite out of every unit, and a returned mattress is rarely resellable. Advertising had to work at a genuinely low ACoS, not an aspirational one.

  • An aggressive review climate

    Bedding and foam is one of the more cutthroat categories on Amazon, with persistent black-hat behaviour: manipulated reviews and listing hijackers. Defending the listings was ongoing work, not a one-off cleanup.

  • Hundreds of variations to keep intact

    Size, colour and firmness across a large catalog. Without strict parent-child hygiene, review velocity splits across children and every variant competes from a weaker position than the family would have together.

Strategy

Three decisions that shaped the account

1

Position for the commercial buyer and the multi-pack

Rather than fighting exclusively for hyper-competitive consumer head terms, we positioned variations explicitly at commercial buyers, bulk orders and hospitality clients. That meant writing and targeting for the buyer furnishing twelve rooms, not the one buying a single mattress.
Why it worked: Commercial and bulk intent carries a higher average order value and far less competition than the consumer head terms, so the same ad dollar buys a larger order against a cheaper click.
2

Take the oversized units out of standard FBA

We shifted oversized units onto Seller Fulfilled Prime and localized third-party warehouses. Prime eligibility was preserved, which conversion depends on, without paying Amazon’s oversized rates on every unit sitting in a fulfilment center.
Why it worked: Bypassing Amazon's oversized storage and fulfilment fees protected the margin that made a 10.53% ACoS viable in the first place. Advertising efficiency and logistics are the same problem on a heavy catalog.
3

Defend the brand's own traffic

We allocated deliberate budget to brand defense campaigns, keeping lower-margin competitors from poaching high-intent buyers off our core listings. In a category with active hijacking, that placement is not optional.
Why it worked: On a catalog with this many variations, a competitor taking the product-page placement is intercepting a buyer who had already chosen. Brand defense is cheap relative to re-winning that buyer through open search.
Execution

How the campaign structure changed over eight years

  1. 2016 to 2018
    Control first

    Tightly controlled exact match and product-targeting campaigns. Search terms kept deliberately narrow to prevent budget bleeding into high-volume keywords that would not convert on a considered, high-ticket purchase.

  2. 2019 to 2022
    Scale out

    Single product ad groups for the hero listings, so each could be bid and budgeted on its own economics, alongside structured Sponsored Brands video campaigns to hold top-of-search real estate.

  3. 2023 to 2024
    Automate the harvest

    Broad and auto campaigns feeding harvested winners into high-bid exact match, with heavy use of Sponsored Display view remarketing to recover shoppers in a long consideration cycle. Furniture is rarely bought on the first visit.

Turning Points

Two moments that moved the account

  • The 2018 pivot

    Between late 2017 and mid-2018 we pushed advertising hard to secure organic ranking on the core product lines, accepting a worse ACoS while it ran. Once organic rank stabilised we pulled spend back sharply. ACoS dropped and sales held, because organic was now carrying what paid had been buying. That spike and retreat is visible in the account's spend history.

  • The 2022 demand boom

    By mid-2022 the catalog optimization had fully matured and landed against peak seasonal home-goods demand. High-intent traffic and conversions spiked together, which is the combination you want: the account was ready when the demand arrived rather than scrambling to catch it.

The Counterintuitive Bit

Why we kept the click-through rate at 0.45%

A 0.45% CTR looks like underperformance. Paired with a 9.50x ROAS it is the opposite, and it was engineered.

On bulky, high-ticket items an accidental click from a casual browser is expensive and almost never converts. The buyer who wants a queen-size, medium-firm, 12-inch foam mattress is a different person from the one idly browsing “mattress”, and only one of them is worth paying for.

So we filtered deliberately: highly specific long-tail keywords, aggressive negative keyword lists, and explicit pricing and sizing callouts written into the copy and images. Shoppers who were not a fit self-selected out before clicking. The ones who clicked already knew the size, the spec and roughly the price.

Return on ad spend9.50x
ACoS10.53%
Click-through rate0.45%

Plotted on the same scale to make the point: the metric that looks weakest is the one doing the filtering. Fewer clicks, better clicks.

Worth saying plainly: this is not universal advice. On a low-priced impulse product you want the opposite, because volume is the point and a wasted click costs cents. Deliberate filtering pays on considered, high-ticket purchases where a wrong click costs real money and converts at nearly zero.
Honesty

What did not go to plan

  • Supply chain disruption, 2020 to 2021

    Global shipping constraints caused sustained out-of-stock periods on hero items. We lost top organic rankings we had spent years earning, and had to buy the positions back later at a worse margin than holding them would have cost. Inventory is an advertising problem, and this is the clearest example of it we have.

  • European expansion did not work

    We took the heavy catalog into European marketplaces and the localized freight economics did not support it. Margins came in poor enough that we retracted back to North America rather than subsidise the expansion. The lesson held: on a freight-sensitive catalog, geography is a margin decision before it is a demand decision.

Outcome

How the engagement ended

In October 2024 the brand was acquired by an aggregator. Operations and advertising moved to the buyer’s internal enterprise team and our engagement wrapped up with the transition.

That is the ending we want on an account like this one. A brand doing $100K to $200K a month on Amazon in 2016, built over eight years into something an acquirer wanted to buy and run themselves.

If your catalog looks like this one

This playbook transfers to heavy or bulky products where fulfilment cost decides the margin, to categories with an active B2B or bulk buyer, to large variation families that need their review velocity kept together, and to considered purchases where filtering the click matters more than winning it. It works because advertising, logistics and catalog structure were treated as one problem rather than three.

We do not guarantee sales or rankings, and we will tell you when the data says no. Send us your last 90 days of Business Report and advertising exports and we will return a written audit with a plan of action, risks included.