When the right answer is to stop advertising

Where they were

Fernbrook were running 31% gross margin before Amazon touched it. After referral fees, FBA, and the storage that a slow moving 250g bag accumulates, the 250g product had roughly $1.80 a unit to play with. Their ACoS on it was 24%.

You cannot fix that with better keywords. At that margin there is no bid that makes the 250g bag profitable on paid traffic, and every month spent optimising it was a month not spent on the products where the maths worked.

What we did

1. 01

   Work out which products can carry advertising at all

   We built the margin model per SKU and then worked out the break even ACoS for each. Two products came out under 9%, which in a competitive grocery category is another way of saying no. Three came out above 20% and had real room.

2. 02

   Turn the ads off and watch

   We stopped advertising the 250g bag and the sampler pack completely, rather than reducing bids, because a small budget on an unprofitable product is just a slower version of the same problem. Both had enough organic rank and enough subscription customers to hold. Revenue on the pair moved less than 4% over the next quarter.

3. 03

   Move the budget to the 1kg bag

   The 1kg bag carries five times the contribution per unit and gets bought by people who have already decided they like the coffee. That is where the entire remaining budget went, along with a listing rebuild that put the roast date and the grind options above the fold.

4. 04

   Make the subscription the product

   Coffee is the ideal Subscribe & Save category and Fernbrook were treating it as an afterthought at a flat 5%. We restructured the tiers around the 1kg bag, set the delivery frequencies to match how fast people actually drink it, and put the subscription option into the image stack rather than leaving it to Amazon’s widget.

Where they are now

Ad spend went from about $5,400 a month to $2,100 with revenue essentially flat, which is a 22% improvement in contribution margin without selling a single extra bag.

Subscription is now 58% of revenue, which changes what the business is worth as well as what it earns. Fernbrook can forecast a month ahead for the first time.

They cut our ad spend by 61% and our revenue did not move. It took me a while to understand that this was the whole point rather than a failure to grow us.

Figures come from the client’s Seller Central and advertising accounts over the period described. Past results are specific to that brand, that catalogue, and that category.