Case studies

Real accounts.
Real results.

These are anonymised accounts worked on across 10+ years in Amazon agencies. The problems, actions, and outcomes are all real.

Case study 01
Ad Cannibalization is Hurting Your Brand. How We Cut Ad Spend and Grew Profit.
Home Improvement UK Marketplace £25k–£30k/month
READ FULL CASE STUDY +
The situation

This brand was in a paradox that is more common than most sellers realise. Despite holding a top 1–2 ranking in their main category. With both a Best Seller and Amazon Choice badge. They were operating at a loss every single month.

Their product retailed at £25 (ex. VAT), with COGS at approximately 25% of retail price. Amazon and FBA fees, inflated by the product's large product dimensions, consumed a further 45–50% of revenue. After accounting for all costs including the agency retainer, the maths pointed to a maximum viable TACoS of around 5% to reach profitability.

The problem? At the time we took over, 80% of their sales were being driven by paid ads. And their TACoS was sitting at around 15%, putting the account firmly in the red.

Before vs. After
Before
TACoS~15%
Sales from ads80%
Organic sales20%
COGS~25%
Amazon + FBA fees~45–50%
Net marginLoss-making
After (8 weeks)
TACoS~5%
Sales from ads20%
Organic sales80%
COGS~25%
Amazon + FBA fees~45–50%
Net margin+10–15%
The diagnosis. Ad cannibalization

Before touching a single campaign, we used SmartScout to analyse their category market share. Even at that point, they were a top-3 brand in their category. Outperforming Chinese competitors priced 50% lower. We then used Data Dive to audit their organic keyword rankings. Out of their top 5 high-volume keywords, all five were ranking in the top 1–2 organic positions on terms with over 3,000 daily searches.

The root cause became clear: the brand was paying to advertise on keywords where they were already winning organically. Shoppers were clicking sponsored ads instead of the organic result directly beneath. And the brand was paying for every one of those clicks. This is ad cannibalization, and it was silently draining their margins.

What is ad cannibalization?
Cannibalization occurs when paid ads compete with your own organic listings for the same search terms. Driving up costs without adding incremental sales. In this case, the brand was essentially paying Amazon to show an ad above an organic result they already owned for free.
What we did
01
Listing overhaul
Professional 7-image carousel with lifestyle photography. SEO-optimised title and bullet points. Full A+ content and Brand Story rebuilt from scratch.
02
Keyword audit & tracking
Mapped all high-volume keyword rankings using Data Dive. Confirmed top 1–2 organic positions across primary search terms before touching ad spend.
03
Ad restructure
Removed keyword targeting budget entirely. Redirected all ad spend to ASIN targeting, competitor targeting, and category targeting instead.
+15%
Net profit margin
~5%
TACoS achieved
8 wks
Time to turnaround
The lesson

Most inexperienced advertisers would never turn off keyword advertising for a brand doing strong numbers. The assumption is that ads are driving performance. But when organic rankings are already dominant, running ads on the same terms actively hurts profitability.

The brands most at risk are those that are heavily ad-dependent. Because if ads stop for any reason (a credit card issue, a budget cap, a technical glitch), sales can collapse overnight. Building a business where 80% of revenue comes organically is not just more profitable. It is significantly more resilient.

Case study 02
Lost the Buy Box Due to Pricing. How We Got Back to Selling Without the Rollback.
Automotive UK Marketplace £360–£395 product
READ FULL CASE STUDY +
The situation

This brand had been selling a high-ticket automotive product on Amazon UK at £360 for over two years. A price that was stable, trusted, and well within Amazon's Buy Box expectations. Then their distributors and partners moved the product to £395 across all platforms, and Amazon was next.

The moment the price went up, the Buy Box disappeared. The product had held the same price for over two years, and Amazon flagged the sudden jump as a potential customer dissatisfaction risk. With the Buy Box gone, it was costing the brand sales every single day.

Keepa price history. Buy Box suppression visible from Apr 2025
Keepa price history chart showing Buy Box suppression after price increase
Why Amazon suppresses the Buy Box on price changes
Amazon awards the Buy Box to the most competitive offer. Not the highest price. For high-ticket products, even a price increase without any competitor undercutting can trigger suppression if the new price looks abnormal against historical pricing. The algorithm treats sudden spikes as a customer satisfaction risk and pulls the Buy Box immediately.
The numbers
The problem
Original price£360.00
New price£395
Price increase~10%
Time at original price2+ years
Buy Box statusSuppressed
The outcome
Original price£360.00
New price held£395
Price rollbackNone needed
Buy Box statusRestored
What Seller Support asked for. And why it stalled

Amazon's Seller Support demanded proof of legitimacy and confirmation the new price was justified. The client refused to lower the price. Their distributors had already adopted the new pricing everywhere, and reverting would damage those relationships. Weeks passed with zero sales. The risk of losing the client entirely was very real.

The fix. Thinking outside the box
01
Researched external pricing
Searched the product name on Google and manually reviewed every website selling it outside Amazon. Looking for sites pricing at £395 or higher.
02
Built the evidence pack
Compiled 3–5 distributor and retailer websites showing the same or higher price. Screenshots, URLs, live pricing. A clean, credible evidence pack showing market alignment.
03
Submitted to Seller Support
Presented the external pricing evidence as proof that the new price was not inflated. It was the market price. Buy Box was eventually reinstated.
0
Price reduction required
Weeks
Lost before Buy Box returned
3–5
External sources submitted
The lesson

Seller Support is not always the answer. Waiting on them while sales are bleeding is not a strategy. Sometimes the simplest move is the right one. Find the evidence Amazon is asking for, give it to them, and move on. In this case, the answer wasn't in Seller Central. It was on Google.

Case study 03
Why We Refused to Join Prime Big Deal Days. And Had the Best Two Sales Days of the Month.
Jewellery UK Marketplace £15k–£18k/month Oct 2024 vs Oct 2025
READ FULL CASE STUDY +
+30%
Revenue vs. October 2024
£0
Discounts given
<10%
TACoS held all month
October 2025 sales. £17,462 total, +30% vs. October 2024
October 2025 sales chart showing £17,462 total. 30% above October 2024
The situation

A UK jewellery brand operating at 15% gross margins was approaching Prime Big Deal Days (Oct 7–8, 2025). The default advice from every corner of the industry was the same: submit a deal, offer a Prime Exclusive Discount, increase the ad budget, and ride the traffic wave. The problem was the maths. At 15% margins, a mandatory 20% discount doesn't just reduce profit. It turns every sale into a 5% loss before ad spend is even factored in.

Why discounting at thin margins is a trap
Prime Exclusive Discounts require a minimum 20% price cut. On a 15% margin product, that turns every sale into a 5% loss before ad spend. Add deal fees (£40–£780) and CPCs that spike 60–80% during event windows, and participation becomes a guaranteed loss. Not an opportunity.
What participation would have actually cost
Scenario Revenue Discount Deal fee Profit
No participation (actual) £2,200 £0 £0 +£330
Prime Exclusive Discount (20%) £1,760 £440 £40 -£88
Lightning Deal (20%) £1,760 £440 £390 -£438
Best Deal (15%) £1,870 £330 £780 -£830
The four rules we held all month
01
Hold full pricing
No deals, no coupons, no Prime Exclusive Discounts. Jewellery is a considered purchase. Event-day shoppers were already interested buyers.
02
TACoS below 10%
Non-negotiable. This enforced bid discipline and prevented panic spending on inflated event-week keyword auctions where CPCs spiked 60–80%.
03
Budget ready for the surge
Ensuring daily ad caps could absorb the event traffic spike without cutting out mid-day when buyer intent was at its peak.
The lesson

Amazon markets Prime Big Deal Days as an opportunity. For thin-margin brands, it is often a trap. This brand's best two sales days of October came with zero discounts and zero deal fees. The 30% monthly lift is the headline. The margin they kept is the real story.

Case study 04
How We Restored Brand Control After Distributors Hijacked the Listings.
Automotive Aftermarket Parts Brand Protection
READ FULL CASE STUDY +
The situation

An automotive aftermarket brand sold through a network of authorised distributors on Amazon. Their model was simple: all resellers sell under one product listing, controlled by the brand owner. Before the client came on board with us, their catalogue was already in a mess due to numerous duplicate listings of the same product.

Some distributors had registered misspelled versions of the brand name, enrolled those fake variants into Amazon Brand Registry under their own accounts, and created their own separate listings. This meant good distributors were stuck with inventory they couldn't sell, and the brand owner had completely lost control of their own product page on Amazon.

The three-way deadlock
The distributors with the fake brand names had Brand Registry rights, which meant they could lock other sellers out. The good distributors had stock sitting in a warehouse going nowhere. And the brand couldn't push too hard against the bad actors because those same people were also their suppliers. Push too hard and they cut off supply entirely.
Why standard enforcement failed

We tried everything by the book first. We asked Amazon to merge the listings into one. Denied. We filed trademark infringement reports against the fake listings. Rejected, because each fake brand was already enrolled in Brand Registry, so Amazon treated it as a dispute between two brand owners and refused to get involved. We applied account health pressure. That worked for a while, but the bad-actor distributors threatened to stop supplying the client entirely, so we had to back off.

The fake trademarks were the problem. Getting them cancelled through official legal channels would take anywhere from 12 to 36 months. The blocked distributors could not wait that long.

What we tried and why it was not enough
01
Catalogue consolidation
Requested Amazon merge the fragmented listings into one. Denied. Amazon won't consolidate listings tied to different registered brand names.
02
Trademark infringement reports
Filed infringement reports against the misspelled-brand listings. Rejected. Each fake brand was Brand Registry–enrolled, so Amazon treated it as a dispute between two rights holders.
03
Account health pressure
Applied enforcement pressure through Amazon's account health tools. It worked. But triggered a wholesale retaliation threat that forced us to retreat strategically.
The solution. Build around the deadlock

Instead of keep fighting a battle Amazon was not going to resolve for us, we went around it. Using the client's legitimate brand ownership, we got a GTIN exemption from Amazon and created a brand new, clean product listing under the correct brand name. This time the client had full control from day one.

We then moved all the legitimate distributors onto the new listing. The fake listings were left alone. The bad-actor distributors still had their Amazon business, so they had no reason to retaliate. The problem did not get solved head on. It got made irrelevant.

0
LEGAL ESCALATIONS
100%
WHOLESALE RELATIONSHIPS PRESERVED
Day 1
FULL BRAND CONTROL RESTORED
The lesson

Amazon disputes with difficult sellers rarely get resolved with one clean move. When the normal tools stop working, sometimes the best thing you can do is stop fighting the existing mess and build something clean alongside it.

No lawsuits, no escalations, no lost revenue. Everything we used was already available to the client as the legitimate brand owner. Sometimes the smartest move is the one that does not need anyone else's permission.

Case study 05
Four Markets. Two Prime Events. Zero Discounts. Full Margin Every Time.
GPS Trackers UK & US Prime Day & PBDD 2025
READ FULL CASE STUDY +
$84,990
US. JULY 2025
£39,977
UK. JULY 2025
£0
DISCOUNTS GIVEN ACROSS ALL 4 EVENTS
The situation

A motorcycle and bicycle GPS tracker brand ran a strict no-discount policy across both 2025 Prime events — July Prime Day and Prime Big Deal Days in October — across UK and US Amazon marketplaces. Four event windows, four different market responses, and zero deals submitted, zero discounts offered, zero deal fees paid.

The 2024 comparison figures look dramatic because 2024 was heavily disrupted by inventory stockouts. The meaningful number isn't the percentage lift. It's that every pound and dollar above was earned at full margin, across all four windows.

Results across all four windows
MARKET EVENT TOTAL SALES STANDOUT DAY
UK July 2025 Prime Day £39,977 Jul 21 (~£2,300. Post-event peak)
US July 2025 Prime Day $84,990 Jul 19 (~$5,800. Post-event peak)
UK October 2025 PBDD £24,578 Oct 7 (~£1,750. Prime Day itself)
US October 2025 PBDD $39,934 Oct 9–10 (~$2,600/day. Post-event)
Why no discounts. Three reasons that made it obvious
01
Channel discipline
The brand had built distributor relationships beyond Amazon. Discounting on Amazon would undercut those partners and damage relationships core to long-term growth.
02
Margin protection
Prime Exclusive Discounts require a minimum 20% price cut. Lightning Deals and Best Deals add £390–£780 in fees on top. On hardware margins, the maths simply doesn't work.
03
Recovery momentum
After a 2024 disrupted by stockouts, training customers that "this brand discounts during Prime" would have damaged full-price sell-through for the rest of Q4.
The data. Four markets, four different patterns

The same no-discount strategy produced four genuinely different results. The most important insight: in three of the four markets, the highest sales day of the month fell outside the event window. "Prime Day = your best days" is a marketing assumption, not a data reality.

UK — JULY 2025
£39,977
UK July 2025 sales
Peak day: Jul 21, post-event
US — JULY 2025
$84,990
US July 2025 sales
Peak day: Jul 19, post-event
UK — OCTOBER 2025
£24,578
UK October 2025 sales
Best day: Oct 7 (PBDD Day 1)
US — OCTOBER 2025
$39,934
US October 2025 sales
Peak days: Oct 9–10, post-event
The October-into-Q4 trap
October PBDD isn't a standalone event. It's the start of an 8-week window through Black Friday, Cyber Monday, and December. Customers who don't buy during PBDD often hold their wallets, waiting for a deeper discount later. Panicking during a soft PBDD and slashing prices trains customers that the brand discounts under pressure. Damaging full-price conversion for the rest of Q4. Those buyers come back in November and December once they realise the discount isn't coming.
Who this strategy works for
RIGHT FOR BRANDS WITH
✓ Established product-market fit
✓ Channel partner relationships to protect
✓ Thin hardware-style margins
✓ The discipline to hold pricing when competitors panic-discount
WRONG FOR
✗ New launches needing sales velocity to build rank
✗ Commoditised "filter by deal" categories
✗ Brands that depend on event badges for visibility because organic rank is weak
The lesson

A no-discount strategy doesn't mean ignoring Prime events. It means refusing to let them dictate your pricing, channel relationships, or customer expectations.

Sometimes the event is your best week. Sometimes a regular Tuesday three weeks later beats every event day. Sometimes the post-event recovery is where the real revenue lives. In every scenario above, the brand made more money by not discounting than they would have by participating. Across four markets, two events, and an entire Q4 cycle.

For sellers willing to play the longer game, that's the actual prize.

See a pattern that applies to your account?

Book a free audit and I'll tell you exactly what I see. And what I'd do about it.

Book a free audit