The most expensive problems in Amazon advertising are the ones that never show up as problems. Here are the seven we find most often — and how to spot each one in your own account.
When a campaign fails loudly — ACOS spikes, sales crater — you notice and you fix it. The leaks that actually kill profitability are the quiet ones: spend that flows out month after month, hidden inside averages that look acceptable. Across hundreds of audits, these seven show up more than everything else combined.
Broad and auto campaigns are discovery tools. Their job is to find converting search terms you didn't know about — and then hand those terms over to exact match. The leak happens when the second half never occurs: the campaign keeps buying every loosely-related search, including hundreds of terms that have already proven they don't convert. If your search term report shows terms with 30+ clicks and zero orders still receiving spend, this leak is active. On a typical $20k/month account, it's worth $1,500–4,000 a month.
When branded and generic keywords live in the same campaigns, your reporting lies to you. Branded clicks convert at 3–5× the rate of generic ones, so blended campaigns look healthy while the generic spend inside them quietly bleeds. Worse: you may be paying for sales that would have happened organically. Splitting branded, generic, and competitor terms into separate campaigns is the single highest-clarity change most accounts can make.
The same keyword sitting in three campaigns doesn't get you three times the traffic — it gets you one auction where you're a bidder against yourself, with Amazon free to serve the highest bid. Spend fragments, data fragments, and optimisation becomes guesswork because no single campaign has enough signal to act on.
Look at which campaigns hit their daily budget caps. In most unaudited accounts, the campaigns running out of budget by mid-afternoon are the profitable ones — while underperformers happily spend their full allocation every day. The fix costs nothing: reallocate budget from the campaigns that never cap to the ones that always do. It's the cheapest growth available in any account.
Top-of-search placements can convert at double the rate of rest-of-search — or they can drain budget at premium prices with no conversion advantage, depending on the product and the term. Most accounts set placement modifiers once (or never) and leave them. The data to make this decision properly is in the placement report, and it changes over time. Check it quarterly at minimum.
Conversion rates are not flat across the day. In most categories we manage, evenings convert meaningfully better, while the early hours of the morning produce the worst ACOS of the day — clicks from browsers, not buyers. An account bidding identically at 8pm and 3am is overpaying for its worst traffic and underbidding on its best.
The maximum profitable bid for any keyword is a function of three things: how often clicks convert, what an order is worth, and the ACOS you can tolerate. Most accounts bid on none of these — they bid on habit, suggested bids, or whatever was set eighteen months ago. When bids aren't anchored to value, every other optimisation is rearranging deck chairs.
None of these show up in a top-line dashboard. They live in search term reports, placement reports, and hourly data — which is exactly why they persist for years inside otherwise well-run brands. Our free Amazon Profit Plan quantifies the first four automatically from your own reports; the rest surface in the walkthrough.
Everything in this guide is quantified in your free Amazon Profit Plan — wasted spend, missed keywords, budget inefficiency, in minutes.
Get Your Free Amazon Profit Plan →