Restructuring Amazon PPC: From Chaos to a Profitable 3 Tier System

Most Amazon ad accounts we inherit look like 18 months of one off experiments fossilised into a 200 campaign mess. ACoS is a moving target, search term reports take a week to analyse, and nobody can explain why any particular campaign exists. The first thing we do on a new Amazon engagement is collapse the chaos into a three tier structure.

The three tiers

Tier 1, Discovery

Purpose: find new converting search terms. Structure: broad and phrase match campaigns with auto targeting, low daily budgets, generous bids. We harvest converters weekly into Tier 2 and aggressively negative keyword non converters.

Tier 2, Performance

Purpose: profitably scale proven keywords. Structure: exact match campaigns, one keyword per ad group (or tight SKAGs), bids tuned to a target ACoS. This is where 60 to 70% of profitable spend lives.

Tier 3, Defence

Purpose: own your branded SERP and protect against competitor conquesting. Structure: branded exact match campaigns plus Sponsored Brands and Sponsored Display with product targeting on your own ASINs. Cheap, profitable, non negotiable.

Why the tier model works

• Clear purpose per campaign, no orphans, no overlap.
• Search term flow is one directional: Discovery to Performance.
• Each tier has its own KPI: Discovery = new converters per week. Performance = TACoS. Defence = SOV on branded.
• Budget allocation becomes obvious: protect Defence first, fund Performance, cap Discovery.

The migration

1. Week 1, pull 90 days of search term data. Identify converters by ASIN and migrate them into new Performance campaigns.
2. Week 1, build Defence campaigns: branded exact match, Sponsored Brands video, Sponsored Display ASIN targeting on your own catalogue.
3. Week 2, build Discovery campaigns: auto + broad + phrase per top ASIN, with low caps.
4. Week 2, pause the old campaign mess. Keep them archived for reference.
5. Week 3 onward, weekly harvest cadence: converters out of Discovery into Performance, non converters negative keyworded.

Bid strategy by tier

• Discovery: dynamic bids, up and down. We want Amazon's algorithm finding placements.
• Performance: fixed bids. We've already proven the keyword; we control ACoS via bid, not algorithm.
• Defence: dynamic bids, down only. Cheap and steady.

Common failure modes

• Mixing branded and non branded in the same campaign, blends ACoS and hides truth.
• Skipping the weekly harvest. Converters die in Discovery without it.
• Adding too many keywords per ad group. SKAGs win on Amazon for the same reasons they win on Google.
• Forgetting Sponsored Display. ASIN targeting your own catalogue is the cheapest defence play available.

We've now run this structure on Amazon US, EU, AU and UK across categories from consumer electronics to supplements to apparel. Without exception, the first 90 days look the same: TACoS drops 20 to 40%, organic rank on hero keywords lifts, and the team reclaims hours per week previously spent staring at a 200 campaign spreadsheet.