Okay, let’s be real—who’s tired of watching CPCs go up and ACOS stay stuck? I was there for weeks. Tweaked bids, optimized listings, cut bad keywords… nothing moved the needle. My core campaigns were stuck at 35% ACOS, and I was about to throw in the towel.

Then I started messing around with AMC (Amazon Marketing Cloud) audience bidding on 7 of my SKUs. Holy crap—18% average ACOS drop across the board, and I didn’t even increase my ad spend. Figured I’d share exactly what worked for me, since most sellers I talk to still aren’t using this right.

First, why AMC audiences are better than third-party tools: Before, I only used pre-built audiences from my ERP. Now Amazon lets you build custom ones directly in Seller Central, and they’re way more accurate. No more generic audiences that don’t hit your actual buyers.

The audience I use 90% of the time? Shoppers who searched my core keywords in the last 30 days but didn’t buy. These aren’t cold leads—they already showed intent. They either scrolled past my listing, clicked but didn’t pull the trigger, or added to cart and bailed. Retargeting them with AMC is a game-changer.

Here are the 3 strategies that actually worked for me (all for Sponsored Products, since that’s where most of my budget goes):

  1. SKAG Campaigns + Matching Keyword Audiences

Build an AMC audience for your exact SKAG keyword, add a 15–25% bid adjustment. You’re basically bidding more for people who already searched that term—no wasting money on cold traffic. My result? ACOS went from 37% to 19%, conversion rate up 41% in two weeks.

  1. Auto Campaigns + Segmented Keyword Audiences

Split your Auto campaigns into close match and substitute match. Add 10–30% bid adjustments for high-intent keyword audiences (focus on the keywords that already drive relevant traffic). I had a substitute match Auto campaign with 47% ACOS and 1.2% CVR—after adding the adjustment, ACOS dropped to 24% and CVR hit 2.8% next cycle.

  1. Competitor ASIN Targeting + High-Intent Audiences

If you target competitor ASINs, layer in an AMC audience of people who searched your core keywords. That way, you’re only bidding more for shoppers who are actually looking for your type of product—not just browsing a competitor’s page. I saw 26% lower ACOS than regular competitor targeting.

Bonus tip: Brand audiences + cart abandonment promotions. Build an audience of people who searched your brand keywords, pair it with a cart abandonment deal in SB or SD. These people already know your brand—they just need a nudge. My CVR hit 8% here, which is 3x higher than my standard brand ads.

For high-ticket sellers ($100+ products): Build a custom audience for people with $100k+ household income. You’ll need basic SQL to do it yourself, or a DSP provider if you don’t. Way better conversion quality than generic audiences.

Couple quick mistakes I made at first—don’t repeat them: You need at least 2,000 users for an AMC audience. If it fails, add more ASINs or higher-volume keywords. And test small first—$20–$50/day for 7–14 days. Results vary by category, so don’t scale until you see positive numbers.

My current setup that’s working best: One cold campaign (no audiences) to reach new people, and a duplicate campaign with AMC bid adjustments to retarget high-intent shoppers who didn’t convert. Clean data, no overlap, easy to scale.

Alright, enough from me. Have any of you tested AMC audiences? What worked, what flopped? Drop your results below—I’m always down to compare notes and tweak strategies together.