Most Amazon PPC problems are not bid problems. They are structure problems.
Sellers spend hours adjusting bids on campaigns that are built in a way that makes good performance impossible. The bids are not the issue. The foundation is.
Here is how to build it correctly.
The single most important rule
One product, one campaign. Or close to it.
Mixing products with different margins, different price points, and different conversion rates into the same campaign means you cannot set a meaningful ACoS target. The campaign's performance is an average of everything inside it, and averages hide what is actually happening.
A $14.99 item and a $89.99 item in the same campaign. The $89.99 item has a 50 percent margin and can tolerate 40 percent ACoS. The $14.99 item has a 20 percent margin and needs to stay under 15 percent ACoS to be profitable. Amazon optimizes the campaign as a unit. You have no visibility into what is happening at the product level unless you dig into placement and ASIN reports. And you cannot set appropriate bids for two products with completely different economics.
Separate them. It is more campaigns to manage, yes. It is also the only way to actually know what is working.
Auto before manual
If you are launching a new product and have no search term data, start with an auto campaign at a modest daily budget for three to four weeks. Let Amazon spend your money discovering which search terms your customers actually use.
At the end of that period, pull the search term report. Find the terms with two or more orders and low ACoS. Those go into a manual exact match campaign. Find the terms with clicks but no orders. Those become negatives in the auto campaign.
This is not a new idea. Most experienced sellers do it. The mistake is skipping it on new products because it feels slow, then building manual campaigns targeting terms that turn out to be irrelevant for your actual customer.
The three-campaign structure for established products
For an established product with some performance history, a basic structure looks like this:
Campaign 1: Manual Exact. Your proven, high-converting keywords. Tight control. Bids set based on your actual ACoS data for each term. No broad match, no phrase match. Exact only so you know exactly what you are paying for.
Campaign 2: Manual Phrase. Mid-funnel terms where you want visibility on variations but still have some control. Broader than exact, but not the free-for-all of auto.
Campaign 3: Auto. Still running in the background, at a lower budget than it ran during launch. Its job now is to catch new search terms you have not thought of and feed them into your exact match campaign when they prove out.
This structure is not complicated. The problem is most sellers never get here because they launch an auto campaign, find some terms that work, build a manual campaign targeting them, and stop. They end up with a loose collection of campaigns that all touch the same keywords, compete against each other in the same auction, and have no clear division of purpose.
Match types are not interchangeable
Broad match on Amazon is aggressive. It will match your keyword to searches that are adjacent, tangentially related, or sometimes just loosely related. The click volume is high. The conversion rate is often low. The search terms you end up paying for can be genuinely surprising.
Broad match is a discovery tool, not a targeting tool. Use it with a defined budget for finding new terms, add negatives regularly, and harvest what converts into exact match. If you are running broad match campaigns that you set up six months ago and have never added a negative keyword to, there is almost certainly waste in there.
Exact match is for terms you know convert. It costs more per impression because you win a narrower set of auctions. It produces more predictable ACoS because you know exactly what you are bidding on.
Most sellers have too much broad, not enough exact. The fix is pulling the search term report, sorted by spend, descending, and moving anything that has converted more than twice into exact match campaigns.
What to do if your structure is already a mess
You have 40 campaigns. Half of them are auto campaigns you launched at different times. The other half are manual campaigns targeting the same keywords with different bids. Nobody remembers why three of them exist.
Do not try to fix it all at once. Start by identifying your top five products by revenue and build clean structures for those. Pause the old campaigns for those products, let the new structure run for 30 days, compare performance.
If the new structure performs better, and it usually does, extend the same process to the next five products. This takes months if your catalog is large. That is fine. The mistake is waiting until you have time to overhaul everything before you start.
Scheduling and budget rules on top of structure
Good campaign structure is the foundation. Dayparting and budget rules are the layer on top that makes the structure perform at its best.
A well-structured campaign running 24 hours a day, spending budget overnight when nobody is buying, is still wasting money. Structure tells Amazon what to show and to whom. Scheduling tells it when. Both matter.
Agencies managing multiple client accounts build structure standards first, then apply scheduling on top systematically across accounts. The ones who skip the structure work and go straight to scheduling optimization find it only helps so much when the underlying campaigns are built wrong.
Off Hours adds automated scheduling and budget control on top of your campaign structure. Start a 14-day free trial and have your first dayparting rule running in under five minutes.