Automatic and Manual Amazon Campaigns: How They Work Together

The question is usually posed as a choice, and it is not one. Automatic and manual campaigns do different jobs, and an account running only one of them is either not discovering anything or not controlling anything.

You give Amazon a product and a budget, and it decides which searches to show it against, using the listing and whatever it knows about the category. That sounds like laziness and it is the opposite. It is research. Real shoppers use words nobody in your business would write down, and the search term report from an automatic campaign is the only place those words appear before you are paying for them deliberately. What it cannot do is set a sensible price. One bid covers everything it decides to match, so a term worth two dollars and a term worth twenty cents are bought at the same price, and one of the two is always wrong.

A term you have chosen, at a bid you have chosen. That is the whole of it, and it is the only way a good term gets the money it deserves. The cost is that you can only pay for words you already know. A manual only account is a permanent record of what somebody assumed in the first week, gradually going stale as the way people search changes around it.

The loop is short and most accounts run about half of it. The automatic campaign discovers a term. It gets enough clicks to be judged. If it converts, it goes into a manual campaign as exact match, with its own bid set on what it is actually worth. And then, the step that gets missed: add it as a negative exact in the automatic campaign. Otherwise both campaigns keep bidding on it, you are competing with yourself, and you pay more for the same sale while the reporting splits across two lines and understates both. Terms that do not convert become negatives too. The mechanics of doing that without overreaching are in negative keywords, the cheapest work in Amazon PPC.

Enough to keep finding things. Once the good terms have been promoted, what remains is mostly the waste, so an automatic campaign left to grow becomes an expensive way to buy the matches you rejected. Treat it as a standing research budget: small, steady, reviewed weekly, never the largest line in the account. The moment it becomes the largest line, something has stopped being read.

Automatic campaigns split into close match, loose match, substitutes and complements. They behave differently enough to be worth separating: substitutes and complements target other products' pages rather than searches, and they usually want a lower bid than the two search groups. Splitting them is ten minutes of setup and it stops one expensive group hiding inside an average.

The handover only works if somebody reads the search term report on a schedule, which is the first item in the weekly optimization routine. When it stops happening, an automatic campaign does not fail loudly. It just keeps spending on terms that were promoted months ago. If the reading never happens because there is never time, that is a capacity problem rather than a discipline one, and seven clear signs you have outgrown manual management is the checklist for it. Which campaign types to run alongside these is covered in Amazon ad types and when each one earns its budget.

Both campaign types bid into the same auction, and how Amazon pay per click works is the shortest description of what they are bidding into. Keeping the handover running week after week is part of Amazon advertising management: it is the least interesting work in the account and the first thing to lapse.