Amazon auto vs manual campaigns: what 65 ad accounts show

Key findings

• Product targeting converts best. Manual ASIN and category targets converted a median 12.9% of clicks across 37 accounts. Manual keywords converted 8.7% across 62 accounts, and auto campaigns converted 5.9% across 57 accounts.
• Auto clicks are cheap. When we compared the two inside the same account, an auto click cost a median 36% less than a manual keyword click (54 accounts).
• Cheap clicks are what make auto more efficient. In the accounts that ran both auto and manual keyword campaigns, auto had a lower ACoS than manual keywords in 59% of them, which is 32 of 54 accounts.
• Manual keywords still carry most of the budget. The median account put 60% of its Sponsored Products spend on keywords, 24% on auto and only 8% on product targets.
• Not all auto targeting is equal. Inside the auto campaigns, close match converted 8.1% of clicks, while substitutes converted only 3.5%.

Auto vs manual: which converts better?

Manual targeting converts better than auto, and product targeting converts the best of all three. In the table below, you can see the median account for each targeting type.

We only counted a targeting type in an account if it got at least 100 clicks during the month so that a handful of clicks could not distort the result.

In the table above, conversion rate means the number of orders divided by clicks, using Amazon’s 7-day attribution. The share of spend shows how the median account split its budget, which is why the three medians do not add up to exactly 100%.

One note on the ACoS column. These are medians across accounts, where a small account counts as much as a large one. Pooled across all US spend in September ($671,757), ACoS was 21.0%, because larger accounts run more efficiently, as our Amazon ads benchmarks show.

For those who wonder why product targeting converts so well, the answer is the following. When you target a specific competitor ASIN, your ad shows up on a page where the shopper is already looking at a product like yours.

Quite often, this product is worse or even more expensive than yours, and the shopper is already close to buying. Auto campaigns, on the other hand, work differently. They test many searches and product pages at the same time, and a lot of them are only loosely related to what you sell.

If you want to understand conversion more broadly, you should take a look at our guide on how to calculate the Amazon conversion rate.

If auto converts worse, why is its ACoS lower?

To put it shortly, auto clicks are far cheaper. ACoS is based on two factors: what you pay per click and how many clicks convert to a sale. Auto loses on the second part, but it wins on the first. And to ensure that a comparison is fair, we compared the targeting types within the same account, eliminating any impact from currency or product category.

In the table, the CPC ratio is the CPC of the first type divided by the CPC of the second type, within the same account.

This means that in most accounts, auto converted worse than keywords, and only 17% of accounts saw auto convert better. However, auto clicks cost about 36% less, which was enough to give auto a lower ACoS in 59% of accounts.

Product targets cost less per click than keywords and also converted better, so they had a lower ACoS in 63% of accounts.

We need to add one honest note here. Manual keyword spend includes the hardest and most competitive terms, which are the ones that every brand in the category bids on. Auto spend, on the other hand, goes wherever Amazon finds cheap traffic.

A lower ACoS for auto doesn’t mean that you should move all of your keyword budget into auto campaigns, but it means that auto is a good place where you can buy cheap, broad traffic, as long as you keep watching it.

For a deeper look at what a “good” ACoS actually is, you should read what ACoS means and our ACoS vs TACoS guide.

Which auto targeting groups work best?

The best part of an auto campaign is close match, whereas substitutes are the worst.

Amazon ends up splitting each of the auto campaigns into four different targeting groups. You can assign a different bid to each group, but many sellers will set all of them to the same bid.

Close match shows your ad on searches that are closely related to your product, and it behaves almost like a good keyword campaign.

Substitutes targeting shows your ad on competitor product pages, and it converts at less than half the rate of close match, but because the clicks are cheap, ACoS ends up only a few points higher.

Complements targeting shows your ads on the pages of products that go well with yours. It gets very little spend in most of the accounts, which is why only 18 accounts reached our 100-click minimum for this group.

How do manual keyword match types compare?

According to our data, phrase match was the most efficient keyword match type. Broad and exact match had similar conversion rates, but both of them had a higher ACoS than phrase.

A lot of sellers get surprised by the fact that exact match has the highest ACoS.

The likely reason for that is that exact match is where the brands put their biggest and most competitive keywords, and these are the ones that cost the most.

Phrase match, by contrast, catches longer and more specific searches which are cheaper and closer to a purchase. If you want to build your campaign structure around this, you can check our post on Amazon PPC campaign architecture. For the full breakdown, including why exact match looks expensive, see our Amazon match types study.

What this means for your brand

The data does not say that “auto is good” or “manual is good” because each targeting type has its own job to do, and here is how we would use these findings.

1. Keep auto running, but keep it small and supervised. In the median account, auto took about a quarter of the spend. That is a one-month snapshot across accounts at very different stages. For an established account, our rule is roughly 5 to 15% of budget in auto, with more during a launch. Auto finds new search terms and ASINs cheaply; however, without negative keywords, it also wastes money on loose traffic. Our guide to Amazon auto campaigns explains how we set that up.
2. Split your auto bids by targeting group. Close match earns a higher bid, whereas substitutes and complements should get lower bids, since they convert at about half the rate.
3. Move the winners from auto into manual campaigns. Every few weeks, you should read through your search term report. The search terms and ASINs that sell in auto belong in manual campaigns, where you control the bid. Our guides on keyword harvesting and reading a search term report show you how to do this.
4. Give product targeting more room. It converted best and won on ACoS in most accounts, yet it took only 8% of the spend in the median account. A good place to start is with competitor ASINs that have a higher price or weaker reviews than yours.
5. Keep an eye on exact match costs. Exact match is not automatically efficient. You need to check whether your top exact keywords will still earn their bids and lower the bids where they do not. Our guide on how to reduce ACoS has more steps you can take.

Keep in mind that your numbers will differ from the median. Use your account history first, and use these figures as a check against it. You can find more of our data studies on the Olifant Digital research page.

How we collected this data

• Source: Amazon Ads Sponsored Products targeting reports, downloaded from the ad accounts that Olifant Digital manages. Sponsored Brands and Sponsored Display are not included.
• Period: September 1 to 30, 2026. Sales and orders use Amazon’s 7-day attribution.
• Sample: 65 ad accounts with Sponsored Products activity during the month, from about 46 brands. There are brands that have several accounts, one for each country, and each account counts once. We left out the accounts that had no data in that period.
• Grouping: Auto means Amazon’s four automatic targeting groups (close match, loose match, substitutes and complements). Manual keyword means broad, phrase and exact keywords. Manual product means ASIN targets and category targets. We left out a small amount of spend on other expressions. Our match types study reports ASIN targets on their own, where the median conversion rate was 11.8%.
• Threshold: A targeting type only counted in an account if it had at least 100 clicks in the month, and this is why “n” is different from row to row.
• Medians, not averages: Every figure is the median across accounts, so one large account cannot skew the result. Comparisons inside the same account use ratios, so currency does not matter.
• Currencies: The accounts report in USD, EUR, GBP, CAD, AUD, SEK and MXN. For this specific reason, we do not publish CPC in dollars, and we only show CPC as a ratio between types in the same account.
• Limitations: This is one month of data from one agency’s accounts across a mix of categories. Manual keyword campaigns include branded terms, which usually convert well and cost less, and we did not split them out here. Attribution is Amazon’s own and it can give ads credit for sales that would have happened anyway.

Frequently asked questions

Are Amazon auto campaigns better than manual campaigns?

Neither one will win on every metric because they do different jobs for your account. Auto is the cheaper way to reach shoppers and find out what they are searching for, whereas manual targeting gives you control over the bid and usually brings buyers who are closer to a purchase.

Auto had the lower ACoS of the two and manual converted better in our accounts.

This distinction is the main reason why most brands that we work with run both at the same time.

How much budget should go to auto campaigns?

The median account we manage put around 24% of its Sponsored Products spend into auto in September 2026. Treat that as an observation across accounts at different stages. For an established account, our rule is roughly 5 to 15% of budget in auto, with weekly harvesting into manual campaigns. Our guide to Amazon PPC budget allocation covers the rest of the split.

The picture changes with a new launch. New products have no sales history and you don’t know yet which searches bring buyers, so it often makes sense to give auto a bigger share in the first weeks and then shift the budget into manual campaigns as the winning terms appear.

Why is my auto campaign conversion rate so low?

In most cases, nothing is wrong. Auto puts your ad on many searches and pages that only loosely relate to your product, so you’re going to get a lower conversion rate.

What you can do is see what targeting group the clicks are coming from. If the majority of your auto spend is on substitutes and complements, the rate is typically increased by decreasing those bids and adding negative keywords for searches that never sell.

Is ASIN targeting worth it on Amazon?

Yes, for most brands it is worth testing. When your ad shows up on a competitor’s product page, the shopper is already viewing something like yours, and that is a powerful moment. It definitely works best if your product has a clear advantage on that page, such as a lower price, better reviews or a bigger pack size.

If the competitor beats you on all of those, you have likely wasted that click.

Should I turn off auto campaigns once I have manual campaigns?

We would not advise this approach. Shopper search evolves, new competitors come online and Amazon keeps launching new products into your category.

An auto campaign will pick up those changes before you notice them, so you can move selling search terms and ASINs into your manual campaigns. Just make sure you keep its budget small and add your manual keywords as negatives in auto so the two campaigns don’t end up bidding against each other.

Which keyword match type has the lowest ACoS?

In our data, phrase match had the lowest ACoS, outperforming both broad and exact match types.

That’s not to say that you should move everything to phrase, because each match type still has a purpose. Broad helps you discover new search terms. Phrase captures the longer searches. Exact lets you control the bid on keywords you know will sell.

Get a second opinion on your auto and manual split

Each brand splits its budget a little differently, and the right mix of auto and manual campaigns depends on your products, your competitors and how long you have been selling on Amazon.

Olifant Digital is a full-service Amazon agency, and one of the first things we look at when we start working on a new account is the balance between auto and manual targeting.

If you want a second opinion on your own setup, read more about the way we approach Sponsored Products management, or get your free marketing plan, and we’ll take a look at where your budget is going.