Amazon Auto Campaign Targeting Groups: Close, Loose, Substitutes, Complements

Amazon Sponsored Products auto campaigns use four targeting groups: close match, loose match, substitutes, and complements. Close and loose match show your ad on search results for related queries, while substitutes and complements show it on product detail pages of similar or related products. You can set a different bid for each group or turn any of them off.

Most sellers create an auto campaign, set one default bid, and never look at the groups again. That leaves money on the table, because the four groups often perform very differently.

What each targeting group does

Amazon decides what counts as close, loose, similar, or complementary based on your listing and shopper behavior. That's why a well-written title, bullets, and backend terms make your auto campaign more accurate. See backend search terms vs PPC keywords.

How the groups usually perform

Results vary by product and category, so treat these as tendencies, not rules:

• Close match is often the most efficient group because the searches are highly relevant. It is usually your best source of keywords to harvest.
• Loose match finds broader searches and tends to produce more irrelevant traffic. It can still find good terms but often needs more negatives.
• Substitutes put you next to direct competitors. This works well when your price, rating, and reviews compare favorably, and poorly when they don't.
• Complements reach shoppers who are buying something else. Conversion can be lower, but some products (accessories, consumables) do very well here.

How to set bids by targeting group

When you create an auto campaign, choose 'Set bids by targeting group' instead of a single default bid. For an existing campaign, open the ad group's targeting tab to edit each group's bid. A practical starting point is to bid close match highest and the other three lower, then adjust after two weeks of data.

Worked example

You sell a $35 French press with a 30% target ACoS. After 30 days with a single $0.80 bid across all groups, the targeting report shows:

The campaign as a whole looks close to target, but that average hides a strong close match group and a weak complements group. A useful formula for each group: new bid = current average CPC x (target ACoS / actual ACoS), capped at around a 30% change per step.

Reading the groups alongside placements

Targeting groups and placements are related but not the same. Close and loose match mostly produce search results traffic, which shows up as top of search and rest of search in the placement report. Substitutes and complements mostly produce product page traffic. If your auto campaign has a top-of-search placement adjustment, it affects close and loose match more than the product page groups. That's one reason to keep placement adjustments modest on auto campaigns: you would be paying a premium on your broadest search traffic. Raise placement adjustments in your exact match campaigns instead, where the search terms are proven.

A two-week adjustment routine

1. Pull the targeting report for the last 30 days (or since launch if shorter).
2. For each group with at least 2 orders, apply the bid formula above.
3. For each group with spend above 1.5x your target cost per order and zero orders, cut the bid 30%.
4. Check the search term report and add negatives for irrelevant terms and ASINs.
5. Harvest converting search terms and ASINs into manual campaigns.
6. Write down every bid change with the date so you can judge it in two weeks.

Two weeks is a reasonable gap between bid changes on auto groups, because they need time to collect enough clicks for the new bid to show a result. Changing bids every two days mostly creates noise.

When to pause a targeting group

• It has spent more than 2x your target cost per order over 30 to 60 days with very few sales.
• You've already lowered the bid twice and ACoS is still far above target.
• The group keeps matching irrelevant traffic even after you've added negatives.

Pausing is reversible. If you pause loose match and later launch a new variation or improve your listing, try it again at a low bid.

Use negatives inside auto campaigns

Auto campaigns accept negative exact and negative phrase keywords, plus negative ASINs and brands. Use them to block irrelevant searches from close and loose match, and to stop substitutes and complements from showing on product pages that don't convert. The weekly process is in negative keywords on Amazon PPC.

Why your listing shapes every group

Auto campaigns have no keywords, so Amazon relies on your product's content and category to decide what's close, loose, similar, or complementary. If your title says 'coffee maker' but not 'French press', close match may lean toward drip coffee searches. If your product is in the wrong browse category, substitutes can place you next to products that have nothing to do with yours. Before blaming the targeting groups for poor results, check that your title, bullets, backend search terms, and category describe the product accurately. Small listing fixes often clean up auto traffic more than any bid change.

One auto campaign or several?

Some advertisers split each targeting group into its own auto campaign so each gets its own budget. That gives tighter control over spend, especially when one group would otherwise eat the whole budget. For a single product with modest spend, one auto campaign with separate bids per group is usually enough. Either way, harvest winners regularly into manual campaigns as described in how to harvest keywords from auto campaigns.

Auto vs manual: where the groups fit

Auto targeting groups are best for discovery. Once you find converting searches and ASINs, move them into manual keyword and product targeting campaigns where you can bid on each one. The broader comparison is in Amazon auto vs manual campaigns.

FAQ