Introduction
Keep your Amazon automatic campaigns running, but as a low-bid discovery engine, not as the campaign that owns your budget. In established accounts, we keep auto at roughly 5% to 15% of ad spend on low bids, move its winning search terms into manual exact campaigns every week, and put the rest of the budget behind keywords that already convert.
We manage over $100M in annual client revenue across 50+ brands, and the examples below come from those accounts and audits on our client calls. Here’s what an Amazon automatic campaign does differently from manual, the four targeting groups, a short harvest routine, and when to cut back.
Should you still run Amazon auto campaigns?
Yes. An auto campaign isn’t a beginner setting to switch off or something to leave on the same bid forever. What makes it essential to your Amazon PPC strategy is that it finds search terms and products you haven’t thought of and buys clicks you would never bid on manually.
How much of your budget should auto campaigns get?
Auto campaigns need a low bid and small share. This is because auto lets Amazon pick the searches and placements and so a big auto budget is a budget you don't control. Also, when auto takes most of the budget, spend drifts away from top-of-search placement and the keywords that are already converting.
In established accounts that we manage, auto gets roughly 5% to 15% of the budget on low bids. New listings with no data can lean on auto more at the start, then move toward this range as winners graduate to exact.
Most of the remaining budget goes to proven, exact match keywords, with product targeting and branded campaigns on their own lines.
For the full split by brand stage, see our Amazon PPC budget allocation guide.
Amazon automatic campaigns vs manual: What does auto do differently?
The difference between auto and manual campaigns is who picks the targets. In an auto campaign, Amazon uses prior shopping queries and product info to automatically match your ads with keywords and products similar to the item you’re advertising.
By contrast, in a manual campaign, you pick keywords by match type (exact, phrase, or broad), or specific products and categories.
Read the table below as a division of labor: auto finds terms; manual scales proven ones. Amazon's own guide to Sponsored Products says the same. Use the auto campaign's search term report to pick the best keywords and products for your manual campaigns, and run both together.
| Factor | Auto campaigns | Manual campaigns |
|---|---|---|
| Who picks targets | Amazon | You |
| Bid control | One bid per targeting group, plus negatives | One bid per keyword or product |
| Best use | Finding new search terms and products | Scaling proven terms at set bids |
| Budget share (our method for established accounts) | Roughly 5% to 15%; low bids | The rest |
How do you set up the four auto targeting groups?
Amazon lists four automatic targeting groups:
- Close match
- Loose match
- Substitutes
- Complements
Close and loose match work on shopper searches. Substitutes and complements show your ad on product detail pages. You can set one default bid for the whole campaign or a bid for each group.
| Targeting group | Where your ad may show | How we start bids |
|---|---|---|
| Close match | Searches closely related to your product | Highest of the four, since intent is closest |
| Loose match | Searches loosely related to your product | Below close match, since the net is widest |
| Substitutes | Detail pages of products similar to yours | Near close match if your price and reviews win the side by side |
| Complements | Detail pages of products that pair with yours | Lowest, then raise it if conversion holds |
Pro tip: Set bids by group and keep every group's bid low, because one default bid lets a weak group spend like a strong one. The job of auto is to find terms cheaply, not to win expensive clicks.
Why a low bid matters: Examples from our accounts
Our September 2026 data across 66 Amazon ad accounts reveals that auto and product targeting together took 44% of the spend at 31% advertising cost of sales (ACoS). This figure combines auto with manual product targeting (competitor and category pages). As such, it doesn’t show what auto alone earns, and it isn’t a reason to give auto a big share.
To illustrate the importance of keeping your bids low, consider the following examples from our accounts:
When a consumer tech brand approached us, our audit of its account revealed that most of its German ad budget was sitting in auto campaigns, at a cost per order that Alex Stoykov, our founder, estimated at about $1,000. Alex’s advice wasn’t to switch auto off, but rather drop the bid from about $2 per click to $0.70. This way, Amazon could serve cheaper traffic instead of premium traffic and allow the campaigns to continue finding terms without taking a big slice of the budget.
For a kitchen tools brand we manage, a low-bid auto campaign produced $1,251 in sales from $170 in spend at 13.6% ACoS in the week of 28 September to 4 October 2026. It outperformed a scaled exact-match campaign on another product in the same account which ran at 87.8% ACoS.
Low bids are why it worked. The auto campaign only won cheap clicks, and it was one of the best returning lines in the account that week.
One caution from the same audit: auto suits cheaper, simpler products better. For an expensive product that shoppers think about for weeks, auto can buy a lot of clicks before it finds a buyer. For this reason, keep its share at the low end and lean on exact match and competitor pages.
How do you harvest auto search terms into manual campaigns?
Once a week, move converting auto search terms into manual exact campaigns. Then, add each one to auto as a negative exact so the two campaigns don't bid against each other.
Our data shows why this pays. In a typical account, about half of the converting search terms generate 80% of the sales, and the 10 biggest terms take 28% of the spend. Those few terms deserve their own bid in exact, not a shared auto bid.
On the other hand, our September 2026 data across 66 Amazon ad accounts reveals that search terms with 10+ clicks, yet no sale, took only about 5% of the spend in a typical account. In this case, you should add negatives for any of these auto search terms with 10+ clicks and no sale.
For the full weekly routine and thresholds, follow our Amazon keyword harvesting process. Then, for advice reading the report itself, here’s how to analyze an Amazon search term report.
How should you use negative keywords in auto campaigns?
Negative targeting works in both automatic and manual campaigns and so auto needs the same negative routine as your manual campaigns. This routine involves:
- Using negative exact for single wasted terms and negative phrase for a word that’s wrong in every search that contains it
- Blocking irrelevant products from the substitutes and complements groups
- Leaving terms with only a few clicks alone, because cutting too early starves discovery
When should you cut back an auto campaign's budget?
Only in rare cases should you turn auto off entirely. An example of such an instance is when a high-price product keeps buying clicks that don't convert even at a low bid.
The table below lists examples of when it will be better to reduce the budget instead of cutting it off completely:
| Trigger | What to do |
|---|---|
| Auto drifts above 15% of spend | Lower the budget or the group bids until it sits back in its discovery range |
| Auto ACoS runs well above your exact match ACoS for a month | Lower the bids first and if it still doesn't come down, lower the budget |
| Few new converting terms show up | When manual campaigns already cover what auto finds, keep a smaller auto campaign running for discovery |
| Wasted spend climbs | If terms with 10+ clicks and no sale take more than about 5% of auto spend, tighten negatives and group bids |
Pro tip: Before you raise bids instead, remember Amazon's dynamic bids come as "up and down" or "down only". Down only is the safer choice for auto groups that haven't proven themselves.
Frequently asked questions
Do auto campaigns still work in 2026?
Yes, auto campaigns are still effective as discovery on a low bid because they keep finding search terms and products your manual campaigns miss. Keep them at roughly 5 to 15% of budget in an established account, harvest weekly, and let exact match carry the rest.
Should new listings start with auto only?
No. Launch with auto for discovery plus a manual exact campaign on keywords you already expect to convert. Auto collects search term data, while exact protects your core terms. Next, harvest the converting terms into exact-match campaigns every week.
How often should I review the auto campaign budget share?
Harvest weekly and review the share monthly. Check that auto still sits in its range and that its ACoS isn't running far above your exact-match campaigns.
What is a good ACoS for an auto campaign?
Any figure below your break-even ACoS (in other words, the ACoS at which ad spend uses up your full profit margin). With low bids, auto can run well below exact match, as in the kitchen tools example above. If it runs far above exact, lower the bids first.
What to do with your auto campaigns this week
Auto campaigns tend to get set up and forgotten. Here’s what’s worth a second look this week. A few small adjustments to your auto campaigns now can make a noticeable difference.
- Pull your campaign report and check what share of spend sits in auto versus manual
- If auto is above roughly 15% of spend, lower its group bids before anything else
- Compare your auto campaign ACoS with your exact and broad match ACoS
- Run a search term report on your auto campaigns and harvest converting terms into exact
- Add negatives for any auto search term with 10+ clicks and no sale
If your auto campaigns are spending your budget instead of finding new terms, our Amazon PPC management team can rebuild them. Our specialists have 7+ years of experience, work with our in-house AI platform, Olifant AI, and build accounts with our 1-1-1-1 Scaling Method. If performance does not improve in 60 days, we refund our management fees.
.webp)
Alex is the founder and CEO of Olifant Digital, where his team manages over $100M in annual client revenue across 50+ e-commerce brands. He sets the agency's strategy and methodology, audits client accounts, and reviews the team's work to make sure every brand is on track to hit its growth and profitability goals. He also leads team training, so the performance data and lessons from across all accounts shape how every account manager runs their clients.
.png)
Mike reviews every Amazon article on this blog for strategic and technical accuracy before it publishes. As Director of Amazon Growth at Olifant Digital, he sets marketing strategy across client accounts and personally audits PPC at every stage of growth. He brings 8 years of daily Amazon operations across 7 and 8-figure brands including Beauty by Earth, Ekster, and Bullstrap, the kind of hands-on depth most agency directors delegate away.
