Single Keyword Ad Groups on Amazon: When SKAGs Help and When They Hurt
A single keyword ad group (SKAG) holds exactly one keyword in one match type, usually exact. On Amazon, SKAGs make sense for your top revenue keywords, where precise bid and budget control pays off, and rarely for long-tail terms, where the extra management time outweighs the benefit.
Below you'll see how SKAGs work on Amazon specifically (it differs from Google Ads), when to use a single keyword campaign instead, and a decision rule for which keywords deserve isolation.
How SKAGs work on Amazon
On Google Ads, SKAGs were popular because each ad group could have its own ad copy matched to the keyword. Sponsored Products has no ad copy: the ad is built from your listing. So on Amazon the benefit of a SKAG is purely about control and data clarity, not message match.
What you gain:
- Clean reporting: every impression, click and order in that ad group came from one keyword.
- Precise negatives: you can add negatives at the ad group level without touching other keywords.
- Easy bid testing: changing the bid affects one target only.
What you don't gain: a separate budget. Budgets and placement adjustments live at the campaign level. If you put ten SKAGs inside one campaign, they still compete for one daily budget. That's why many Amazon advertisers go one step further to single keyword campaigns for their top terms.
SKAG vs single keyword campaign
| Setup | Own bid | Own negatives | Own budget | Own placement adjustments |
|---|---|---|---|---|
| Many keywords in one ad group | Yes (per keyword) | Shared | Shared | Shared |
| SKAGs inside one campaign | Yes | Yes | Shared | Shared |
| Single keyword campaign | Yes | Yes | Yes | Yes |
Single keyword campaigns make the most sense for your hero term, the one keyword that drives a large share of sales and where top-of-search placement matters. You can give it a top-of-search adjustment without inflating bids on everything else. Learn how those adjustments work in Amazon placement bid adjustments.
When SKAGs help
- The keyword drives meaningful spend: as a rough rule, 30+ clicks a week or a noticeable share of product sales.
- You need to defend or win a specific placement for it.
- Its ACoS behaves very differently from the other terms in its ad group.
- You want to run a clean bid test (for example, raising the bid 15% for two weeks and comparing).
When SKAGs hurt
- Low-volume terms: a keyword with 5 clicks a month gives you almost no signal, isolated or not.
- Management overload: 300 SKAGs means 300 bids to review. If you can't review them, the structure is worse than a simpler one.
- Thin data per entity makes automated bidding tools less effective, because each target has too little history.
- Duplicate traffic: if you forget to negate the SKAG's keyword elsewhere, your own campaigns can end up competing for the same search term.
A decision rule for isolation
Rank your exact-match keywords by sales over the last 30 to 60 days. Then:
- Top 3 to 5 keywords (for a single product these often carry a large share of ad sales): consider single keyword campaigns.
- Next 10 to 20 keywords with steady orders: SKAGs or small themed ad groups (3 to 5 keywords) inside one exact campaign.
- Everything else: themed ad groups of 10 to 20 keywords, managed with bid rules.
Revisit this list monthly. Keywords move up and down, and your structure should follow the data rather than the other way around.
Worked example
A seller of a cast iron skillet has an exact campaign with 25 keywords and a $50 daily budget. Over 30 days, 'cast iron skillet' accounts for $3,000 of $6,500 in ad sales but runs at 38% ACoS against a 30% target, while the other 24 keywords average 24% ACoS.
Because they share a budget, the hero term often uses up the money by mid-afternoon, so the efficient long-tail terms lose impressions late in the day. The fix:
- Move 'cast iron skillet' into its own exact campaign with a $35 daily budget.
- Add 'cast iron skillet' as a negative exact in the original exact campaign (and in auto and broad campaigns).
- Drop its bid about 10% and test a modest top-of-search adjustment, watching whether conversion rate holds.
- Keep the original campaign at $30 a day for the remaining 24 keywords.
Two weeks later, compare: did the hero term's ACoS move toward 30%, and did the long-tail campaign stop capping out? If both improved, keep the split. If the hero term lost too much volume, raise its bid in 5% steps. You can work out a starting bid with the PPC bid calculator.
Negatives keep SKAGs clean
Isolation only works if traffic actually goes where you want it. Every time you create a SKAG or single keyword campaign, add that exact term as a negative exact in every other campaign that could catch it: auto, broad, phrase and other exact groups. Without that step, multiple campaigns may serve for the same search, and you lose the clean data you wanted. Amazon PPC keyword isolation explains the full negative map.
How to set up a single keyword campaign in the console
- In the Amazon Ads console, create a new Sponsored Products campaign with manual targeting.
- Name it clearly, for example 'SP | Skillet | KW | Exact | Hero', so you can find it later.
- Set a daily budget sized to the keyword's recent spend plus some headroom (if it spent $25 a day before, start at $30 to $35).
- Choose a bidding strategy. 'Dynamic bids - down only' is the safest starting point; test 'up and down' later if conversion rate is strong.
- Create one ad group, add the products that should show for the term, and add the single keyword in exact match.
- Set the bid. A good default is the keyword's previous bid, adjusted toward your target ACoS based on its last 30 days.
- Go to your other campaigns and add the keyword as a negative exact everywhere it could otherwise match.
After launch, don't touch the bid for at least 7 days unless spend is clearly out of control. Attribution takes a few days to settle, and early data on a single keyword can swing a lot from day to day.
Common SKAG mistakes
- Isolating a term but leaving it active in the original ad group, so two of your own bids enter the same auction and your reports split the data.
- Creating SKAGs for keywords with no history, based only on search volume estimates. Prove the term first in a broad or exact group, then isolate.
- Copying the old bid without checking placement. If most conversions came from top of search, the isolated campaign may need a placement adjustment to keep them.
- Never consolidating. When a hero term fades, fold it back into a themed ad group instead of keeping a near-empty campaign alive.
Do SKAGs still make sense in 2026?
Yes, selectively. Automated bidding works best with enough data per target, which argues against isolating everything. The modern approach is a hybrid: isolate the few terms that matter most and group the rest. For the bigger picture of how campaigns, ad groups and keywords fit together, read Amazon PPC campaign structure and ad group best practices.
If maintaining isolated keywords and their negatives by hand feels like too much, AdSpark's keyword isolation feature handles the moves and negative placements, and you can reverse any change it makes.
Isolate your top keywords without the spreadsheet work.
Start your 14-day free trialFAQ
Do SKAGs lower ACoS on Amazon?
Not by themselves. They give you better control and data, which can lead to better bid decisions and lower ACoS. Isolating a keyword without adjusting its bid or negatives changes very little.
Should I use exact match for SKAGs?
Usually yes. Exact match gives the tightest control. Some sellers also run a phrase-match SKAG for a key root term to collect variants, but that is a research setup, not a performance one.
How many keywords should be in an Amazon ad group?
For most themed ad groups, 5 to 20 related keywords is manageable. Your top terms can live alone. See ad group best practices for details.
Can I put multiple products in a SKAG?
Yes, if they are near-identical variations that all deserve the keyword, such as sizes or colors. Unrelated products should not share a keyword ad group.