SKAG vs STAG
In short: Both are rules for how many keywords share one ad group in a Search campaign. SKAG puts exactly one keyword in its own ad group so the ad copy can echo that single phrase. STAG groups a handful of closely related keywords under one shared intent, one set of ads, and one landing page. SKAG was built for an era when exact match meant exact; close variant matching quietly broke that premise. If you are building or fixing an account today, start with STAG and treat SKAG as something to migrate away from, not toward.
By the AdFlint research team · Fact-checked against current Google and Meta platform behavior · Last reviewed July 2026
SKAG
A legacy Google Ads structure putting one keyword in its own ad group so every query maps to tightly matched copy.
Single keyword ad groups gave advertisers manual control over message match back when exact match meant exact. Close variants ended that: the same query can now enter several ad groups anyway, and splitting keywords fragments conversion data that Smart Bidding needs. Treat SKAGs as history rather than a plan. Accounts still built this way usually show hundreds of low-volume ad groups, none with enough data to bid intelligently.
Full definitionSTAG
Groups keywords that share one intent into a single ad group, so the ads and landing page can address that theme.
Single theme ad groups sit between SKAG fragmentation and full consolidation: a handful of semantically close keywords, one set of responsive search ads, one landing page. Most modern Search accounts land here because it preserves relevance while pooling enough conversions for automated bidding. The mistake is drawing themes too narrowly, which recreates SKAGs under a new name and starves each ad group of signal.
Full definitionSide by side.
The differences that actually change what happens in your account.
| SKAG | STAG | |
|---|---|---|
| Keywords per ad group | Exactly one keyword, plus whatever close variants Google now matches to it. | A handful of keywords, typically single digits, that share one search intent. |
| Why it existed | Built when exact match meant exact, so the single keyword and the ad copy could match word for word. | Built as the replacement once close variant matching made keyword-level message match unreliable. |
| Ad group count in a real account | Explodes - hundreds of ad groups for a modest keyword list. | Stays proportional to the number of real intents, a fraction of the SKAG count. |
| Conversion data per ad group | Thin - each ad group only sees the volume that one keyword drives. | Pooled across the keywords in the theme, giving Smart Bidding more signal per ad group. |
| Negative keyword upkeep | Heavy - needs cross-ad-group negatives so overlapping keywords don't compete against each other. | Lighter - negatives mostly separate one theme from another, not keyword from keyword within a theme. |
| Landing page mapping | One page per keyword, sometimes with dynamic text insertion to fake a 1:1 match. | One page per theme, written to answer the shared intent rather than one exact phrase. |
| Where it breaks | Query overlap from close variants routes traffic into the 'wrong' SKAG anyway, undermining the premise. | Themes drawn too narrowly quietly turn back into SKAGs and reintroduce the same data-thinning problem. |
What actually separates them.
SKAG isolates one keyword per ad group so ad copy can echo a single phrase; STAG accepts several related keywords per ad group and writes copy to the shared intent instead of any one phrase.
Close variant matching means a SKAG's 'exact' keyword also fires on paraphrases and reordered words, so the query can legally land in a different SKAG than the one built for it - STAG absorbs that variation inside the theme instead of fighting it.
STAG concentrates conversions across its keyword set, giving Smart Bidding a denser signal per ad group than the same keywords would produce split into individual SKAGs.
Negative keyword lists in a SKAG account exist mainly to stop sibling ad groups from cannibalizing each other's traffic; in a STAG account negatives exist mainly to separate one theme from a different theme.
Rewriting an ad group's copy in a SKAG account means editing one keyword's worth of ads; in a STAG account it means editing copy that has to satisfy every keyword sharing that theme, which takes more judgment but produces fewer ad variants to maintain.
Which one should you use?
Use SKAG when
- You inherited an account built years ago and are deciding whether to keep the existing SKAG structure rather than rebuild it from scratch.
- You are auditing an underperforming account and SKAGs are the reason ad groups have too little data for Smart Bidding to work.
- You need to isolate a single brand or trademark term for legal or bid-control reasons and are willing to accept the reporting cost for that one keyword.
- You are studying how an older account was built to explain historical performance before a restructure.
Use STAG when
- You are building a new Search campaign today and need a starting structure for grouping keywords.
- You are consolidating an old SKAG account and want the smallest change that fixes the data-thinning problem.
- You want each ad group to have enough conversion volume for Smart Bidding to optimize against.
- You are writing responsive search ads and want one coherent message per ad group instead of one message per keyword.
- You need ad groups small enough that the landing page still feels relevant to every keyword inside it.
Common questions.
Are SKAGs still worth building in 2026?
No, not for new campaigns. Close variant matching broke the message-match premise SKAGs depended on, and splitting keywords this finely starves Smart Bidding of the pooled conversion data it needs. If you're starting fresh, build STAGs instead.
I have an old SKAG account that performs fine. Should I rebuild it?
Not urgently, but restructure at the next real opportunity - a new campaign, a product update, or a bid strategy change. Consolidating SKAGs into themed ad groups usually improves Smart Bidding performance because it stops splitting the same conversion history a dozen ways.
How many keywords should a STAG actually contain?
There is no fixed number - the test is whether one piece of ad copy and one landing page can honestly answer every keyword in the group. Once you are stretching the copy to cover a keyword that doesn't fit, split it into its own theme rather than forcing it in.
Does STAG still work with broad match?
Yes, and it's the more common pairing now. Broad match plus Smart Bidding plus a themed ad group lets the algorithm find query variations you would never have listed manually, while the shared theme keeps ad copy relevant to what does show up.
Or stop choosing between them.
AdFlint picks the setting, writes the ads, and keeps optimizing inside the Google and Meta accounts you already own.
Related comparisons
- Hagakure Structure vs SKAG
- Consolidated Account Structure vs SKAG
- Granular Account Structure vs SKAG
- Consolidated Account Structure vs STAG
- Granular Account Structure vs STAG
- Consolidated Account Structure vs Hagakure Structure
- Granular Account Structure vs Hagakure Structure
- Consolidated Account Structure vs Granular Account Structure