Should I use single keyword ad groups (SKAGs)?

What SKAGs are (and why they used to work so well)

A single keyword ad group (SKAG) is exactly what it sounds like: one ad group built around one core keyword (often replicated across match types), with ads and a landing page tightly aligned to that term. SKAGs became popular because they made it easier to control relevance, steer traffic, and diagnose performance issues quickly—especially back when match behavior was more literal and ad testing was more “one ad = one message.”

Today, the platform behaves differently. Match types are meaning-based, close variants apply across match types and can’t be disabled, and responsive search ads (RSAs) are designed to mix-and-match multiple assets. On top of that, when multiple keywords could match the same query, the system uses prioritization rules and relevance signals to choose what enters the auction. All of that reduces the “hard control” SKAGs used to provide, while keeping most of the operational overhead.

The real benefits of SKAGs (when they still make sense)

1) Cleaner message-to-query alignment for a small set of critical terms

If you have a handful of high-intent, high-volume queries that deserve highly specific messaging (for example, one service line, one product category, one flagship offer), a SKAG-like setup can still help you keep ad copy and landing page intent extremely focused. When the offer, qualifiers, pricing model, or compliance language differs meaningfully by query, isolating that theme can improve clarity and conversion rate.

2) More controlled experimentation—when you’re testing the landing page, not “keywords”

In mature accounts, the keyword is often not the main lever—the landing page, offer, and lead quality are. SKAGs can be useful as a test harness: you isolate one intent cluster so you can run clean A/B tests on landing pages, forms, pricing presentation, or conversion flows without mixing in multiple intent types that muddy interpretation.

3) Easier diagnostics for a small number of “must-win” auctions

When a term is strategically important (brand defense for a specific product line, a core commercial category, or a key competitive battleground), a dedicated ad group can make it faster to spot issues with ad strength, asset coverage, policy/disapproval risk, budgets, or conversion tracking anomalies.

The drawbacks of SKAGs in modern Google Ads (and why they often underperform expectations)

1) Close variants and meaning-based matching reduce the “purity” of a SKAG

Even if you build the perfect one-keyword ad group, that keyword can match to searches that are similar in meaning or intent, not just identical strings. That means a SKAG can still attract a range of queries—so the promise of “one keyword = one query set” is no longer reliable. The practical outcome is that you may still need negatives, and you may still see query blending across what you thought were perfectly separated ad groups.

2) Duplicates and “very similar keywords” don’t buy you extra control

Building many near-identical SKAGs (plural/singular, reordered words, close phrasing) usually creates administrative weight without creating additional leverage. When similar keywords are eligible for the same search, only one can trigger an ad for that search. So the account may behave like you have lots of knobs, but you’re often just moving complexity around while the system selects one eligible path based on priority and ad rank.

3) Ad group prioritization and relevance signals can override your intended routing

If multiple ad groups can match the same query, the platform uses prioritization logic. Identical exact-match keywords have the highest preference when the search term is identical (including spell-corrected terms), and then identical phrase/broad/search-theme style targeting can take precedence over non-identical targeting. When the search isn’t identical, relevance signals (including the meaning of the search, the keywords in the ad group, and the landing pages in the ad group) can determine which ad groups are even considered, and then ad rank decides among equals.

This is a big reason SKAGs can backfire: by stripping an ad group down to a single keyword, you sometimes reduce contextual signals that help the system understand what that ad group truly represents. In practice, tightly themed ad groups (a small cluster of near-identical intent terms) often route traffic more predictably than hundreds of one-keyword silos.

4) RSAs reward variety; SKAGs often lead to repetitive assets and weaker ad strength

Modern Search creative is built around RSAs, where you provide multiple headlines and descriptions so the system can assemble combinations and learn what works. Repetitive, keyword-stuffed assets—common in SKAG builds—can limit variety and reduce the number of meaningful combinations. Pinning can enforce strict messaging, but heavy pinning reduces combinations and is generally not recommended unless you have mandatory language that must always appear. In other words: the SKAG mindset (“lock everything to the keyword”) often conflicts with how RSA optimization is designed to work.

5) Operational overhead becomes a performance tax

SKAGs multiply everything: ad groups, keywords, negatives, assets, settings, experiments, and reporting views. That overhead tends to delay optimizations that actually move ROI (search term pruning, landing page fixes, conversion tracking validation, budget reallocation, and offer testing). Unless you have a very strong process and enough conversion volume per ad group, SKAGs can turn into a high-maintenance structure that learns slowly and obscures decision-making.

A practical decision framework: SKAGs vs tightly themed ad groups

When I’d consider SKAGs (or “SKAG-lite”) today

• You have enough conversion volume on that one intent to justify isolation (otherwise learning is slow and results are noisy).
• The intent is genuinely unique and deserves a distinct landing page or offer (not just a wording variation).
• You need strict messaging control for compliance, qualifiers, or brand requirements—and you’re willing to use pinning carefully and sparingly.
• You’re running a focused test (new pricing, new landing page, new value prop) where isolating variables matters more than minimizing account complexity.

When SKAGs are usually the wrong move

• Your SKAG plan is mostly “one keyword per ad group” at scale across lots of synonyms and close variations.
• You rely on RSAs to explore messaging and need healthy asset variety across themes (SKAGs often lead to repetitive assets and over-pinning).
• You’re using broader matching and automation to expand reach efficiently—because the control you think you gain from SKAGs often isn’t real control anymore.
• Your team’s bottleneck is execution speed (search term hygiene, landing page work, tracking, budgets). SKAGs add moving parts.

How to get the upside of SKAGs without the downside (my recommended structure)

1) Build “intent clusters,” not single keywords

Instead of one keyword per ad group, aim for small, clearly themed ad groups where every keyword shares the same intent and can credibly use the same landing page. This gives the system more context (keywords and landing page alignment) while still keeping reporting and optimization clean.

2) Use match types to manage reach, and negatives to manage intent

Because meaning-based matching and close variants can blur edges, negatives are often the real control lever. Use them to block clearly wrong intents (free, jobs, DIY, definitions, unrelated product types), and to prevent internal overlap when two ad groups represent genuinely different intents.

3) Let RSAs do their job—write for combinations, not for one “perfect” ad

In each ad group, write assets that cover: the core offer, 2–3 differentiators, 1–2 proof points, pricing/financing if applicable, and 1–2 strong calls to action. Pin only what must be guaranteed (for example, a required disclaimer or a non-negotiable qualifier), and if you pin, consider pinning multiple options to the same position so you keep some flexibility.

4) Use the search terms report as your ongoing control system

The fastest way to decide whether SKAGs are “working” is not to stare at the keyword line item—it’s to review the actual queries coming in, then decide whether you need negatives, a new intent cluster, or a landing page adjustment. Also remember that the match type you see for a search term in reporting can be narrower than the match type of the keyword that triggered it, so interpret those labels carefully and focus on query intent first.

5) If you must test SKAGs, run a contained experiment

If you’re still unsure, don’t rebuild the whole account. Pick 5–20 of your highest-value intents, create a SKAG-lite variant (one ad group per intent, not per spelling variation), and measure against a comparable control group on conversion rate, cost per conversion, lead quality (if you can import offline outcomes), and the amount of negative keyword work required to keep intent clean.

Whether you choose SKAGs, SKAG-lite, or tightly themed intent clusters, the hard part is keeping query intent clean, ads and landing pages aligned, and structure changes grounded in real search-term data—not just theory. If you want help doing that consistently, Blobr connects to your Google Ads account and runs specialized AI agents that surface practical, prioritized actions, like finding new keyword opportunities, curating negative keywords to prevent overlap, and spotting where a high-value intent deserves its own ad group and landing page. For example, its Keyword Landing Optimizer can highlight mismatches between keywords, RSAs, and URLs and suggest where isolating a “must-win” term makes sense, so you can test SKAG-lite ideas without rebuilding everything by hand.