The debate between SKAGs (Single Keyword Ad Groups) and STAGs (Single Theme Ad Groups) has divided Google Ads practitioners for years. On one side, SKAG advocates champion the granular control and precise ad-to-keyword matching that made this structure a best practice for nearly a decade. On the other, STAG proponents argue that Google's machine learning has evolved to the point where hyper-granular structures actually hurt performance by fragmenting the data algorithms need to optimize.
The truth is that both camps have valid points—but the optimal answer in 2026 depends heavily on your specific situation. This guide will help you understand when each structure excels, how to evaluate your current setup, and how to migrate between approaches without destroying your account performance.
Understanding Single Keyword Ad Groups (SKAGs)
SKAGs represent the pinnacle of manual account control. In a pure SKAG structure, each ad group contains exactly one keyword, historically in multiple match types: exact match, phrase match, and broad match modifier (before Google deprecated BMM in 2021). The philosophy behind SKAGs is simple—when you isolate each keyword in its own ad group, you can craft ad copy that matches the search query with surgical precision.
This structure emerged when Google Ads (then AdWords) rewarded tight keyword-to-ad relevance with higher Quality Scores, lower costs per click, and better ad positions. A search for "red running shoes women" triggering an ad with that exact phrase in the headline would consistently outperform a generic "Shop Running Shoes Today" ad pulled from a broad ad group.
The classic SKAG structure
A traditional SKAG setup for a running shoe retailer might look like this:
- Ad Group: red running shoes women
- Keywords: [red running shoes women], "red running shoes women"
- Ad Headlines: "Red Running Shoes for Women" | "Shop Women's Red Runners"
- Ad Group: blue running shoes women
- Keywords: [blue running shoes women], "blue running shoes women"
- Ad Headlines: "Blue Running Shoes for Women" | "Shop Women's Blue Runners"
This structure provides maximum control over the search-to-ad experience. You know exactly which keywords trigger which ads, making troubleshooting straightforward. When a keyword underperforms, you can adjust its bid, pause it, or rewrite its specific ad without affecting anything else.
Advantages of SKAGs
- Precise keyword-to-ad matching: Every search triggers an ad specifically written for that query
- Granular bidding control: Set different bids for each keyword without relying on modifiers
- Clear performance data: Instantly see which exact keywords drive results
- High Quality Scores: Tight relevance historically boosted QS significantly
- Easy troubleshooting: Problems are isolated to specific ad groups
The problems with SKAGs in 2026
Despite their historical advantages, SKAGs have become increasingly problematic as Google's platform has evolved. The issues fall into three main categories: match type changes, smart bidding requirements, and algorithmic evolution.
First, exact match no longer means exact. Google now matches exact match keywords to close variants, meaning your [red running shoes women] keyword might trigger on "women red running sneakers" or "running shoes for women in red." The granular control SKAGs provided has been partially removed at the match type level itself. For a deeper understanding of these changes, see our keyword match types guide.
Second, smart bidding strategies like Target CPA and Target ROAS require conversion data to optimize. When you spread your conversions across hundreds of single-keyword ad groups, each group might only see a handful of conversions per month. This data fragmentation means Google's machine learning never has enough signal to optimize effectively for any individual ad group.
| Structure | Conversions/Month | Conversions per Ad Group | Smart Bidding Effectiveness |
|---|---|---|---|
| 100 SKAGs | 200 | 2 average | Poor - insufficient data |
| 20 STAGs | 200 | 10 average | Moderate - learning phase |
| 5 Themed Groups | 200 | 40 average | Strong - optimization enabled |
Understanding Single Theme Ad Groups (STAGs)
STAGs take a fundamentally different approach to account organization. Instead of isolating individual keywords, you group keywords that share the same search intent and theme. The principle is that if multiple keywords would be well-served by the same ad copy and landing page, they belong in the same ad group.
This approach aligns with how Google's algorithm now thinks about relevance. Rather than matching keywords to ads based on exact text overlap, Google evaluates semantic relevance and user intent. A search for "best running shoes for marathon training" and "marathon running shoe recommendations" express the same intent, even though they share few words.
A STAG structure example
Here's how the same running shoe retailer might organize with STAGs:
- Ad Group: Women's Running Shoes - Color Variants
- Keywords: red running shoes women, blue running shoes women, pink running shoes women, black running shoes women, white women's running shoes
- Ad Strategy: Dynamic keyword insertion or responsive search ads that adapt to the color searched
- Ad Group: Women's Running Shoes - Purpose
- Keywords: women's marathon running shoes, trail running shoes women, women's racing flats, cushioned running shoes women
- Ad Strategy: Ads focused on finding the right shoe for your running style
Notice that color variants share an ad group because the intent is the same—someone wants women's running shoes in a specific color. The landing page would filter by color, and Responsive Search Ads can dynamically match the color in headlines. But purpose-based keywords form a separate group because someone searching for trail running shoes has fundamentally different needs than someone searching for racing flats.
Advantages of STAGs
- More data for machine learning: Consolidated groups give smart bidding enough conversions to optimize
- Reduced management overhead: Fewer ad groups means less time spent on routine maintenance
- Better Responsive Search Ad performance: More search queries per ad group provides better RSA learning
- Alignment with Google's direction: Google explicitly recommends consolidation for smart bidding
- Sustainable scaling: Adding new keywords doesn't require creating new ad groups
When STAGs fall short
STAGs aren't perfect for every situation. The main limitation is reduced control over the search-to-ad experience. When you have 15 keywords in an ad group, you can't write headline copy that perfectly matches each search query. Dynamic keyword insertion helps, but it can produce awkward or truncated headlines when keywords are long.
Additionally, reporting becomes less granular. In a SKAG structure, you can immediately see that your "red running shoes women" keyword generated 50 conversions at $15 CPA. In a STAG, you see that your "Women's Running Shoes - Color Variants" ad group generated 200 conversions at $14 CPA, but determining which specific color queries performed best requires diving into search term reports.
The Impact of Broad Match on Account Structure
Perhaps no single factor has changed the SKAGs vs STAGs debate more than Google's push toward broad match combined with smart bidding. Google now recommends using broad match keywords with automated bidding as the default approach for most advertisers. Understanding why requires grasping how modern broad match actually works.
Broad match in 2026 is fundamentally different from the broad match of 2015. Previously, broad match keywords would trigger on any remotely related query, leading to massive wasted spend. Today, broad match uses Google's understanding of intent, your landing page content, other keywords in your account, and the user's previous behavior to determine relevance. Combined with smart bidding, Google adjusts bids in real-time for each auction based on the likelihood of conversion.
How broad match changes the structure equation
With broad match handling query expansion automatically, the granular keyword variations that SKAGs were designed to capture become redundant. Your single broad match keyword "women's running shoes" might trigger on thousands of relevant queries that you'd need dozens of exact match SKAGs to cover manually.
| Approach | Query Coverage | Management Time | Algorithm Data |
|---|---|---|---|
| 50 exact match SKAGs | 50-100 queries | High | Fragmented |
| 10 phrase match STAGs | 500-1000 queries | Moderate | Sufficient |
| 3 broad match themed groups | 5000+ queries | Low | Abundant |
The trade-off is control. With broad match, you're trusting Google to make good decisions about which queries to show your ads for. This works well when you have sufficient conversion data and properly configured conversion tracking, but can lead to wasted spend in accounts with low volume or poorly defined conversion goals.
Making the Decision: SKAGs, STAGs, or Hybrid?
The right account structure depends on your specific circumstances. Here's a framework for deciding which approach fits your situation:
Choose SKAGs when:
- You have very high-value keywords where individual bid control is critical
- Your account has low conversion volume (under 30 conversions/month)
- You're in a highly regulated industry where ad copy must be precisely controlled
- You're testing new keywords and need clear, isolated performance data
- You're using manual CPC bidding and need keyword-level bid control
Choose STAGs when:
- You're using smart bidding (Target CPA, Target ROAS, Maximize Conversions)
- Your account generates 50+ conversions per month
- You have many keywords that share the same intent and landing page
- Account management time is a constraint
- You're scaling and need a sustainable structure
The hybrid approach
Many successful advertisers use a hybrid structure that takes the best of both worlds. High-value, brand keywords might remain in dedicated ad groups for control, while mid-funnel and discovery keywords are organized into themed groups. This provides the precision where it matters most while allowing consolidation where it benefits performance.
For example, a B2B software company might structure like this:
- Brand campaigns: SKAG structure for competitor and brand terms
- Product campaigns: STAGs organized by product category
- Solution campaigns: Broad themed groups around problems solved
Migration Strategy: Moving from SKAGs to STAGs
If you've decided to consolidate from SKAGs to STAGs, the migration process matters as much as the destination. Abrupt restructuring can tank performance, lose historical data, and reset learning phases. Here's a proven approach for safe migration.
Phase 1: Analysis and grouping (Week 1)
Before changing anything, analyze your current SKAGs to identify logical groupings. Keywords that share these characteristics should be combined:
- Same search intent (informational, navigational, transactional)
- Same landing page destination
- Similar conversion rates and CPAs
- Semantically related themes
Create a spreadsheet mapping each existing SKAG to its proposed STAG. Flag any keywords that seem like outliers—these might need their own groups or different treatment.
Phase 2: Parallel testing (Weeks 2-4)
Rather than replacing SKAGs directly, create new STAG campaigns alongside your existing structure. Use campaign experiments or draft campaigns to test the consolidated structure with a portion of your budget. This lets you compare performance directly without risking your entire account.
- Start with 20-30% of budget in the new STAG structure
- Run both structures simultaneously for 2-4 weeks
- Monitor key metrics: CPA, ROAS, conversion volume, impression share
- Allow the learning phase to complete before judging results
Phase 3: Gradual shift (Weeks 5-8)
If the STAG structure performs comparably or better, gradually shift more budget from SKAGs to STAGs. Don't pause SKAGs entirely until you're confident the new structure can handle the full volume. A typical progression:
- Week 5: Shift to 50/50 split
- Week 6: Move to 70% STAG / 30% SKAG
- Week 7: Shift to 90% STAG / 10% SKAG
- Week 8: Pause remaining SKAGs (don't delete—keep for reference)
Phase 4: Optimization (Ongoing)
Once migrated, continue optimizing your STAG structure. Watch for ad groups that become too broad—if a single group has keywords with dramatically different conversion rates, consider splitting it. Conversely, if two small groups always have similar performance, they might benefit from combination.
Modern Account Structure Best Practices for 2026
Beyond the SKAGs vs STAGs debate, several best practices have emerged for Google Ads account structure in the current landscape. These principles apply regardless of which organizational approach you choose.
Campaign-level organization
Your campaign structure should reflect your business goals and budget allocation needs, not just keyword themes. Consider organizing campaigns by:
- Business objective: Separate campaigns for acquisition vs retention
- Product line: Different products may need different ROAS targets
- Geographic region: Especially for location-based bidding strategies
- Funnel stage: Brand, competitor, and generic keywords often need different treatment
This organizational approach gives you budget control where it matters while allowing consolidation within campaigns. For a complete guide, see our search campaigns guide.
Responsive Search Ads best practices
RSAs work best with STAG structures because they have more query variation to learn from. For optimal RSA performance:
- Provide 10-15 unique headlines that cover different angles and keyword variations
- Include at least 4 descriptions with varied value propositions
- Pin critical messages (like brand name) rather than entire ads
- Use ad strength as a guide, but don't sacrifice relevance for "Excellent" ratings
Smart bidding configuration
Smart bidding performs best when it has clear signals and sufficient data. To maximize effectiveness:
- Set conversion actions that reflect true business value
- Use value-based bidding when conversion values vary significantly
- Allow 2-4 weeks of learning phase after any significant change
- Avoid making multiple changes simultaneously—isolate variables
Negative keyword management
With broader match types and consolidated structures, negative keywords become even more critical. Implement these practices:
- Review search terms weekly for the first month of any new structure
- Create shared negative keyword lists by theme or exclusion type
- Use negative keyword lists at the campaign level for efficiency
- Don't over-negate—broad match needs room to find converting queries
Measuring Success After Restructuring
Evaluating whether your structure change worked requires looking beyond simple CPA or ROAS comparisons. Consider these metrics holistically:
| Metric | What to Watch | Healthy Change |
|---|---|---|
| Conversion Volume | Total conversions, not just rate | Stable or increasing |
| CPA/ROAS | Primary efficiency metric | Within 10% of baseline |
| Impression Share | Are you reaching more of the market? | Stable or increasing |
| Search Term Relevance | Quality of queries triggering ads | Maintaining high relevance |
| Quality Score | Account-level average | Stable after initial adjustment |
| Management Time | Hours spent on account tasks | Decreasing over time |
Give any structure change at least 4-6 weeks before drawing conclusions. Early fluctuations are normal as algorithms learn and Quality Scores stabilize. The true measure of success is sustainable, scalable performance over time.
Platform Comparison: Structure Strategies Across Ad Networks
Understanding how account structure differs across platforms provides useful perspective. While Google Ads has evolved toward algorithmic optimization, other platforms have their own structural best practices.
Meta Ads takes a fundamentally different approach—instead of keywords, you're organizing around audiences and creative. Meta explicitly recommends campaign consolidation with Advantage+ features handling much of the optimization. The parallel to Google's broad match + smart bidding is clear.
TikTok Ads similarly favors simplified structures. TikTok's algorithm performs best when given creative flexibility within broad targeting parameters, making it more similar to modern Google Ads strategy than many advertisers realize.
The cross-platform trend is unmistakable: algorithms are getting better at optimization, and advertisers benefit most by providing good creative and clear conversion signals rather than trying to micromanage every targeting detail.
Conclusion
The SKAGs vs STAGs debate ultimately comes down to matching your account structure to how Google Ads actually works in 2026. SKAGs made sense when keyword-level control was the primary driver of performance. Today, with smart bidding, broad match, and machine learning driving optimization, most advertisers get better results from consolidated STAG structures that give algorithms the data they need.
That said, the best structure is one that works for your specific situation. Low-volume accounts, highly regulated industries, and scenarios requiring precise control may still benefit from more granular approaches. The key is understanding the trade-offs and making an informed choice rather than following outdated best practices.
Ready to optimize your Google Ads account structure? Benly's AI-powered platform can analyze your current setup, identify consolidation opportunities, and monitor performance through any structural changes—helping you make data-driven decisions without the guesswork.
