SKAG (single keyword ad group) Google Ad campaigns are campaigns in which you create a structure where you have one keyword per ad group. This allows you to write highly specific ads for that specific keyword, which maximizes the relevance, CTR, Quality Score, and conversion rate.
With that in mind, SKAGs should be a cornerstone tactic in your Google Ads arsenal.
That said, your entire account shouldn’t be built around the SKAG-structure. It makes for a rigid, unbalanced structure, which doesn’t prioritize your work or balance impact versus workload.
In this blog post, we will go in-depth with the human approach to creating and managing SKAG campaigns. We will cover the basic principles for when SKAGs are a good idea, and when they’re not.
NOTE: this blog post covers tactics that are slightly different than what we do internally in SavvyRevenue. We always aim to implement automated setups using data feeds; then, we manually build campaigns for generic keywords using the 80/20 rule.
However, most people don’t have access to our tools. If I just came out and explained everything that we did, you wouldn’t be able to replicate it anyway— so what’s the point in making it sound fancy?
Before jumping into SKAGs, then it’s important to highlight that this advice is mainly for eCommerce stores. The value of the SKAG structure isn’t the same for all businesses. National Lead Generation campaigns might not see the same value (more on this below).
How Are SKAGs Different For eCommerce Stores?
In many ways, SKAGs are more important for eCommerce stores than other advertisers. This is because you have so many category, brand, and product keyword combinations that will require relevant ads.
Failing to do this in this day and age will undoubtedly result in a lackluster performance.
One can argue that someone looking for plumber prices, sink repair, plumber Dallas, etc. will react equally well to an ad and landing page for Plumbers in Dallas:
I need a plumber, so who really cares about whether the ad says plumber prices or sink repair. I know that a plumber will fix my sink. I just have to find the right one.
It’s different with eCommerce. Consumers searching for Adidas Ultrafoam running shoes vs Adidas Boost running shoes will react a lot better not only to a specific ad, but also a landing page that only shows the specific running shoes they’re looking for.
Therefore, the power of SKAGs is greater for eCommerce stores than with regular lead generation campaigns.
What You’re Missing By Not Using Single Keyword Ad Group (SKAGs) Campaigns
The general guidelines around Single Keyword Ad Groups are still the same:
Great for maximizing relevance and therefore improve CTR, Quality Score, Conversion Rate and ultimately your profits. It’s truly a win-win.
The reason why SKAG campaigns are so effective is that you can write better ads and link to landing pages that match the searcher’s intent with SKAGs.
The opposite of SKAGs is adding a lot of keywords to the same ad group.
Let’s say you have one ad group for all the following keywords:
- Running shoes
- Marathon running shoes
- Trail running shoes
- Lightweight running shoes
They’re all running shoes, so what does it matter, right?
However, this greatly limits your ability to write a good ad.
First and foremost, you can’t use the words marathon, trail, or lightweight as one word wouldn’t be relevant for all keywords.
So you’d end up writing a generic ad that applies for all the ad groups:
Not a bad ad, right? But they can be better.
Imagine you create a SKAG structure, so you get the following ads:
But wait – we are not done.
The ads can be further enhanced with ad customizers:
Depending on how your data feed is structured, you can even take it even further with benefits, USPs, etc. for each shoe in the ad copy.
That’s how you win in Google Ads. But it requires a SKAG approach.
Many PPC rockstars have different opinions on SKAGs. My good friend Duane Brown sums it up great:
“To many people started using SKAG as a strategy vs the very tactic it started out as. SKAG is not a one size fits all approach for client’s. There are as many ways to structure an account as there are to grow it. “–Duane Brown
If you don’t understand why you are running SKAGs and just running them because it’s a best practice, then you’re missing the point.
SKAGs are great though and one reason why many love to hate them is that they do complicate an account a bit if you don’t know how to work with.
Honestly, and I know I speak for many pros – we wouldn’t know how to keep an account streamlined without using SKAGs! Accounts would be a mess, we wouldn’t find keywords when we needed them and that would result in duplicated keywords.
You know how I know? Because I’ve seen it hundreds of accounts.
The Reasons Why Pros Love SKAG Campaigns and Amateurs Love Hating On Them
Not everything is rainbows and unicorns when it comes to SKAG campaigns.
They can be hard to scale and very time consuming if you’re trying to build campaigns manually or don’t have a good structure for optimizing your campaigns subsequently.
It takes forever to perform any form of ad testing for all but keywords with the highest amount of clicks. The more ad groups you create, the longer it takes to create the new ad variations, and it also takes longer to actually test the new ads because you have less data.
A hundred (and often more) campaigns become impossible to manage for most PPC managers. I’ve spoken on several occasions about what it takes to become great at managing large-scale PPC campaigns.
The truth is that a SKAG campaign structure can take an otherwise easy account to a nightmare of exponential complexity.
You need to have systems and processes in place in order to manage a SKAG structure without losing sight of what’s most important.
SKAG campaigns are directly opposite of a lot of Google’s Best Practices when it comes to getting their automation working perfectly. Yup, Google actively recommends their largest advertisers to group keywords together and let their algorithm figure out how to match keywords with ads.
Whenever Google recommends something contradictory to the advice of the search pros, it becomes a struggle to discuss due to the inherent weight advert new advertisers put on Google’s recommendations. They’re from Google, so they should know best, right?
Wrong. Google gives a lot of great advice, but all of it is meant to apply for most advertisers. Google Reps are also incentivized to have their clients implement Google features, so they will be pushy abut it.
How to Make SKAG Campaign Structures Unsuck (Yes, it’s a word!)
Luckily, there are ways around making it easier to manage SKAG campaigns.
Use a Tool for Building SKAGs
There are a number of tools out there to build SKAGs.
Of course, you can also just use a spreadsheet, but if you don’t have strong spreadsheet skills, then you need a tool.
We use an internal tool (and feed-based campaigns – more below) to work with SKAG structures, but the two tools I list above are good places to start.
Controlled Chaos: Not Everything Has To Be Perfect
This is probably one of my biggest pet peeves. Yes, you need to be vigilant that you don’t create a mess, but going out of your way to create an ad group for every single keyword can sometimes be counterproductive.
Remember, the reason for creating SKAGs is to maximize the keyword to ad relevance.
If you have keywords where you don’t need a new ad (i.e. competitor keywords), then it’s generally not needed with standalone ad groups.
As your account becomes very large, you want to keep the complexity to a minimum.
The way I solve for this is to have controlled chaos ad groups where I add keywords that don’t deserve special attention, but I don’t want them mixed in with my hyper-optimized SKAG structure. The 80/20 rule is everything in this setting.
For example, consider the following ad group:
Bose QC 35ii headphones
However, you have search terms like:
Bose QC 35ii headphones fry’s electronic
You don’t really need a SKAG for that keyword. Especially as you find dozens of small competitors across the country, then it will become overkill.
You’re not going to write Fry’s Electronic in the ad (I hope). So unless the keyword has crazy search volume, there’s no need for it to have its own ad group.
Just stick it in your CC (controlled chaos) ad group:
- Bose QC 35ii headphones fry’s electronic
- Bose QC 35ii headphones walmart
- Bose QC 35ii headphones best buy
Oh yeah, professionals might call these ad groups for miscellaneous, but I like CC.
Commit to Using Multi-Ad Group Ad Testing…
… which you should anyways.
The challenge with diluting data when you have SKAGs is easily overcome with multi ad group ad testing. It’s really one of those things that sound much harder than it actually is. Once you get the hang of it, you’ll be surprised you were ever able to test ads the old fashioned way.
I’ve written a bit about multi-ad group ad testing here, but will not go into further detail as it’s an entire subject in itself. Keep tuned for a guide on the subject, or a video below!
Process for Working with SKAG Campaigns
Review Search Terms and Create New SKAGs for Good Keywords. Exclude Irrelevant Keywords.
But it takes time! And hardly anyone does it consistently, but it’s so worth it.
At least once a month, go through the search terms from your Broad Match Modified and Phrase Match keywords:
Check what other terms your keyword is triggering, and add them to a new single keyword ad group:
Don’t go overboard when you have an ecommerce account: Singular/Plural and other minor word variations stay in the same ad group:
Rule of thumb = if you require a new ad message, then you should create a new ad group.
Harvest your Broad Match Modified keywords so that you continue expanding your SKAG approach. The better you are at finding all keywords before you launch, the less work this will be. For the first 3-6 months of a new campaign, expanding your SKAG structure should be a routine task every two weeks.
Exclude Poor Performing Search Terms
At the same time, you should (obviously) add any search terms that are poor performing, or irrelevant, as negative keywords.
I like tools like Optmyzr, or working with search query reports in Excel for this task.
Working with a search query report in the interface is like shaving by removing on hair at a time.
Not. Worth. It.
Test New Messages Across Multiple Ads With Multi-Ad-Group Testing.
I like testing two ads in most accounts. It generates the quickest statistical significance. But for very high-volume ad groups, I’ll test 3 or 4 ads at a time.
I highly recommend using Adalysis for multi ad group ad testing.
The idea behind multi-ad group ad testing is that you’re testing the same element across multiple ad groups.
This means that if you have the following ad groups:
- Reading Glasses
- Cheap Reading Glasses
- Reading Glasses for Men
- Reading Glasses for Women
- Designer Reading Glasses
- Quality Reading Glasses.
Then the first part of the ad will most likely be the keyword:
In the second headline you can see the part “3 pairs only €59 – save 33%”. This particular ad is a pricing message.
However, the rest of the ads don’t have to be pricing focused. You can change the message to perform ad tests on other elements:
Or any combination.
Whatever pattern/message you choose to add to your ads is the one you should test across all your ad groups within a campaign.
Here’s a great video on how to get started with multi ad group ad testing:
Match Types and Ad Groups
Oh, the age-old question. Should each match type have its own ad group or campaign?
As with many things, then it depends on your personal preference and your industry. If you’re running campaigns for a large lead gen account (i.e. insurance), then you will not have too many keywords (compared to a 10,000 SKU eCommerce campaign). Splitting campaigns or ad groups into individual match types can amke sense.
If you’re running match type specific ad groups or campaigns for an eCommerce store with +20,000 products, you’ll quickly run out of space in the account.
However, some will argue that at that level, you’ll benefit from focusing on exact match keywords to better control the exact search queries.
Of course, it depends on your specific case, so I can’t say much without writing another 1,000 words. However, my advice can best be summarized in this way:
- Start with not splitting in match types
- Once you’ve hit a performance plateau, revise your strategy and test if you can increase performance by splitting based on match type
- Consider splitting ad groups following the 80/20 rule for where you can get the greatest impact
The biggest issue with splitting your campaigns based on match types is that it will split up your data. The more prevalent audience-bidding becomes in Google Ads, the more important it will be to join data.
Another aspect of splitting campaigns in match types is if you want to push more budget towards the better performing match type. Then it makes sense to split campaigns into match types.
But just splitting campaigns into match types because you read someone does it is not smart. It’s dumb.
Whatever complexity you add to your account has to be defensible. When you add complexity it’s because you’ll use it to improve performance. Otherwise you’re just overcomplicating something that should be simple.
How To: Tool Suggestions for In-House
There are quite a few tools that can set up SKAG campaigns for you.
A lot of PPC managers will opt to use a spreadsheet, which is definitely possible. However, it’s highly likely that you don’t have the skills to set up a spreadsheet that can do the job.
A simple tool like fiuti.com (whoever came up with that name needs a crash-course in marketing) is super useful for building out the campaigns. The most time-consuming part is adding Ad Extensions.
It typically takes me more time to add proper ad extensions than building the actual campaigns including keyword research and ad writing, which is really, really sad.
The Not-So-Well-Known Approach: Build Feed-Based Campaigns
Stay tuned for a video series on how hot build feed-based campaigns 😉
An approach that is quickly becoming more accessible is building SKAG campaigns using a Google Shopping feed (or another feed).
It’s automatic, easy, and super fast once you have set it up. If your store is growing quickly with a lot of new products, then it’s a great way to capture new products, categories, and brands without having to manually check hundreds of categories manually.
Feed-based Google Ads campaigns also allow you to easily build hyper-focused ads with pricing, inventory and other data directly in your ads.
The Challenges with Using a Feed
You’ll end up with a very narrow keyword focus. Because you will only have one “keyword” per category, then you will miss out on a lot of keywords if you only have feed-based campaigns.
An example of this is the category desk fans. A desk fan is also called a table fan, and most will also look for USB fans when they are really looking for desk fans.
So it’s your job to ensure that you expand upon this.
You have two (three) options:
- Create a second manual feed where you add the necessary keyword variations.
- Have a feed optimization tool that allow you to create variations
- Create “manual” campaigns for high-converting categories
Expand With Category and Brand Keywords
Due to the rise in campaigns based on a Google Shopping feed, then most feed-based campaigns are focused on product keywords (for eCommerce). However, this really misses the mark.
Yes, product keywords are great and should be a part of your Google Ads campaign structure, but there are many other keywords that can bring even more value to your campaigns.
The most noteworthy keywords are brand and category keywords.
I will write another post on how to approach this part of feed-based campaigns, but here is a quick summary:
- Clean your feed, so you have a field that can be used for the category keyword (i.e. no > > layers)
- Add the necessary brand and category links to your feed to use them to build campaigns (either directly from your store, or in your feed optimization tool).
You need to spend a lot of time building your feed. However, comparing to building campaigns manually, then you are really just moving time from one manually building campaigns to setting up the foundation for automatic campaigns, which is a lot more sustainable and will provide great ROI over time.
The One Action with the Highest Impact
Honestly, SKAGs suck for people who haven’t tried them yet. If you have an account right now that does not use SKAG, the sheer thought of moving to a SKAG structure is probably overwhelming.
My best advice is to take just one action based on this blog post:
- Sort your Google Ads Search campaigns based on cost.
- Take the top campaign, and restructure it to SKAG campaigns. Depending on the size of this campaign you will have to restructure it into several campaigns, so it’s properly structured
- Experience the improvement in performance, so you can get inspired to do the same for the next campaign and so on.
This will help you build experience.
Consider reading my guide to Google Ads account structure for eCommerce before you start so you know how to structure your account.