I recently stumbled across a Google AdWords video by Derek Faylor describing how to boost AdWords relevancy. He suggests picking one keyword that is core to your business, setting it to exact match and giving the keyword its own ad group with its own tailored ads. The idea is this: if your ads closely match your keywords, you will be seen by Google as being highly relevant, so your Quality Score will increase. This will lead to a higher ad rankings, higher click-though rates (CTR) and lower costs per click (CPC).

It makes sense, and I completely agree that a highly relevant approach such as that outlined by Derek is essential to achieve great results in paid search.

However, although Derek emphasises that his one keyword per ad group strategy should only be applied to one keyword which is core to your business, there will rarely be a case where a business will only want to advertise on a single keyword. There will likely be hundreds of possible phrases that will be highly relevant to a business, and having a portfolio of hundreds, even thousands, of long-tail keywords (instead of just bidding on one or two highly generic short-tail keywords) will often achieve better results.

So is Derek’s strategy of one keyword per ad group practical if applied on a larger scale?

Let’s have a look at the pros and cons.


1. Highly Relevant

Having one keyword per ad group makes it possible to write ads which very closely match the keyword. If the ad group contained only the keyword “brown leather shoes”, the ad could include the words “brown”, “leather” and “shoes”, possibly with prices of brown leather shoes, and take users through to the brown leather shoes landing page. However, if the ad group contained the keywords “brown leather shoes”, “blue suede shoes” and “red wellington boots”, at best, a generic “shoes” ad could be shown. Having very different keywords in the same ad group makes it impossible to create a highly relevant PPC campaign.

2. High Quality Score

As previously pointed out, having one keyword per ad group would likely achieve high Quality Scores, high CTR and strong ad rankings. Conversion rates are also likely to benefit, as highly relevant ad text will make users more pre-qualified before clicking.

3. Easy to Optimise Ads for Quality Score

If you notice Quality Score for one of your keywords is low, having one keyword per ad group makes it is relatively easy to identify which keyword / ad combination is performing poorly and make appropriate changes to improve its Quality Score.


1. Unnecessary

Having every one of your keywords in its own ad group is unnecessary. If you have two keywords, “blue suede shoes” and “suede shoes blue”, what benefit is there of having each of those keywords in a separate ad group? The keywords are so similar, so you could not possibly write a more relevant ad for one if it were in its own ad group.

2. Unmanageable

Imagine a Google AdWords ad group with 2,000 keywords. If each had its own ad group, that would mean 2,000 ad groups. Imagine how difficult it would be to manage 2,000 ad groups, most of which would fail to see a single click. Since many of the keywords would be so similar, the same ads would likely be used across multiple ad groups. There would be a lot of duplication of ads and it would take forever to download reports or update changes. Quite simply, your AdWords account would become incredibly time-consuming and frustrating to manage.

3. Diluted Ad Text Performance Data

If each keyword had its own ad group, impression and click data for ads would be diluted over a greater number of ad groups, making analysis and optimisation of ads difficult and less meaningful. If , however, 10 of your very similar keywords were grouped together in one ad group, impression and click data for those 10 keywords would be aggregated for the ad group’s ads, making it easier to spot which ads are performing well and which need changing.


Although having one keyword per ad group would be nice in a perfect world, considering that there are potentially thousands of keywords that could be relevant to your business, having one keyword per ad group is taking paid search to an unmanageable level. While having one keyword per ad group is one extreme of paid search management, putting all of your keywords in one ad group is the other. The best results in paid search are achieved from a balanced approach, somewhere in between the two extremes.

So don’t just separate out keywords into their own ad groups for the sake of it. Your account will soon become unmanageable and you’ll dilute your ad text data. Instead, group very similar keywords together, even if they are broad matched. As long as your keywords are very similar, 5-20 keywords per ad group is fine. I regularly achieve Quality Scores of 9 and 10 using this approach. The key is to make sure the keywords in each of your ad groups are very similar, and that your ads are highly relevant to the ad group’s keywords.

Start with maybe 10 closely related keywords in each ad group. Once you have some data collected, use the 10% Clicks Rule to decide which of your ad groups and keywords could benefit from being split out and given their own tailored ads. It will keep your time and effort focused only on the parts of your campaign which deserve your time and effort.

As with most things in life, it’s about finding a balance. As you continually strive to improve the relevancy of your keywords and ads, make sure the strategy you are adopting is achievable and sustainable. If you put all your keywords in the one ad group, you’ll receive a poor Quality Score as users fail to engage with your ads. If you over complicate your keyword / ad group structure, you’ll end up creating a bloated paid search account and start to lose focus of your long-term goals. Find a balance that works for you.

Rule of Thumb

So unfortunatley there isn’t really a rule for the number of keywords an ad group should contain. There isn’t a ‘best’ number of keywords you should aim to have in each ad group. It’s about finding what works best for your business, for your products or services, for your set of keywords. But if you ever find yourself unsure whether a keyword should be split out into its own ad group, ask yourself this:

A keyword should only be give its own ad group if you think you could write a more relevant ad (or show a more relevant landing page) for that keyword if it were in its own ad group.

Alan Mitchell

Alan Mitchell is a Google AdWords PPC specialist, based in Melbourne, Australia, with a proven track record at improving return on investment (ROI) from Google AdWords. Find out how Alan can help your business.

  • http://www.clickequations.com/blog Craig Danuloff

    Good post and blog (which I’ve just discovered today.) Enjoying the style and substance of your analysis as I sample the last few posts.

    I think this post has an interesting tact, but seems to risk putting the cart in front of the horse a little bit. What I mean is, relevance is primarily an input to Quality Score, so the only reason to split up keywords into smaller ad groups is to improve Quality Score. (There is a slight chance you’d want to do it to write ads which resulted in higher conversion rates, but the correlation between QS and conversion is usually quite high. You comment about targeting a better landing page obviously makes sense too.)

    All I’m suggesting is that before anyone should consider all this, they should check whether or not they need to improve Quality Score on the keywords in question. If so, then tightening up ad groups and writing more specific and click-worthy copy is a great idea. Otherwise, doesn’t seem worth any effort.

    Looking forward to my new subscription to your blog.

  • Alan Mitchell

    Interesting point Craig. I hadn’t really thought about it from a Quality Score perspective.

    I agree that keywords with a low Quality Score are often much better candidates for ad group expansion than keywords with a high Quality Score. I also agree there is generally a correlation between Quality Score and conversion. However, I think a stronger correlation exists between that of Quality Score and CTR, and if I am right in assuming that a higher CTR generally means a higher Quality Score, I would not feel confident saying that a higher CTR necessarily leads to a higher number of conversions.

    In my opinion, Quality Score should be seen as an indicator of a successful campaign, rather than the cause of a successful campaign. I don’t think Quality Score should be the end goal of PPC management; I think relevancy itself should be. If high keyword / ad group relevancy is your end goal, Quality Score will naturally follow. But if Quality Score is your end goal, there is a danger of focusing too heavily on improving CTR, and in doing so opening up your messages too widely, making them unqualified, unfocused and unoptimised for conversion.

    So although keywords with a Quality Score of 5 will rarely outperform keywords with a Quality Score of 10 (in terms of ROI), and you’re completely right that Quality Score can be useful in identifying areas for ad group tightening, I think Quality Score should only be used as a rough guide in PPC management. Relevancy itself, in my opinion, should be the focus, and everything else will follow.

    Thanks for commenting. Appreciate the feedback and hope you enjoy the blog!

  • http://www.espinteractivesolutions.com/ bay area web design

    One key per ad group is not a good approach. i think 10 to 20 keywords per ad group is a good choice.You can easily cover all keywords in ads copy and manage high quality score.