I’m going to focus my first post on what I believe is the most fundamental concept in PPC: relevancy. Giving users what they are looking for. Directing them to where they want to go. Answering their questions.

Why? Because paid search relevancy can pay massive dividends. Not only is a highly relevant pay per click (PPC) campaign more likely to receive a higher click-through rate (CTR), higher Quality Score, higher ad rankings, lower costs per click (CPC) and benefit from less wasted spend, but users will more qualified so bounce rates are likely to fall (the number of people who immediately ‘bounce’ back), conversion rates increase and return on investment (ROI) will ultimately improve. So a highly relevant paid search campaign is definitely a good thing.

To achieve PPC relevancy, keywords, ads and landing pages need to work together in tandem. Messages in ads need to match users’ search queries, landing pages need to match messages in ads and landing pages need to relate to users’ original searches. (For a more detailed explanation of how each component interlinks, you might like to consult Acquisio’s great article on AdWords relevancy and Quality Score).

Closely matching ads and landing pages to keywords to encourage only targeted and qualified users to visit your site is a simple theory, and one that’s been around since the dawn of Google AdWords.

So nothing new then – does that mean relevancy is no longer relevant?

Well, not exactly, for two reasons…

Impossible to achieve

Firstly, relevancy can never be achieved in its perfect form. There will always be ways a PPC advertiser can improve his keyword selection, negative keyword list, match type strategy, ad copy matching and landing page selection to give the user a more engaging and personalised experience. Just like there will never be a 10.0 film on IMDB, there will never be a perfectly relevant PPC campaign. One can only strive towards perfection – towards the Holy Grail of relevancy.

Higher expectations

Secondly, the bar of relevancy is constantly being raised. As search engine continually improve their algorithms to provide users with more relevant organic search results, so paid search ads will have to improve to keep up with the growing expectations of searchers. Ads which might have been considered ‘quite relevant’ last year may be considered ‘not so relevant’ now. Ads which sufficiently answer the questions of searchers today may not do so next year when people start to demand a more personalized and tailored service.

So not only is the Holy Grail of relevancy (a perfectly relevant campaign) impossible to achieve, but it is getting more and more impossible to achieve as we speak.

But all is not lost. No-one expects perfection, after all. Just being better than the competition can reap massive benefits for advertisers. And as I’m about to point out, getting better than the competition doesn’t need to be difficult. There are opportunities everywhere.


Say you’re interested in visiting Sydney and want somewhere to stay. Load up Google, search for ‘Sydney hotels’ and look at the paid search results.

Of the 10 PPC ads, 9 mention the words ‘Sydney’ and ‘hotels’. Most of the ads are calling out to the user, “I have hotels in Sydney! Come to Me!” Most of the ads are relevant to your search.

But let’s say you know a little bit more about your Sydney hotel requirements. After all, you can’t be bothered clicking through each of the hundreds of paid search results (or the 22,900,000 organic results for that matter). Come to think of it, you are interested in going to Sydney next weekend, you your refine your search query.

Search for ‘weekend breaks in Sydney’ and look at the paid search results. Although most ads mention ‘Sydney’, not a single advertiser includes the words ‘weekend’ or ‘break’ in their ads. No-one is shouting out to the user, “Yes! I have weekend breaks in Sydney! Come to me!” Every ad appear to be a generic ‘Sydney Hotel’ ad that may or may not be relevant to your weekend requirements.

Suppose, instead, when searching for ‘weekend breaks in Sydney’, you saw one of the following ads:



Suppose they took you through to a special ‘weekend break’ page, specifically designed for people looking to stay at the hotel over the weekend. Along with suggestions of local Sydney sights, activities and restaurants that could easily be fitted in over a weekend were reviews from people staying at the hotel on Friday and Saturday nights.

Would you be more likely to consider this hotel in your plans? I know I might.

More opportunities

Okay, only 36 people searched for ‘weekend breaks in Sydney’ in June. But these were 36 people who knew what they were looking for and were delivered poor, generic, one-message-fits-all ads.

‘Weekend breaks in Sydney’ is just one example. Imagine all the hundreds of similar qualified searches people could make to find your products or services. 880 people searched for ‘Sydney CBD hotels’ in June, but most advertisers fail to mention ‘CBD’ or even their location in their ads. ‘Sydney hotels the rocks’ had 390 searches, but only one advertiser mentions the phrase ‘The Rocks’ in their ads. 73 people searched for ‘3 star hotels in Sydney’ but only a handful of advertisers mention ‘3 star’ or ‘3*’ in their ads. These are people who know what they want are willing to part with their cash if they can find it.

It doesn’t take long to find hundreds of other examples of keywords that have significant search volume and are being poorly served. There are opportunities everywhere.

What’s more, as demand for better search results grows, people will start making more of these 3, 4, 5 and 6-word searches and expect better, relevant, more personalised results. The winners will be the advertisers who cater for them. The losers will be the ones who don’t.

So how do I go about improving the relevancy of my AdWords campaign?


I’m not going to pretend there is a quick overnight fix (because there isn’t). A highly relevant AdWords campaign takes patience, commitment and dedication.

Nor are there techniques that work for everyone. The whole purpose of this blog is to share with you the PPC techniques I have found to work in my experience, although I recognise they will be far from the be-all-and-end-all of paid search management so I welcome your ideas and comments. Paid search is an ongoing battle to become better and better, and it isn’t going to stop any time soon.

But to keep things nice and simple, here’s a quick 5 minute run-down of the essentials of creating a highly relevant PPC campaign:

Keyword Research

Research keywords that people are searching for. There are free tools out there, such as Google’s keyword tool, so use them. Build up a comprehensive keyword list. Not just with generic, high-volume keywords, such as ‘cheap Sydney hotels’, but also with long-tail keywords such as ‘cheap hotels in Sydney CBD’ and ‘cheap hotels Sydney Darling Harbour’. Long-tails can collectively be of significant volume and provide a great opportunity for tailored ads.

Then research negative keywords, lots of them. Why waste money on clicks you know are completely irrelevant? Use the keyword tool to identify keywords that might broad match to ‘cheap Sydney hotels’. Go through each result, making a note of anything you think is irrelevant. Is your Sydney hotel miles away from Sydney Airport? If so, add ‘airport’ as a negative keyword. Keep brainstorming negatives until you have at least a hundred.

Ad Group Structure

Once you have done your initial research, group your keywords into small, closely themed ad groups of generally no more than 20-30 keywords each. Write ad descriptions that are relevant to the ad group’s keywords and include the ad group’s keywords in your ads where possible. If you think you could write a more relevant ad for a keyword if the keyword was in its own ad group, split out that keyword into its own ad group and write a more relevant and tailored ad for it.

Think of the keyword as the question and the ad as the answer. Keep asking yourself, “If I searched for this keyword and saw this ad, is it answering my question?” If not, change it so it does.

Include offers and prices that are relevant to the keyword. In your ‘Sydney Hotels Christmas 2009’ ad group, how about mentioning Christmas 2009 prices or early booking discounts?


You’re trying to make your ads as relevant as possible to the user’s searches so think about how you could angle your products or services to appeal to users searching each of your keywords.

Next, deep-link your keywords to the most relevant page on your site. Don’t have a relevant landing page for a set of keywords? Write one.

Rinse and repeat until you have hundreds of ad groups, each with tailored ads and landing pages that match the keywords they contain.


Then get optimising. Test new keywords. New ads. New landing pages. Two keywords in the same ad group getting a lot of volume? Split the two keywords out into separate ad groups and write new ads that better match those keywords.

Run search query reports to highlight searches your keywords have broad-matched and phrase-matched to. Are they relevant? If so, ad them as new keywords in new ad groups and write tailored ads for them. If not, add them as negative keywords to prevent your ads showing for them again.

It may seem like a lot of work but it’s worth it. With patience, your CTR will start to increase. So will your Quality Score. People will start to spend longer on your site and view more pages. Returning visitors will rise as people decide to come back. Conversion rates will grow and sales volume will increase.

The Holy Grail of relevancy is not something you can achieve overnight, or achieve at all for that matter. It is only something you can strive towards. PPC success favours the dedicated. So keep testing and optimising.

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.

  • Pingback: SEOKudos()

  • hearmeout

    Nice article. Fascinating to see how much potential there is in paid search if it’s done properly.

  • Pingback: PPC keyword list building software Should Help Most Internet marketers | Online Business News, Ideas, Strategy()

  • http://inchoo.net Toni Anicic

    Great first post. Love the blog template, I’m also using it on one of my blogs.

    On topic:

    I believe that there’s something about PPC that even most experienced AdWords users tend to ignore: the changing habits of searchers.

    Most of the people will fine tune the campaign until they’re satisfied with conversions and simply ignore all the research after that. The truth is, searchers’ habits change over time. Some campaigns that use to be almost perfectly optimized a year ago might not do the so good today.

  • Alan Mitchell

    Completely agree Toni – I think not only are habits of searchers changing, but the average searcher is getting more intelligent. Every time I look through search queries I notice a higher frequency of “phrase quotes”, long-tails and unusual word orderings.

    If more people are searching like this: ‘hotels sydney cheap’, and less like this: ‘cheap hotels in sydney’, not only will tomorrow’s search marketers have to place more emphasis on different word orderings, but they will also have to reassess the importance of conjunctions such as ‘a’, ‘in’ and ‘and’.

    Constantly keeping up with changing searcher habits is by no means easy – as you rightly pointed out. But for advertisers who take the time out to really understand their customers’ changing needs, and adjust their PPC strategy accordingly, the rewards are great.

  • San Diego SEO

    Great article Alan. You nailed it. I appreciate your time and efforts in writing a helpful article.

  • http://www.atlantarealestateinfo.com/ Rob


    One of most well written posts on the subject I have read. I recently wrote a series of AdWords articles on a Blog I write for.

    Also, you guys are correct about how it all changes over time. I had a very good campaign going for about two years. It worked so well, I never revisited it or modified it.

    Recently, it started rolling over on me, so I went back in and reconfigured it, added 400+ KWs, reorganized, etc., etc., and it’s back up and running again, even better than before.

    You can ignore it for a while but eventually it starts to change.

  • Pingback: Twitted by alanmitchell()

  • http://righteousmarketing.com/the-holy-grail-of-search Holy Grail of Search

    Definitely agree on the need to test. I find that testing and optimizing are one of the key drivers of all other inputs to achieving great quality scores.

  • http://www.mikemeisner.com Mike Meisner

    Great post man. I have been hesitant to put my Adwords account to use, and although I’m fairly knowledgable about keyword research, long tail keywords, and SEO in general, I gleaned a couple good tips from this. I’m investing a lot of time and effort into a new ecommerce venture, selling a nice, profitable niche product, I want to put that account to use.

    It’s basic, I know, but the negative keywords will really help. Also I think ranking for and bidding on long tail phrases coupled with their higher conversion rate is totally worth the extra effort.

  • Alan Mitchell

    Hi Mike,

    Glad you found the article useful.

    Long tails are definitely a great way to achieve relevancy from paid search and provide searchers with highly-targeted and controlled ads. I did some research on long-tail keywords and found they can be up to 50% cheaper and convert up to 200% better than short-tail, generic keywords. Definately worth using them throughout the bulk of your campaigns.

    Good luck developing your niche. Remember to keep an eye out for which of your search queries are being matched to which of your ads to ensure you continually provide the best ads and user journey possible.


  • Pingback: Why The PPC Affiliate Model Is Flawed - Affiliate Marketing News & Information - Aff PR – All about Affiliate Marketing()

  • http://www.myruby.com Arnold

    Hi, great article, very helpful, especially as I have just started looking into AdWords PPC.

    However whenever I split up well similar keywords, and try to be EVEN MORE relevant, I get an ad conflicting error. The same goes for when I split them up to different match type groups, which I have heard is another good technique.

    Any ideas?

    Thank you

  • http://www.calculatemarketing.com/who-i-am/about-me.html Alan Mitchell

    Hi Arnold,

    I would suggest using the Ad Preview Tool to get more insight into the problems you’re having with your ads.

    Although with the highly-granular ad group structure it looks like you’re adopting, it sounds like there may be multiple keywords which could trigger certain search queries. If this is the case, Google, will trigger the keyword/ad combinationm they believe is the best match for the given search query.

    With broad match, it can be difficult to control which of keywords/ads are triggered for each search query. For example, two very different broad match keywords ‘cheap flights to hong kong’ and ‘low cost hong kong flights’ could potentially be triggered for the search ‘discount flight deals to hong kong’.

    Modified broad match is one solution to increase control over which keywords/ads are shown, while using strategic negative keywords at the ad group level will also help to funnel the right searches to the right ads.

    Splitting keywords with different match types into their own ad groups, as you suggest, is of course another stragegy you could try, although I’ve found it to only be beneficial for high-volume and more generic keywords. Having a seperate ad group for each match type could uneccessarily bloat your account for little benefit.

  • Pingback: The Facebook PPC Opportunity | Search Engine People | Toronto()

  • Pingback: How Will Siri Affect Your Brand’s/Startup’s SEO Strategy?()

  • Pingback: How Will Siri Affect Your Brand’s/Startup’s SEO Strategy?()

  • Pingback: How will Siri affect your brand's / startup's SEO & Digital Strategy? | Digital Brandy | Digital Brand Strategy & Metrics()