Running effective PPC campaigns in B2B markets is a competitive task that can be extremely expensive. That’s why it is so important for search marketers to run efficient campaigns that create maximum value for those high-cost clicks. So what can you do to be more competitive and increase ROI, even as these tough economic times are shredding your budgets?

Measurement. Better yet, effective measurement. Because, as we all know, you can’t improve what you don’t accurately measure.

If you’re like most search marketers, you’ve often wished for more insight into your campaigns than what AdWords alone can provide. You may also be looking for a ‘one-stop’ place to analyze all of your PPC data from all engines and efforts. Recent upgrades and enhancements to Google Analytics make both possible.

While those of us who rely heavily on Google Analytics (GA) are still getting up to speed on the new features released in October, it is important to get back to GA 101 and look at the features that can really give you the most bang for your PPC buck.

In this article, I’ll present four steps to improve PPC performance by using Google Analytics:

Before we continue, there are some assumptions and necessary steps for success. Most important, however, is that GA has been properly installed on the website pages that you are tracking. If you are unsure about this, please stop here and read this.

Another assumption is that you have used the GA URL Builder, or compatible application, to build your keywords for the other engines. More information on this is available within Google Analytics support documentation.

Linking your AdWords account to your GA account

The biggest benefit to using Google Analytics to measure you PPC efforts is its ability to have all of your data in one place. Linking your AdWords account to GA brings all of your PPC data to a single, customizable repository you can use for measurement and reporting.

Before you link your account, ensure that you’ve used the same Google Account email address for both your Analytics and AdWords account, and that the AdWords login email address has Admin access on the Analytics account. If you haven’t used the same Google Account, simply add your AdWords login email address to your Analytics account as an Account Admin.

• Go to Reporting>Google Analytics
• If you already have an Analytics account, click I already have a Google Analytics account
• From the Existing Google Analytics Account drop-down list, select the name of the Analytics account you’d like to link to
• Click Link Account
• Next time you log in to AdWords, go to  Reporting>Google Analytics and you should be linked up

Setting up your GA Dashboard

One of the greatest strengths of GA is the ability to customize your initial screen or dashboard view to display data that is most important to you every time you log in.

All GA reports have an “Add to Dashboard” icon in the upper left side of the screen. If you have a report that you use daily/monthly, click this button to add it to your front-page dashboard for easier access. The four fundamental reports that we will review should definitely be part of your GA Dashboard.

Setting up GA goals

As I said earlier, you can’t improve what you don’t accurately measure. And you can’t accurately measure your PPC performance without clearly defined goals or metrics for success. There are many factors that go into creating goals. You should consider the following issues when setting goals:

• Did you or the client help with these goals?
• Match types/Case Sensitive. Are they working?
• What are your conversion goals for the site?
• Do you have a good idea of what the user is doing while on your site to get to that goal?

Google Analytics is most powerful when you are tracking these additional steps as well as the goal itself. Be sure to get buy-in from your client or your boss before creating these goals. What you think is a good goal may not be right for them. You all need to be on the same page to prove the success of your campaigns.

Creating reports to measure success

The Google Analytics interface is full of pretty charts and graphs that can be intimidating to even the most seasoned search marketers. To turn that stream of data into useful and actionable information, you must create reports that give you the information you need when you need it. They can be canned GA reports or custom reports you create yourself. Either way, these reports will give you the ability to extract and present your successes and show you where there is room for improvement.

In Part two of this step by step guide, I will discuss four fundamental GA Reports to help you manage your campaigns.

There are two key factors crucial to assuring the success of any large SEM optimization process: First, tuning the campaign so that it is efficient and second, scaling the account while maintaining the efficiency you’ve gained. In any SEM account there will be a relatively small set of keywords that are naturally efficient (usually but not limited to branded terms) but that only drive limited volume. As a result, the goal is to use the tools available to identify incremental opportunity without losing profitability. In my work, I use three processes to identify qualified incremental opportunity: keyword expansion, impression share analyses and sophisticated bid management logic. When used together, these three processes can effectively identify opportunities for growth while keeping ROI/CPA under control.

The most basic approach to identifying incremental opportunity is through keyword expansion. Blindly expanding keyword sets has the significant risk of being unprofitable. However, if you have to expand blindly, be sure to implement a robust negative keyword set—limiting the potential detrimental impact of new, unproven terms. A more efficient methodology is to use the AdWords search query report to identify queries that have been broad or phrase matched by terms that are currently in your account. Adding terms which have converted (assuming the account is using Google conversion tracking) in exact and broad match pre-qualifies their relevance, mitigating risk as these terms are likely to continue to convert. While this process is not as fast, risk is negligible.

If you’ve had a meeting with a Google account team in the last 6 months or listened in on any holiday webinar, you’ve heard of impression share—a fantastic tool for identifying incremental traffic opportunities. Impression share (IS) metrics are available at the campaign level and consist of four unique data points:

1. Impression share: the percentage of impressions garnered relative to total impressions available
2. Lost IS (rank): the percentage of lost impressions due to rank
3. Lost IS (budget) the percentage of impressions lost due to budget limitations
4. Exact match IS: the percentage of exact match impressions garnered relative to total exact match impressions available.

If you take the sum of impression share, lost IS (rank) and lost IS (budget), the total will be 100%. These indicators are directional in the sense that they do not give truly actionable insight outside of whether you need to increase rank or budgets to get more impressions. That said, if you’re using Google conversion tracking or have the ability to track revenue/conversion data at a campaign level, it’s easy to see which campaigns are converting and have incremental volume opportunity.

This report starts to get really useful when you add an additional column, Ad distribution: with search partners. In general, Google.com tends to deliver more qualified traffic than its syndicated search partners, so with distribution-level insight at the campaign level, you can determine the average rank necessary to maximize Google.com traffic without maximizing syndicated traffic. Analyzing keyword level data to identify terms that are performing well with a rank that is above (the integer value is greater than) the “ideal” rank will provide an actionable data set to work with.

On the topic of ranking, before maximizing volume by pushing terms to a specific rank, it’s worth running a rank-based analysis to identify the ideal rank in terms of performance. Holding quality score, ad copy and landing pages constant, there is an “ideal” rank range where terms will maximize CTR relative to CPC (for more insight on ad rank calculations, see my post, how to get more PPC traffic for less money). In other words, analyzing the relationships between rank vs. CTR and rank vs CPC will yield a range (for example, 2.5 to 3.5) which has the most lucrative ratio of CTR to CPC. The image below shows an example using branded data:

When determining the ideal range, be sure to factor in conversion data to identify the max CPC you are willing to pay (through substitution and knowledge of the AdWords ad rank algorithm, we can determine: Max bid = goal ROI * average order value * conversion rate). When looking to maximize volume, aim for a rank that is at the more aggressive end of the ideal rank range. As this is simply a rank-based analysis, conversion rate is not expected to change as rank changes. If the goal is to boost conversion rates, other optimization techniques such as pre-qualifying traffic via ad copy and landing page tests are more likely to help.

The third tool for scaling with efficiency is bid management. It’s easy to lose track of ROI when managing terms to a given rank. This is why it’s important to identify a range within which you can manage for efficiency. Adjust bids to keep average rank at the keyword or match type in the ideal range (for example between ranks 2 and 7 to ensure first page exposure), then you can increase bids when ROI is above goal and decrease bids when ROI is below goal. This task can be tedious, which is why AdWords has developed conversion optimizers and rank-based optimizers (however both tools cannot be used in tandem). This is why most SEM software providers will allow logical rules to be implemented at the keyword or even match type level to automate the process, respecting the rank and efficiency parameters.

There’s a lot to think about when trying to efficiently scale an SEM effort. As always, when using each of these tools, often times together, don’t lose track of the underlying goals of the account (ROI, CPA, CPC, etc) as a byproduct of driving incremental volume.

Can you apply the spirit of social media to other marketing channels?

At this year’s SMX East, after my presentation on Landing Page Usefulness—emphasizing a “usefulness” mission over “usability” tactics—it struck me: great landing pages can bring many of the ideals of social media to paid search marketing campaigns.

Here are five principles of social media marketing that can energize your landing page program:

1. Engage in specific conversations, not generic one-size-fits-all talk.

When a company engages in social media, the worst thing it can do is echo canned, cut-and-paste responses to every incoming comment. It’s painful just to imagine! Yet many paid search marketing campaigns commit that very faux pas: a user clicks on a keyword/ad combination with a specific promise, and then they are unceremoniously tossed to a general-purpose page. Such “message mismatch” between keywords/ads and their associated landing pages damages brands and hobbles conversion rates.

The reason I advocate deploying dozens—or even hundreds—of landing pages is because doing so lets you deliver focused and well-matched introductory dialogues with respondents, framed in their terms. As I said in my presentation, the goal is have respondents exclaim, “thank you, that was exactly what I was looking for!”

It’s not about optimizing one page to rule them all—an illusory, marketer-centric fantasy—but deploying many separate pages that each speak authentically to their niche. That’s the kind of respect that honest social media marketing shows to people reaching out to you, and a good landing page strategy can live up to the spirit of that goal.

2. Embrace “constant content,” continually releasing new ideas out into the world.

From blogging to tweeting, the engine of social media is the frequent generation of content. Hopefully it doesn’t take a committee or half a dozen pairs of hands to put up a new blog post or to update your Facebook fan page. The incentives in social media are to be fast, prolific, experimental, relevant and real.

The same tenets should apply to landing pages.

Sometimes, when I suggest that people should publish dozens or hundreds of landing pages, I get a look of incredulity: how could we ever create so many landing pages? Yet organizations who embrace social media marketing produce 10-times as much content without breaking a sweat. The resistance to such agile production of landing pages is often a hang-up from the bygone days of long-cycle web development. Today, deploying new landing pages should be as easy as—maybe even easier than—posting to your blog.

If you have a good content management system (CMS), a nice collection of page design templates, a shared library of images, maybe a few reusable Flash components, and a standardized mechanism for data collection and analytics tracking, then you’re ready to crank out landing pages on demand. And if you don’t have all of those pieces yet, none of them are particularly difficult to put in place.

3. Harness fast feedback to learn about your audience.

Arguably the best feature of social media is that it lets you tap into candid and immediate feedback from your market, albeit in an unstructured manner. It’s a wonderful environment to put ideas out into the community and quickly gauge reaction.

However, you can also solicit a different kind of feedback—more quantifiable and more directly connected to sales—through rapid experimentation with landing pages and keyword buys. Participation is more predictable with such PPC experiments, and the results can be easily benchmarked against your e-commerce or lead funnel metrics. It’s a small, low-risk investment that can help you discover big wins.

Struck with a novel theory about an unaddressed customer segment over your morning coffee? Don’t just hypothesize about it or file it for the next quarterly planning meeting. Launch a targeted search ad and tightly matched landing page for it before lunch and have real-world feedback by the next day. It doesn’t have to be perfect. You can test and tweak as you go along—an ongoing feedback loop.

Ads and landing pages also lend themselves to A/B tests, in a more controlled fashion than variations in social media tactics. If you structure your tests with good hypotheses, you can learn a lot about audience preferences and personas.

4. Open up a dialogue by asking relevant questions—and respecting the answers.

Social media is a conversation, not a soliloquy. People can ask questions, usually quite informally, to help identify the content or information that’s most relevant to their interests. This allows a single discussion to adapt itself to many different participants.

A similar dynamic can be achieved with landing pages. Sometimes, you have to field clicks from keywords/ads that appeal to several different segments of respondents. Instead of reducing the specificity of your content to a bland common denominator—the ill-fated, one-page-to-rule-them-all approach—start by offering them a few meaningful choices. Are you more interested in A, B, or C? Based on their one-click selection, you then deliver more detailed content that’s tailored to their needs.

This technique is known as multi-step landing pages or conversion paths. It can be a tremendous source of feedback, especially when you test different types of choices. However, it’s crucial that the choices genuinely help respondents find what is most useful to them—you want segmentation that benefits users, not just marketers. Remember, we’re striving for that “thank you, that was exactly what I was looking for” effect.

5. Champion transparency and authenticity over cleverness and technology.

The essence of social media is its authenticity, plain and simple. You can try to manipulate it with gimmicks and complicated machinery, but such machinations tend to fall flat. People love what’s real in social media, not what’s artificially crafted to appear real. Human trust is more important than plastic perfection.

Certainly this holds true with landing pages as well. There’s no shortage of sophisticated software you can use to dynamically alter your pages to users based on their IP address or behavioral profile. You can layer rules upon rules to calculate the optimal offer for each respondent. But inevitably, such overly processed experiences lose their authenticity.

Similarly, you can play UI tricks to try to force people to engage with your page (e.g., you must fill out this form before continuing!), but it’s almost always more of a turn-off than a successful hard-sell tactic. If you’re going to remove your regular navigation choices from a landing page, do so because it helps eliminate clutter for a respondent in that context—but still always give them an option to easily jump to your main site.

Be genuine, creative, open, and enthusiastic in your landing pages, and you will win more converts.

Landing pages, like social media, are something that you get better at by doing. So release your inhibitions and make more landing pages.

The importance and impact of online testing on your conversion rates cannot be understated. Yet as powerful as testing can be, it is a double-edged sword that can actually set you back if you are not careful. Poorly designed tests can take years to complete, and even worse, they might not provide concrete insights to what elements will convert more visitors into customers. Here are six of the most common mistakes I see clients fall into while conducting online testing.

Using the wrong type of test (A/B vs. multivariate test)

A/B tests allow you to test two versions (or more) of an entire page against each other to see which one works best. If you are testing three different versions of a page, then you are conducting an A/B/C test and so on. A/B tests are especially useful because they allow you to test major design decisions by putting two or three completely different designs against each other to find out which one of these works best with your visitors.

Multivariate tests (MVT) allow you to test multiple elements of a single page at the same time. So you are able to test different headlines, different images, or different colors on a single page. Although it may seem like testing different elements on one page is less complex than testing one or more pages against each other, the opposite is true. You can think of A/B tests as simplified versions of multivariate tests.

One of the first questions you must ask yourself is whether you should start with an A/B or multivariate test for a particular page. And while multivariate tests are indeed powerful, they require considerably larger number of visitors to the test page for you to reach a decisive conclusion.

So, should you go with A/B or multivariate testing for your particular case? There is no universal answer. If you are starting out with a brand new landing page or have limited number of visitors, we usually recommend conducting an A/B experiment to start out. However, if you already have an existing page, we might conduct a small multivariate test (less than 12 or so different scenarios). The goal of this type of test is to learn which elements have the highest impact on conversion. Based on the results we learn from that MVT, the team will decide to conduct further MVT or A/B tests.

A successful test will take into account the number of visitors and number of conversions on a page (not the site). Although your site may have 30,000 visitors a month, your test page may have as few as 500 visitors and one conversion per month.

A/B tests are good for alternate designs of an entire page, whereas multivariate tests are helpful in determining the most successful elements at different locations on a page.

Testing too many elements

In the case against multivariate testing, I discussed the perils of letting multivariate testing software do the thinking instead of using an online marketer’s judgment, intuition and persuasion to guide the testing process.

There are plenty of software solutions out there which allow you to test tens of thousands of combinations on a single page—there is software that actually allows you to test millions of combinations. The problem with testing so many variations is the amount of time and resources it will take to conduct such tests.

And yes, some websites have sufficient traffic that they feel testing tens of thousands of combinations is just fine.

I disagree. There is no art in testing millions of scenarios and hoping that one of them will convert better than the current page you have. I’ve found that testing fewer than 100 scenarios can increase conversion rates from 4% to 15% in a matter of a few months. It’s not about testing only, it’s taking a common-sense, holistic approach to the process testing that leads to improved results.

Running tests that take too long

There are two kinds of factors that can have an impact on your conversion rate. There are internal factors that you can control, such as your design, messaging, copy, etc. But there are also external factors over which you have little control. If your competitor is running a 50% discount sale, you should expect your conversion rates to suffer. If the economy gets worse, you can expect your conversion rates to suffer as well.

One of the dangers of conducting testing is not accounting for the impact of these external factors. There is no real way of controlling them; however, there are few things you can do to minimize their impact:

• Keep track of external factors and understand that they might have an impact on your results. Being aware and not being sidetracked by a plummeting economy is probably a good idea.
• Minimize the time it takes to run your experiments. We do not advise conducting tests that take months to complete. I usually like to see tests completed in four weeks at the most. If that is not possible, then go back to the drawing board and rethink your test scenarios.
• Stay on top of what your competitors are doing. There are several tools that allow you to “spy” on competitor activity, including whether or not they are employing testing or some type of conversion optimization. I’ll write about those tools in a later article.

Failing to monitor your test as it takes place

Conventional wisdom states that you can set up an experiment, start the test and just sit around and wait for the results. That works well in many cases. But there are times where you might want to change an experiment in mid-stream. Closely monitor experiments to decide if you should eliminate some elements when there is enough evidence that they are not producing the results you are looking for.

This point is especially applicable when you’re using Google’s website optimizer as your testing tool. Website Optimizer is a great testing tool but running experiments on it may take too long. Now, you must be careful, seeing that one scenario is not producing any results after only 20 or so visitors does not mean anything. You also have to know when it is the right to time to interfere.

Failing to conduct follow-up experiments

Let’s say that you conducted a test, and as a result you were able to double your conversion rates. But then you stopped testing. That is a typical story.

We had a client who started with a conversion rate of around 3%. After we did the initial redesign and testing, we were able to increase their conversion rates to 9.8%. Our clients was so pleased with the results, they decided to suspend the campaign they already paid for.

It took about another month to convince them that they should continue testing and that there is the potential to increase conversion rates even further. Designing follow up experiments, and learning which elements worked and which elements did not, is at the heart of conversion optimization. To make a long story short, the client was able to increase their conversion rates to 14.9% with just two more months of testing.

Thinking that testing is a silver bullet

Testing is a great tool to increase your conversion rates. It is also an important step in any conversion rate optimization process you conduct. However, it is only one step in the process. It is a critical step that should happen at the end of conversion optimization projects. If you are looking for double digit conversion rates, then you must start with understanding your target market through the process of persona creation, site analysis, analytics assessments and design and copy creation. Trying to test random scenarios without doing the initial homework is like throwing darts in the dark.

To recap:

• Select the right type of test to conduct based on the number of visitors and other data collected from you analytics.
• Limit the number of scenarios through a holistic approach to testing by considering market segment, persona development, trust factors, fear factors, etc.
• Don’t be side-tracked by external factors that impact your conversion rate. Stay on top of how the economy is performing and what your competitors are doing so you can test and optimize accordingly.
• Stay involved in the tests you are conducting. Don’t minimize the importance of following up and tweaking test elements as results become statistically significant.
• Don’t stop once you see improvements. Get beyond single-digit improvement by continuing to test.
• Approach any project from a number of different angles, because testing alone will not maximize your results.

Choosing appropriate match types for PPC campaigns is one of the most important decisions an advertiser can make, having an impact on such diverse factors as costs, quality scores and ultimately overall success. Advertisers may choose to set their keywords to either “broad,” “phrase” or “exact” match types, each used for a different purpose.

Generally, setting up an account with only phrase or exact matched keywords is a safer tactic than setting all keywords to broad match—broad match, as it implies, generally serves up far more keyword variations in an ad than the other two narrower types. However, we’ve found that a mix of keywords, leveraging multiple match types, coupled with strategic bids, can significantly enhance your account’s performance.

The problem with relying on a single match type is that no single match type offers a perfect solution and each presents its opportunities and risks. Let’s look at the pros and cons of each.

Broad match type

• Pros: Allows for the widest and most varied contextual reach
• Cons: Susceptible to over-exposed keywords, broad match “masks” CTR for terms that may have triggered an ad for a phrase or exact matched query.

Takeaways: By leveraging search query reports, broach match can be a great source for generating new keywords (as well as negative keywords)—but if not monitored closely can lead to excessive spending on irrelevant keywords and suggest misguided CTRs during analysis.

Phrase match type

• Pros: Offers additional control over when your ads are triggered (contextually speaking)
• Cons: Limits your ability to attract clickthrus on slight variations where search query order is not consistent with your phrase-matched term.

Takeaways: Offers more control over keywords and ads, opportunity to discover new keywords, however variations are restricted to the order of the keyword phrase in the AdWords account.

Exact match type

• Pros: Offers precise contextual control over when keywords and ads are shown
• Cons: Most limiting of all match-types, keywords will show only when query is searched on exactly as specified in your AdWords account.

Takeaway: Offers more control, though the possibility of generating clicks on varied search queries is almost entirely nullified, which limits contextual reach substantially.

Now that we understand the pros and cons of each match type, let’s look at a hypothetical example where using a single match type vs. multiple match types can have an effect on your account performance.

Let’s imagine we operate an online guitar store and (for simplicity’s sake) we are bidding on the keywords “acoustic guitar” and “acoustic guitars.”

According to Google’s keyword research tool, there were over 2.6 million searches for these two keywords in August.

Obviously there is no shortage in traffic, but it’s more important to understand what type of traffic you may be exposed to if you bid solely using broad match.

Under the broad match setting, all of the keywords listed above (plus many other variations including conceptually related terms like classical guitars and perhaps banjo) may trigger your ad. Some are potentially relevant and others are not. This introduces a significant account-management problem. No longer is our CTR based on the keywords “acoustic guitar” and “acoustic guitars,” but rather we are managing to an aggregate CTR for all variations triggered by these keywords.

By adding a phrase and/or exact matched version of “acoustic guitar” & “acoustic guitars,” you reduce your exposure substantially—only queries containing the words acoustic & guitar(s), in that order, will trigger your ad for these variations. In doing so, we were able to create a more efficient channel for tracking our keywords without compromising the broad-matched terms.

Won’t multiple match types for the same keyword cause confusion in AdWords auctions? No. According to Google’s help documentation:

”…the more restrictive match type will always trigger the ad, regardless of CPC bids. For instance, if the broad-matched keyword ‘apple’ and the exact-matched keyword ‘apple’ both existed in your account, the exact match would always trigger an ad.”

You may still consider bidding higher on phrase and exact matched keywords in order to rank more competitively. These keywords are naturally better targeted to your audience’s search query and may result in higher conversion rates.

In addition to adding phrase and/or exact matched versions of your keywords you should also constantly monitor your broad matched and phrase matched search query results and supplement your campaigns with appropriate negative keywords. This will help fine-tune your broad and phrase matched keywords and limit irrelevant spending.

By following these simple tactics, over time you will be able to rely to a lesser extent on broad matched terms, decrease overall costs and better target your campaigns via phrase and exact match keywords.

To learn more about negative keyword phrases and keyword research tools see my blog posts on negative keywords and keyword research tools.

Search advertising success is defined by relevance. The concept of relevance is broad, but logical: it’s the relationships between advertiser, keyword, ad copy and the landing page/ post click experience. So, how do we know if our relationships are strong, and our account is relevant?

Simple: Quality Score for Google, or Quality Index for Yahoo.

There are several components of Quality Score, illustrated quite nicely below:

Three of the six components are controllable (keyword relevance, ad relevance, and landing page content); two of the components are byproducts (clickthrough rate (CTR) and historical performance); and the last (other relevancy factors) is simply AdWords reserving the right to make subjective decisions and dash a little bit of their secret sauce on the formula. We know that Quality Score is important, otherwise it wouldn’t exist… but how important is it? The AdWords team recently upgraded the help center to include several articles on how Google uses Quality Score. Most notably, they provide a clean equation for how Quality Score is used to calculate position:

Ad Rank = Quality Score * Max Bid

As each auction (impression) is dynamic (different advertisers, bids, etc), this equation is not linear. That said it is very logical that if there are two advertisers in an auction and advertiser A has a Quality Score of 10 and advertiser B has a Quality Score of 5, advertiser B must bid twice what advertiser A bids to achieve rank 1. Taking this one step further, let’s assume there are several players in the auction and it becomes clear that by improving Quality Score, an advertiser in a lower rank with a relatively low impression share can get the same, if not more traffic for less money. This concept is by no means revolutionary, but it’s helpful to see a real application of Quality Score and further understand its importance.

Now that Quality Score is defined, we can focus on how to maximize the relationships between keyword, ad copy and landing page content—getting these all right will greatly contribute to achieving a high Quality Score.

My previous post on organizing a PPC account for maximum success was centered on creating relevant ad groups and campaigns (at the AdGroup/keyword level—I’m defining relevant terms that can use the same piece of ad copy). This is because Quality Score is a product of more than a single keyword and its ad copy and landing page—the “other relevancy factors” mentioned previously include advertiser legitimacy. That is, whether or not the advertiser is relevant to the keyword being bid on and the advertisement being displayed on the search engine. What this means is that if an advertiser is bidding on what appear to be arbitrary keywords (from the search engine perspective) then it will be more difficult even for relevant keywords to achieve the maximum Quality Score.

Assuming that an account is well organized, ad copy should be uniquely tailored to each AdGroup, simultaneously qualifying clicks while containing compelling messaging. Every impression should generate bold text in the ad (achieved when ad copy contains one or more words from the actual query) yet not appear to be dynamically generated using generic dynamic keyword insertion (DKI). Because DKI injects the query into the ad, it’s a very useful tool when keywords in a given AdGroup are not perfectly similar. This is not to take away from the value of DKI, but generic ads may indicate that an advertiser offers something that is not true, leading to lots of clicks and no conversions. If using DKI, be sure to qualify users with the non-DKI portion of the ad and a strong negative keyword set to ensure honest messaging and mitigate bounce rates.

Whether or not DKI is being used, the landing page plays a significant role in determining relevance. Search engines are looking for the keywords(s) from both query and ad to also be present on the landing page—in text, not in an image. Also, landing pages should not be cluttered. They should feature information about the service or product being offered as well as a clean, direct route to conversion, whether the conversion event is a purchase, registration or additional pageviews. Don’t try to push too much: focus on the message being advertised and the product or service being offered. Furthermore, unless unavoidable (or the business offers only one product or service), the homepage should not be used as the landing page because homepages tend to be generic and focused around the business as a whole instead of the product being advertised (an exception would be ads using brand terms, where searchers would expect to be delivered to the brand’s home page). And last but not least, test and test, and then test again. Testing various combinations of ads and landing pages is critical to constantly improving results while maintaining fresh messaging.

There are a multitude of factors that determine success in search marketing ranging from intangibles (such as advertiser quality) to data driven dynamic strategy development. One can spend thousands of hours micro-managing campaigns only to realize the more work you do, the more work there is to be done. Don’t you wish there was a cheat code? A game genie for SEM? Unfortunately, there isn’t a set-it and forget tool that has proven to function perfectly, so until that’s been developed, we have to settle for best practices with the tools we’ve got at our disposal.

In the first two posts in this series, I covered methodologies behind a match type analysis and prioritizing efforts via top keywords. In this post, I’m going take a step back to discuss organizing an account, the ramifications on relevance and in turn, quality score.

Account organization is the first and most important step in the optimization process—it’s the foundation for success. A well organized account allows effective and efficient management of an SEM effort, freeing up resources that would otherwise be bogged down in the ongoing maintenance of the account. So, what is an organized account? And what differentiates an organized account from an unorganized account? These are good questions and I’ve devised a short questionnaire below to determine if your account is well organized.

At the campaign level:

1. Are brand/branded terms isolated from non-branded terms?
2. Is content match isolated into unique campaign(s)?
3. Are site placements isolated into unique campaigns?
4. Is there a unique campaign for each product/product line?

At the ad group level:

1. Is there visibility into top performing keywords?
2. Is each ad group focused on a unique product?
3. Does associated creative speak to the product?

At the keyword level:

1. Do all terms in a given ad group have a similar quality score?

If you answered “no” to any of the questions above, then there is work to be done. You may have caught onto the theme of my questions: more organization is better than less. That said, there’s a fine line between being well organized and going overboard. Furthermore, not all advertisers are the same, so to understand the logic of organization, we must think about advertiser and search engine limitations as well as the tools available for optimization.

For each advertiser, there are unique limitations. Some advertisers represent one store whereas other advertisers act as an affiliate/agency representing multiple stores and/or product lines (as an example, think of the online Apple store versus Best Buy, or The North Face versus REI). Representing one store is definitely the more simplistic model as there is only one brand to manage. Nonetheless, in both cases, all brand/branded terms should be isolated into a unique campaign. If representing several advertisers, then simply breaking brands into unique ad groups should do the trick. However, if there are too many terms (more than 30 or 40 per brand—this is a guesstimate of a maximum number of terms per ad group based on the logic that it’s difficult to have more than 40 highly related keywords) then it may be necessary to make a unique campaign per brand.

Because quality score is calculated at every possible level (keyword, ad group, campaign and account), it’s important to try to make each level as organized and relevant as possible. So, if you do represent multiple brands, then having unique accounts for each property/brand will boost the overall quality score of each account as relevance across the account will increase. If you are a more significant spender and have the benefit of support from the search engines directly, then request a master my client center account (MCC) from the AdWords team and have them generate unique accounts for each of your brands/properties. An MCC provides the convenience of one consolidated account (dashboard and reporting) with the benefits of unique accounts.

Quality score is determined from the relationship between the exact match of a term (the term itself), the ad copy associated with it, the landing page and historical CTR (assuming there is history). As a result, all match type iterations of a given term (exact, phrase, and broad) can live in the same ad group because the quality score will be the same. The only match types that should be isolated (for both dashboard and optimization purposes) are content match and site specific placements. All content match terms should be in unique campaigns; as content key performance indicators (KPIs) dilute campaign and ad group level data so significantly that it becomes almost useless. Similar to content, site placements are unique and optimizations should not be made in conjunction with search.

In terms of campaign level organization, each product or product group should have a unique campaign (depending on size constraints). Within each product based campaign, ad groups should be organized by type: navigational, informational, and transactional. Top performing keywords should be isolated into unique ad groups for enhanced control. Ad copy should be relevant at the ad group level and specifically tailored to the user’s stage in the buy cycle (informational, navigational, transactional). If possible, use product specific landing pages that contain user-friendly information regarding the product as well as a simple click path to purchase. If all of these processes are in place, then quality score should be consistent within each ad group.

While everything I have mentioned is designed to help in the long run, there is a significant caveat: overhauling account structure is a time consuming process and if not executed properly, quality score will be negatively affected. For this reason, whenever we overhaul an account at our agency, we start with top keyword isolation and ad group level changes as to maintain the keyword/ad copy/landing page relationships, thereby preserving quality score on the highest volume terms. Mirroring campaigns and overhauling [the newly mirrored campaign] is also an effective best practice as it’s easy to revert in case the short term negative effects on QS are significant.

Highly profitable PPC accounts aren’t for everyone.  Here are some tongue-in-cheek tips you can implement to play it safe, reduce traffic and lower your profits with a PPC advertising campaign.

1. Set a low daily budget

If you can get a few conversions a day with a $100 daily budget, shouldn’t you just be content?  If you raise your budget, some crazy clicking event may bankrupt you.  Better to set your budgets low, then you’ll never have to worry about whether or not you can pay your PPC bill.

Yes, a worse-case scenario would also involve more traffic to your site and likely more conversions.  But what if the extra traffic doesn’t convert?  Play it safe: decide how many conversions will make you happy and set your budget accordingly.  You won’t be able to take advantage of news-driven spikes in traffic, but what you don’t know won’t hurt you.

2. Don’t bid on keywords where you rank well organically

PPC should never be allowed to steal from SEO, even if it brings you more total traffic and makes you more money.  You probably have pretty good organic listings for some keywords, especially your own brand name.  This is free traffic.  You spent a lot of time on SEO and now’s your chance to reap the rewards.  It would be stupid to pay Google to have your ad listed in the search results when you are already there organically. Besides, “nobody” clicks on search ads. Google’s billions in annual revenue from paid search is just an anomaly that you should ignore.

If you were to bid on your company name, many people would click on your ad when they would have clicked on your free listing. Others are trying to grab all the profitable traffic they can get, even if they have to pay for it.  Instead, why not just be happy with the money saved.

3. Stick with your current sales funnel

Are you currently making sales from your website?  If it ain’t broke, don’t fix it.  Too many people are discontent with their conversion rate and are always trying something new to get more conversions.  There are so many things that could go wrong when you mess with what’s working… I go crazy even thinking about it.

Yes, if you had a higher conversion rate you could afford to increase your bids to get more traffic.  Testing new landing pages, however can sometimes actually lower the conversion rate.  Besides that, technical problems may arise when adding new pages, sometimes effectively breaking your website.  So don’t mess with a working website that converts.

4. Don’t bid on competitor keywords

Big companies spend a large portion of their marketing budget on branding campaigns in hopes someone will search for them directly by name, land on their website and purchase a product.  You can understand their frustration when another company stands in at the last second to take the sale.  That’s why many PPC advertisers play nice and won’t bid on competitor brand terms.  If you don’t like risk, this is the route you should take.

It doesn’t matter that some of your cheapest conversions from highly qualified traffic will likely come from competitor terms, or that your competitors are probably bidding on your brand name.  Play it safe: competition is good so long as competitors don’t get mad at you.

5. Set bids low enough to leave a healthy initial profit

Once the initial sale takes place, there is only one thing that is certain: the initial sale.  Why pay more for clicks than you can easily pay for with front-end sales.  Additional sales to your customers may or may not happen.  Past performance is no indication of future returns.  So if you can’t make a sale, pay your PPC bill, and still have a healthy profit margin, you need to lower your bids.  Any extra sales to your existing customers are gravy.

In fact, some high-spending companies actually plan to lose money on the initial sale.  That’s why they can afford those outrageous bids at the top.  Surely they are setting themselves up for disaster if those “future” sales don’t pan out.  Not you.  You’re safe with an average position of 6.  Who needs a backend when you’ve got a good front-end?

Take these suggestions if you don’t like to upset the apple cart.  If you’re a contrarian and think you can increase your PPC campaign’s effectiveness, read the converse view: 5 Calculated Risks to Strengthen Your PPC Advertising.

Not all keywords are created equal. The Pareto principle states that roughly 80% of effects will come from 20% of the causes. Search is no exception to the rule. In fact, search marketing efforts often exhibit an exaggeration of the principle because it’s very common to observe 1% of the keywords in a given account driving 50% of the cost and even more of the revenue.

So, what does this mean? Behavior is unique for these high volume/value keywords and as a result, steps need to be taken to make these terms more visible for daily account management and analysis. The first step is to identify and isolate these top terms to help prioritize where to focus.

Here’s a simple exercise to identify the top terms in an account:

1. Run a keyword report and sort the data by cost.
2. Filter out branded terms, as these terms should be isolated in a unique campaign.
3. Filter out terms that constitute the top 50% of the non-banded cost.

Typically this keyword set is very small, but accounts for the same cost as the remainder of the account. It’s important to note that the top terms as I have defined them only refer to top spending terms. If your SEM effort is ROI focused, at this point, take the opportunity to identify if there are terms that are in this keyword set that do not drive revenue.

Now we’ve got this keyword set comprised of the most significant keywords in an account. What to do with it? It has been rumored that isolating a keyword in a unique AdGroup will increase quality score. While this is not necessarily true, it highlights the reason why it’s important to have an organized account: control. By placing each of these top terms in a unique AdGroup, you can create keyword/ad/landing page associations that are more relevant—effectively increasing Quality Score, clickthrough rates (CTR), and subsequently driving down cost per click (CPC). While it seems logical to do this for all terms in an account, it’s not a scalable solution for an account with thousands, if not tens of thousands or more terms. Thus we have identified a small sub-section of terms, our “top terms,” that have a very significant effect on the account. Because this set of terms is a cross-section of the account, we can take successful elements, such as ad copy and landing page combinations, from the top terms and apply them to similar terms in the rest of the account to increase CTR and Quality Score.

As I mentioned before, another important set of terms to isolate and manage independently from the rest of the account is brand terms. Brand terms generally have very high Quality Scores, are cheap, have great CTRs, and observe the best conversion rates, often making them the most efficient terms in an account.

As a result, you can maximize the impact of brand terms by activating them on all match types: exact, phrase and broad (for more on this, see my earlier post Best Practices For Auditing An SEM Account). The same rules for optimization apply: isolate high volume terms into unique AdGroups within the brand terms campaign, create relevant ad copy, and make sure landing pages are as relevant as possible.

Side note: Josh Dreller has been posting articles recently about ways to use excel for SEM analysis beyond standard reporting. In his first post in this series, How To: Excel at Excel for SEM Applications, Part 1, he outlines a process to create a catalog system for keywords so you can pivot your data to display in various ways. Building on his catalog concept, you can set up a label (column) for these top keywords, allowing rolled-up reporting on top terms vs the rest of the account. Furthermore, you can build a second label for Brand and Non-Brand terms. Now you can analyze brand vs. non brand as a sub-set of top keywords vs. the rest of the account. Very cool.

Identifying top terms in both brand and non-brand keyword sets will create a very controlled environment to test and optimize ad copy and landing pages. Small changes in quality score and subsequently CPC will have a very significant impact on the account and the best practices identified in the top terms effort can be rolled out across the campaign.

One of the toughest challenges facing pay-per-click marketers is keyword expansion. Marketers need to be adept at selecting keywords that will not only drive traffic, but more importantly, will drive conversions in a cost-effective manner. As competition intensifies, identifying unique, high quality keywords becomes increasingly important-and difficult. While basic keyword research remains important, search marketers need to look to other tools to gain an edge.

Basic keyword research

Keyword research generally begins with one of the popular keyword-research tools. These include-but are not limited to-Google’s keyword tool, Wordtracker and Keyworddiscovery. These tools allow marketers to research topics, identify search-volume data, and find additional permutations of their “seed keyword.” To illustrate my point, a search in the free version of Wordtracker for “dog training” produces the following:

This list is a great starting point for anyone promoting a dog training service. It includes several highly searched keywords relating to “dog training” as well as a vast number of less searched terms as you move down the “tail.”

While basic keyword tools help marketers generate keyword ideas for their pay-per-click campaigns, they also create a huge problem. As most keyword tools are readily available to the public, there is nothing prohibiting your competitors from including the same highly trafficked keywords in their campaigns. Naturally, as more and more competitors enter your niche, the costs-per-click begin to rise. Following this keyword research methodology will undoubtedly leave you with a long-but thin-list of keywords, not to mention that you will be paying too much for “overbought” keywords.

So what is a PPC marketer to do?

As competition increases, intelligent PPC marketers realize that it is imperative to not only build out long keyword lists, but to also “widen” those same keyword lists. What do I mean by “widen?” Let’s revisit the dog training example from above:

Traditional keyword tools suggest that you include terms like:

• dog training
• dog training tips
• dog obedience training

In general, these potentially useful keywords speak directly to your target audience’s needs. But these are also keywords that are most likely to be bid on by your competitors.

What if we dissected this keyword list and expanded on the term “dog?” Think about how many breeds of dog exist: Golden Retrievers, German Shepherds, Beagles etc. Now, lets replace the word “dog” with specific breeds:

“dog training” and “training a dog” become “Golden Retriever training” or “training a Golden Retriever.”

By incorporating breeds and replacing the term “dog,” we were able to exponentially increase the size of our keyword list.

Some simple math to consider

If you choose to focus on 10 dog breeds and combine those with 50 seed dog-training keywords, the result would be 550 unique keywords! 10 unique dog breeds * 50 seed keywords = 500 + 50 original seed keywords, for a total of 550 keywords. Additionally, the new breed keywords probably have less competition and are much more likely to convert with a targeted landing page (a topic for another time).

New tools from Google can help

Google Sets and Google Squared are amazing free tools that will allow you to quickly identify elements of a set.

Here are examples of both in action:

To summarize, basic keyword tools are excellent for exploring generic topics and developing a set of keyword permutations. However, in order to differentiate your keyword list, and to avoid complete reliance on highly competitive keywords, I encourage all PPC marketers take steps to widen their keyword lists by utilizing Google Sets and Google Squared. This approach will increase the likelihood that you will find less competitive, and more targeted keywords that ultimately produce conversions at more cost-effective rates.