Google has a bidding methodology called Budget Optimizer that attempts to maximize the traffic you receive for the keywords in a campaign. This is useful for early buying cycle keywords. However, every keyword should be reaching some goal regardless of where it falls into the buying cycle. It was difficult to track the effectiveness of these campaigns until recently when Google made some changes to Google Analytics.

Now you can more effectively bid on early buying cycle keywords, or keywords that you want exposure for, but do not have direct returns by combining the new Google Analytics goals with a budget optimizer campaign.

Google Analytics changes explained

Google Analytics expanded the number of goals you can track from four to 20. However, it looks a bit different as you now have “goal sets.” Where each goal set has 4 goals, and you can have 5 sets per profile.

Second, Google added two additional goals:

• Page views per visit
• Time on site

Please remember with Google Analytics, time on site actually means time on site minus the last page visited, as Google analytics only tracks data when there is a click. For instance, if your site visitor exhibited this behavior:

• Page 1: Time 5:00
• Page 2: Time 2:00
• Page 3: Time 1:00
• Exit via closing the browser window

In this example, Google does not know the browser window was closed and that they should stop tracking time on site. Therefore, Google Analytics would report 3 page views and 7 minutes on site for this user.

Using goals to measure branded or early buying funnel keywords

I’m a fan of bucketing keywords by where they fall in the buying cycle. If someone is in the learning phase of the buying cycle they need an informational page. If they are in the buy section of the funnel you need to monetize the click by encouraging them to buy.

This creates situations where keywords (and often display or content ads) reach a user who is at a different point in the funnel than the ads you’re displaying and you might not be able to monetize the user—at least for now.

For instance, the keyword “plasma vs LCD TV” is a learning cycle keyword. Buying a high-end electronic device is expensive, and takes thought and time before even beginning initial searches, much less actual purchases. Therefore, this keyword will rarely lead to a purchase as its a research oriented search. However, that does not mean it’s a bad keyword. When you have keywords that describe your products or services, but cannot be directly monetized, you have three options:

• Set a different conversion goal for the keyword, such as a newsletter subscription; and then set a value for what a newsletter subscription is worth
• Use an attribution management bid system
• Use a budget optimizer campaign

Use budget optimizer campaigns to increase your exposure

The budget optimizer works by your giving control of your bidding to Google in an attempt to maximize the traffic you receive. The optimizer does not care which keywords get clicks, it just cares that your budget is spent and that it gets you the most clicks possible.

The benefits for a publisher who measure RPV (revenue per visitor) are obvious. As long as your RPV is higher than your CPC, you want as much traffic as possible on those keywords. However, for those who sell a product or service, there is a unique way of using the budget optimizer to maximize your company’s exposure.

If you set ROI or profit based bids by keyword then do not use budget optimizer.

If you consider words to have unique values, then do not put those words into a budget optimizer campaign.

However, if you have keywords that, are early in the buying cycle and do not directly lead to revenue, describe your industry and products well but are hard to monetize and you want exposure on because they are words in the news, branded terms or for other reasons then putting these keywords into a budget optimizer campaign can be a useful way to maximize your exposure on non-direct ROI keywords. With the budget optimizer campaign, set a “branding or exposure” budget. The goal of this campaign is to move searchers into your buying funnel so you can monetize them at a later date.

However, you still need to make sure that these keywords are helping you reach goals.

Setting goals for budget optimizer campaigns

If you set a goal for these keywords such as a newsletter subscription or contact, then you should be able to determine what those conversions are worth and then set keyword bids. Therefore, words that lead to those types of conversions should not necessarily be in a budget optimizer campaign.

A few metrics for measuring budget optimizer campaign keywords are:

• Page views per visit
• Time on site
• Video views
• Bounce rate
• Send to a friend
• Bookmark
• Print this page
• Tweet This (or other social actions)
• Download product specs

The above goals do not directly lead to revenue; they lead to exposure and show levels of consumer engagement.

It just so happens that the new analytics goals will let you measure these items much more easily. Follow these easy steps to be able to measure the effectiveness of these early buying cycle keywords:

1. Set goals for the budget optimizer keywords, such as 3 page views per visitor.
2. In Google Analytics, set goals for time on site, page views per visitor, or use one of the expanded goal sets to track the social sharing items.
3. Add early buying funnel keywords to a budget optimizer campaign.
4. Let the budget optimizer campaign send you traffic. It may take up to two weeks for budget optimizer to really get going; have a little patience here.
5. Examine your analytics to optimize your budget optimizer campaign
   • If a keyword is not meeting your goals – delete it
   • If a keyword is meeting other goals (such as newsletter subscriptions) then move it to your normal max CPC or conversion optimizer campaign
   • If a keyword is meeting your budget optimizer goals, but no other goals, then leave it where it is
6. The last step is to measure if your overall revenue is higher while the budget optimizer campaign is running. If yes, then as long as the increased revenue is higher than the money you are spending in the budget optimizer campaign, then you are in good shape. If not, then lower the budget optimizer”s budget.

Every keyword, ad copy and landing page should increase revenue. If it does not, it needs to go away. However, you cannot always measure direct revenue for every keyword or traffic source.

Therefore, you should set appropriate goals for a keyword regardless of where it falls in the buying funnel to make sure your entire budget is working toward turning a searcher into a customer.

Engage consumers early in the buying cycle, so when they are ready to buy your company has already been part of the conversation. However, do not just spend money to receive clicks. Make sure every single click is helping to increase your overall profits.

Each report in your AdWords report helps you to extract data from your account so that you can make meaningful decisions to optimize your campaigns. In part one of this 6/90 series, I looked at six reports which will help you make ninety percent of the decisions you need to make on a day-to-day basis to manage your AdWords account. In part two, I’m taking a deep dive into the keyword report, as it contains so much data that can help you make good decisions it deserves its own article.

Setting Max CPC

If you use the AdWords conversion tracker you can see your conversion metrics, such as cost per conversion, ROI, total conversions, and so on for every single keyword. This is the data that you will use to set max CPCs for every single keyword in your account.

My favorite column in the report is called value/click. This is the total revenue that a keyword brought to your company divided by the number of clicks that the keyword received. Essentially, it’s your break even CPC cost. If you bid based upon ROI, then the formula for setting a bid is:

Max CPC = (value/click) / desired ROI.

For instance, if your value per click is $2, and you wish a 200% ROI on your account, then the formula is:

$1 = ($2) / 2

If you have a hard cost of goods, then you need to take margins into account before you see your true value per click. If you are a profit based bidder, you will use different formulas for determining your bids; however, the starting data is the exact same as ROI bidding.

When you are making bid changes, you should do so based upon hitting your goals. The easiest way to see how all your keywords are performing is to use the keyword report combined with the AdWords conversion tracking script.

Understand your Quality Score

There are many times it is more profitable—and smarter—to work on raising your quality score over raising your bids. If your quality score is under 5, then you shouldn’t even be worrying about your bids. If your quality score is five to seven, then you should be working on both. If your quality score is seven or higher, then you will have to make most of your rank changes based upon changing your max CPCs; however, you should also be testing ad copy for both quality score and conversion purposes.

Quality Score is kept at the keyword level for search. If you have a large account, it can be difficult to find the best places to start working on Quality Score. This is where pivot tables can help. Run a keyword report and include a minimum of keyword, quality score, impressions, and spend. Next, put the data into a pivot table so that you can see the average of quality score and the sum of cost for each ad group:

Please note that this is an average of all the keywords within an ad group. If your keywords have very different impression rates in an ad group, rather than using the average of all the keyword quality scores, instead add a column to your keyword report that is call Real QS that is calculated by the formula (quality score * impressions). Then in the pivot table, divide impressions by Real QS. This will give you the average quality score for your ad groups, normalized by impressions.

Now, look for ad groups with low quality scores and high spends. In the above example, you would make much better use of your time working on the ad group that has spent $20,874 with a quality score of 5 than ad group 1 which has spend of just $25 or ad group 12 that has an average quality score of 7.12.

When you find ad groups that have high spends and low quality scores, examine the quality score factors to see which ones you can improve. If you can raise your quality score from 5 to 7 in ad group 11, you could save yourself a few thousand dollars.

Please note that this one of my experimental accounts. Yours should not look quite so consistent with so many ad groups having exactly a 5 quality score.

First page bids

The first page bid is an estimate of how much you need to bid for your keywords to appear on page one in all the geographies your ad appears. If you are advertising to a small region, then the first page bid is fairly accurate. The larger your geography, the less accurate the first page bid becomes. For instance, if your first page bid is $1 in Fargo, ND but $3 in Chicago; and you only bid $1.25, then your ad will be on page one in Fargo, but beyond page one in Chicago.

Unfortunately, you cannot see first page bid broken down by geography—Google provides just a single number per keyword. In the above example, Google would display $3 as the first page bid as that this the highest number across all geographies even though a lower bid would make your ad appear on page one in less competitive geographies.

While you should set bids from performance data, and not just raise bids so that your ads are on page one, there may be cases when your keywords never get enough page one exposure for you to know if those keywords would perform well on page one.

Use this report to look for places where your first page bid is below your max CPC. You can use conditional formatting in Excel to easily see what keywords are not on page one.

Deciding what keywords to pause, delete or move

If keywords are not performing, then you should pause, delete or move them to a budget optimizer campaign. If you set bids by ROI, these would be the keywords whose bid would have to be at or near $0.00 as they do not directly lead to conversions. Please note, just because a keyword does not directly lead to conversion does not mean its a bad keyword. You will have keywords that start someone down the buying funnel, but as they are not the last keyword clicked the words will appear to have a zero ROI.

If you see the message, “This account is nearly an unmanageable size,” it is an indication that your are reaching your account’s keyword limit. If you wish to keep adding keywords, you need to pause or delete some first. When you run a keyword report, there is an option under “filter your results” to include keywords with zero impressions. Run a report over a long time span and then look for keywords that have never received an impression. These are the keywords that you would want to delete in order to make room for new keywords. If you have a seasonal business you may wish to pause seasonal keywords instead of deleting them so you can unpause them when their season returns.

Each keyword you choose to use in your account should help you reach your marketing goals. It is essential to understand how each keyword performs, converts and helps you achieve your goals. When you want to know detailed information about any keyword, the keyword report should be the first place you look to see the appropriate metrics. The one exception to this rule is when you want to see the search query that triggered your keyword. For that information you should use the search query report.

You can have these reports automatically created and emailed to you as a reminder it’s time to examine keyword level data. You should regularly examine your quality scores, first page bids, and max CPCs to learn from your account, and make adjustments as necessary so that you can extract the most value from your AdWords account.

Google offers numerous AdWords reports that allow you to view more data than most people have time to analyze. One of the tricks to working with PPC is to determine what data you need to look at every day or week, and then determine what data is useful to analyze when improving your AdWords account.

There are six reports that, when used effectively, can give you the majority of insight you need to analyze your accounts on a day-to-day basis. In part 1 of this two-part series, I’ll take you through these reports to find insight into your accounts. In part two, I’ll take an in-depth look at the keyword report to create actionable items from the data.

While these reports are easy to run they are most effective when combined with the AdWords conversion tracking script. This is a different feature than what you find in Google Analytics. The conversion tracking script sends data from your website back to AdWords so that you can see conversion information in your AdWords reports. In walking through these reports, I’ll make the assumption that you are using this script. If you are not, then you can still use these reports; however, you should marry up the data with the analytics package you are using to track conversions to get a complete picture of what’s going on.

The keyword report

The AdWords keyword report should be your starting place for analyzing keyword data. This report shows metrics by individual keywords within your account. For instance, the main data points to examine when reviewing information by keyword are:

• Impressions
• Clicks
• CTR
• Conversions
• Cost per conversion
• Conversion rate
• Value per click

There are three main uses of this report:

1. Set bids based upon cost per conversion or other metrics with your established bid methodology
2. Find low quality score keywords with high spends so you can optimize for quality score (note: it can be useful to use a pivot table to find AdGroups with high spend and low quality scores. See Josh Dreller’s In The Trenches column about mastering pivot tables for more info).
3. Find keywords that are not on page one (where the first page bid is higher than your max CPC). Remember, though, that just because your keyword is not on page one does not mean you should raise your bid to be on page one. If you cannot be profitable on page one, then you need to decide if this is a word that can be on page two; if not, you need to optimize the landing page, ad text, or other factors for this keyword before you raise your bid to be on page one.

In part two of this article, I’ll take a deep dive into the keyword report to show various ways of working with the data it produces.

The search query report

The search query report will show you the query that was actually typed into a search engine that caused your ad to be displayed. For instance, if you bid on the broad match “coffee mugs,” your ad could show for “yellow coffee cups,” “blue coffee mug” or possibly even “tea cup.”

Use the search query report to find words that are not converting yet are consuming your ad-spend dollars. When you find such words add them as negative keywords.

Then use the report to find words that are converting and are not keywords in your account—consider these to be Google’s gift to you to help you improve the overall importance of your campaign. Add these words as keywords so that you can bid on these words based upon their actual returns. Remember, your broad match keywords will never convert higher than your exact match keywords.

When you conduct keyword research, you should always consult the search query report as part of your keyword expansion methodology.

The placement performance report

The placement performance report is only useful if you’re advertising on the content network. It shows you the sites and individual URLs where your content ads have been placed. The first time you run this report, include the actual URLs where your ad has been shown. Click on the URLs and look at the actual pages where your ads have been shown. If your ad is not being placed on the correct types of pages, then you should examine the keywords triggering your content ads.

If the ads are being placed on appropriate pages, run the same placement performance report again, but do not include individual URLs. By excluding the individual URLs, it is easier to gain high level perspective on how each web site where your ads are being displayed is performing for your business. Alternatively, you can also use a pivot table to examine the domain level information from a report that includes URLs. When you find a site that is bringing you high quality traffic that meets your business goals, add it as a placement targeted site so your ad is always shown there. When you find sites that are costing your money, but the traffic is not meeting your goals, then block the site from showing your ads in the future.

The entire work-flow process for managing content sites can be found in this Search Engine Land article.

The ad text report

The ad text report will show you all the ads that your are using along with associated metrics such as cost-per-conversion or clickthrough rate. Use this report when running any ad copy tests.

In addition, run this report on occasion to find ads that are performing below the averages for your account. Then go to those ad groups and write new ads to test which ones perform better. As you can have multiple ads running in any ad group, you should continuously be testing various ad copy. In addition, if one ad suddenly gets disapproved, you will have at least one more ad running so you will still maintain exposure in those ad groups while you fix the issue.

While most PPC marketers are always looking at text ads, do not forget your video or image ads. Use this same report to see the metrics for your other ad types. If you are running video ads, then you should be uploading multiple static images (with the same video) to see which opening image has a higher play rate, clickthrough rate, and cost-per-conversion.

The URL performance report is similar to the ad text report. It will show you where ad traffic is sending searchers on your website, and the associated cost-per-conversion and other metrics for that landing page. Analyze the URL report just like the ad text report. Find landing pages that are not meeting your goals, and test different landing pages and landing page layouts.

The geographic performance report

This is one of the more overlooked reports in AdWords. This report will give you information about where your ads are being shown and the associated metrics by geography. Even if you are a national advertiser, it can be beneficial to write ads for specific local areas.

One of the quality score metrics is clickthrough rate in a geographic region. If your ads do well in San Francisco, but poorly in New York City, then your ad will be shown more in San Francisco and less in New York City. As the New York City metro area makes up more than 10% of the entire United States population, a lack of exposure in that geography can significantly lower your total consumer base.

In the report, look for regions where you are not doing well and that have significant populations. It’s easy to overlay percentage of population data with the report to spot these regions. For instance, create a chart that shows:

• The regions where your ad is displayed
• The percentage of impressions and clicks you receive from that region
• The percentage of the population that lives within that region

Always take other factors into account when examining population data in isolation. For instance, say you have a physical location in Miami and your ads are not doing well in either city. It’s likely more worthwhile to work on getting your ads to perform better in your local market, Miami, before spending time tweaking your Los Angeles campaign, even through Los Angeles has a much higher population.

When you find underperforming regions, and those regions have a high enough population that you are willing to create a campaign just for that region, follow these steps:

• In your original campaign, exclude your ads from being shown in the new geographic area that you want to target
• Duplicate the existing campaign with the AdWords editor, and then change the geographic area to be the new region
• Write ads that are compelling for that region

If the region is not large enough for you to create a specific campaign for a region, conduct keyword research to see if consumers are using location specific words (such as Los Angeles plumber) in their search queries. If they are you can create an ad group that uses geographic keywords with associated geographic ad copy.

The impression share “report”

Technically, there is not an impression share report. The impression share report is an unofficial name that advertisers have given this report. To see impression share data, you need to run a campaign report, and in the reporting options select these options:

• Impression share
• Lost IS (rank)
• Lost IS (budget)
• Exact Match (IS)

Once you have the report, you will see how many impressions you are losing and why you are losing impressions.

If you are losing impression due to budget, then you can easily determine how many more impressions you would receive if your increased your budget. This is very useful for determining what your account could spend based on the current keywords in your account. If you are losing impressions due to rank, then you need to improve either the bids or the quality score to receive higher placement so these ads are shown.

Setting an AdWords reporting schedule

There is more information available in AdWords reports than you could ever hope to analyze and take action upon within any day. To cope with information overload, first define your business goals. Next, define what metrics will give you insight into attaining those business goals. Once you have defined those two items, then run the AdWords reports which will help you gain insight into those goals so that you can find areas where you need to improve your account.

Then create a reporting schedule based upon when you want to see and act upon the data. You can set AdWords report to be automatically created and emailed to you on a daily, weekly (Monday), or monthly (first day of every month) schedule. If those options do not meet your needs use your calendar to remind you to create reports.

Here’s a recommended schedule that’s easy to work with—though remember, based on your goals, budget and time, your schedule might differ tremendously.

Keyword report:

• Daily – examine changes in cost per conversion, conversion rate, and total conversions.
• Every Monday – examine changes in quality score, find areas of low quality score, work on improving your quality score.
• Every Monday (or daily for large account, highly competitive industries, or volatile bidding environments) – examine keywords not on page one.

Search query report: Every other Tuesday. Conduct additional keyword and negative keyword research.

Placement performance report:

• Every Wednesday. Examine changes in cost per conversion, conversion rate, and total conversions for bid changes purposes.
• Every other Wednesday: Examine sites you wish to block or add as placements.

Ad text report /URL report:

• Every Thursday. Finalize tests or create new text ads.
• Every Friday. Finalize landing page tests, create new landing pages, create new landing page tests.

Geographic performance report: Once a month (assuming you are not a regional specific business, then weekly might be more appropriate).

Impression share: Once a month (weekly or daily during high peak seasons) to help create next month’s strategy of how to improve your account.

Always conduct in-depth audits beyond these reports and use the other reports and tools as necessary. Google offers a tremendous amount of data that you can extract from your account. While these six reports will not give you every detail you will ever need to optimize your account, these six reports will give you 90% of the data necessary to improve your account’s performance.

The fundamentals of a good pay-per-click campaign have not changed over the past six years. However, every single day, advertisers new to the world of PPC advertising are making the same mistakes as veterans made years ago.

You can save yourself lots of time and money by examining the mistakes of others so that you do not fall into the same traps of creating unprofitable PPC campaigns. Here are seven of the most common mistakes I see new advertisers make—and more importantly, how to avoid them.

Not setting & measuring campaign goals

Before you start any advertising campaign you should set the goals for what you wish that campaign to achieve. If you do not have goals, there is no way to know which keywords, ad copy or landing pages are meeting your goals or just wasting money. A conversion goal is a metric your define for your account. It could be a form fill, a phone call or an eCommerce checkout. First, set your account goals. Second, find a way to measure those goals.

All of the major PPC engines have conversion tracking scripts that you can put on your website. When these scripts are triggered, it sends conversion information back to your reports so that you can see what keywords are converting and which are wasting your money.

These scripts are useful to use with Google Analytics or other analytics systems as they put data back into your PPC accounts directly. This makes it easier to see conversion data in all the reports you run within your PPC account.

Poor account organization

The most time consuming aspect of PPC accounts is the organization of keywords and ad copy within an AdGroup. The most basic, yet fundamental notion of an AdGroup is that the keywords in the AdGroup only trigger the ads within that same AdGroup. The ad copy within an AdGroup only show when a keyword within that same AdGroup is matched to the search query. Yet many beginners either stuff their AdGroups with too many, or unrelated keywords.

While your ad copy does not have to contain the actual keywords, the closer your ad copy reflects the intent of the keyword search, the higher your clickthrough rate and conversion rates will be.

Use this exercise to organize your AdGroups:

1. Write an incredibly specific ad copy with exactly one search intent
2. Examine the keywords in the AdGroup
   • If the ad copy reflects the keyword, leave the keyword in the AdGroup
   • If the ad copy does not reflect the keyword, put the keyword in another AdGroup

It is better to have ten AdGroups of ten keywords than one AdGroup of 100 unrelated keywords. There isn’t a magic number for how many keywords should be in an AdGroup—at least if you’re targeting the search network. Google says that for ads appearing on the content network, only the first 50 words in an AdGroup are used to determine its theme.

No doubt, organizing your account in this granular fashion is time consuming. However, it will make your ads and keywords more relevant, which will lead to higher quality scores, higher clickthrough rates, and ultimately, a greater payoff for your efforts.

Focusing on traffic rather than conversions

Every day someone adds a very general keyword such as “TV” to an account. What does that keyword even mean? Is the searcher looking for TV listings, Plasma TVs, online TV or something else? The word does not have a defined search intent. While these general keywords can drive a lot of traffic to your website, the purpose of search advertising should not always to drive more traffic for the sake of getting more eyeballs to your website.

Start by focusing on more specific keywords that have a defined search intent, such as:

• Samsung DLP TV
• Buy Samsung TVs
• Samsung hl66sw

While general keywords will drive a lot of traffic, they can also waste a lot of money as they usually have low conversion rates and do not pay for themselves. Using general keywords is OK once you understand how they affect the metrics in your account. However, when you are first starting PPC advertising, use specific keywords until you understand how different types of keywords convert.

Ignoring match types

By default, Google uses “broad match” to trigger results for a search query. Broad match by definition is a loose fit, so a general keyword such as “coffee mugs” can match queries such as “tea cups,” “Starbucks coffee mugs” or “collector coffee mugs.” Those searches are conducted by individuals looking for very different items. This happens because a broad matched word can be matched to similar, misspelled or plural words. In this case, tea is similar to coffee and cups is similar to mugs—that’s why it is possible for your “coffee mug” keyword to be displayed when someone searches for “tea cups.”

There are three match types in Google AdWords and Microsoft adCenter that determine what variations of the search query will match to your chosen keywords and cause your ad to be displayed.

Broad match has its place, but if you exclusively rely on it you will lose control of ad serving. Broad match keywords will never convert higher than your exact match keywords.

In addition, you can use negative keywords in your account to stop your ad from showing. In the above example, if you sold coffee mugs, but you do not sell Starbucks coffee mugs, you could use the negative keyword Starbucks so your ad would not show when someone searched with the word Starbucks.

Negative keywords allows you to control when you do not want your ad to show. This in turn leads to higher CTR, conversion rates and more profit by not paying for clicks that will not convert for your company.

Failing to separate content and search campaigns

There are two different types of advertising that you conduct within the major PPC accounts. The first is search. With search, your ad is only shown when someone actively searches for your keywords.

Your ads can also be displayed across the content network. With the content network, the search engine examines all the keywords within your AdGroup to determine the theme of your AdGroup. Then, Google places your ad on a web page with content related to the theme of your AdGroup (such as NYTimes.com).

Because there are very different ways that the keywords in your AdGroups trigger search or content ads to be shown, and very different user intent (active searching vs. passively reading an article), you should separate your content and search campaigns. You should also use different keywords so that your ad is triggered appropriately and you can optimize the content network.

Google’s matching of ads on the content network has vastly improved in recent years, but it can be difficult to use effectively when first starting your PPC campaigns. I recommend start with only search, and once you have fine-tuned your keywords, ad copy and landing pages, then turn to the content network to further the reach of your accounts.

Sending all traffic to the homepage

Searchers, by definition, are looking for a specific piece of information. Usually a company’s homepage has lots of information about a company, or speaks broadly to their products, but does not speak to any particular search query. If you send searchers to your homepage from your ads, and they have to hunt for the information they’re looking for, you will see people quickly abandoning your site. You should send traffic to:

• A landing page created for a specific search query
• The page that is the furthest, logical step in the buying cycle for your website

Examine the conversion rates by keywords and landing page. Look at bounce rates for keywords and landing pages. When you see high bounce rates and low conversion rates, then the traffic needs to go to another page or you need to optimize the page to increase conversions.

Not testing

The PPC engines allow you to have multiple ad copies in a single AdGroup. Each ad copy can also go to a different landing page. It is very simple to test different ads and landing pages to see which ad copy or landing pages increase your primary search metrics, such a clickthrough rate and cost per conversion.

It is not difficult to get to metrics such as profit per click to see which ad copy and landing page combination leads to the highest most profit for your different keywords and ad copies.

When creating a new AdGroup, always create at least two ad copies. After you have enough data to reach statistical significance, then you can run reports to see which ads, landing pages and keywords lead consumers to do business with your company.

Conclusion

These deadly mistakes should be taken to heart by all new advertisers. However, remember that the more you learn about PPC marketing, the more you can deviate from these basic rules as there will be times when you need to break them to reach particular objectives.

However, learn from the mistakes that others have been making so that you do not fall into the same trap which has caused many PPC accounts to become unprofitable, or worse, fail entirely.

The web has given us the ability to track amazing amounts of data. However, that data is only easy to track when the actions take place online. When you conduct business via the phone, in-store, or other ways that cause the user to go offline, the reliability of data begins to suffer.

There are several companies that offer different types of call tracking solutions. Some of those solutions are quite expensive, and others lead to data loss. Understanding the different ways of tracking phone calls is important when either selecting a vendor or instituting tracking on your own.

There are five main ways in which you can track phone calls. Of course, there are other methods and variations on these methods.

Unique phone numbers

You can have a unique number for each keyword, ad group, campaign, or account.

With this method, the first thing to determine is if you want to use a static number or a dynamic number. With a static number on the page, you will hard code the number into that page. If you hard code the number, then you will want to make sure that there are not additional ways of navigating to that page. If someone can enter your site from the main page and navigate to that page, then you will not have accurate information about which traffic source led to a phone call.

With dynamic numbers you will first have to determine a way to know what account, campaign, ad group or keyword initiated the traffic. The most common way of isolating these variables is to add parameters to your links that your system will read and display the corresponding phone number.

To further improve the tracking, you can set a cookie that corresponds to the traffic source. Then, regardless of what page someone navigates to on your website, you can display the phone number that is associated with that cookie. This is also useful if someone leaves your site and returns later. Since the cookie will still be set, you can still serve the call tracking number associated with the click to determine the quality of those clicks.

Pros: If you use dynamic parameters and cookies, this is one of the most accurate call tracking solutions that is possible. You will have very little data loss. However, you will have to decide additional business rules (for example, if someone clicks on a second ad, do you set a new cookie with a new phone number or keep the first cookie and the original phone number?). These types of rules will affect the type of optimization you can achieve.

Cons: The biggest disadvantage to this method is that it can become very costly. If you need thousands of numbers you could end up with some hefty phone bills. This method also requires some coding. If you are using a third party vendor, it might be as simple as dropping a script where the phone number appears. If you are coding this in-house, then it can take a bit of development work to coordinate the phone numbers with the ad copy, keywords, ad group or inventory source.

Phone extensions

This method is identical to the method above, except instead of using a unique phone number, you will use a unique phone extension. The same rules apply with regards to static versus dynamic phone numbers and additional business rules.

Pros: This method can be much less expensive than using unique phone numbers. However, it can be just as accurate since you are using a unique extension for each ad group, keyword, or campaign.

Cons: The main disadvantage of this method is that not all phone systems support endless extensions, and adding new ones can be difficult or require the intervention of a different department. If your system does not support endless extensions, it might be less expensive to use another method that institute a new phone system. In addition, some consumers could be turned off by calling a number and then having to enter an extension. You should examine how many people call your number but do not enter an extension.

Code identifiers

With this method you will generate a unique code that sits just below the phone number. When someone calls your office, the person answering should start with the dialogue, “To best direct your call, can you please read me the code below the phone number?” That person will then record that number (or punch it into the phone system) and then continue the call as normal. This method also relies on either generating a unique code (instead of phone number) listed in the unique phone number above by examining the link parameters or by hard coding the code into the page (which leads to some data loss).

Pros: This method requires a little bit of developer time, which is the same as any of the above methods, but the cost should be lower than buying a large amount of numbers that you pay for each month.

Cons: The biggest challenge of using this system is training those who answer the phone to ask and accurately record the codes.

Search engine conversion tracking script

All of the major engines have a conversion tracking system that allows you to see conversions within your account. However, those actions rely on someone viewing a page of your website that contains the code. If someone is calling you, then they are not viewing pages of your website. However, if you can be creative and get the caller back to a page on your website, then you can trigger the script to record the action. The easiest way to do this is to generate online receipts. When someone views their receipt, then the script is triggered and you can potentially tell which keyword, ad copy and landing page generated the conversion. However, you can also use methods such as receive a coupon, take a survey, etc.

Pros: The biggest advantage of using the engine’s conversion codes is you will be able to see phone calls in your PPC reports. The second advantage is that it requires very little additional coding (beyond adding a simple conversion script to the pages which the engines will walk you through when you sign up to receive the code). This is one of the least expensive solutions.

Cons: The biggest disadvantage is data loss. There will be people who will not revisit your website that triggers the code, or will do so from a different computer. Although this could be seen as a disadvantage, the silver lining is that you need to be creative to get someone to visit specific pages of your website. However, by sending someone back to your website, you do have an other opportunity to continue to engage the caller.

Confidence factors

If you have 10,000 keywords, yet only one phone number (or even 100), how do you get keyword level call data? You use confidence factors to determine which keyword is generating the call. For instance, if you time stamp every click, and then track the time that each phone call came through to your business, you can start to see patterns that if this keyword is clicked, there is a percentage chance that our phone will ring in the next ten minutes.

This is a really useful solution for aggregators. Consider a company that does marketing for 1000 plumbers. While each plumber might have a unique phone number, it hurts margins to use multiple numbers for each plumber. While each plumber might only receive 20 calls a month, with 1000 plumbers, you could track 20,000 phone calls per month. In this case, that is enough data to begin learning which keywords, ad copy, keyword, and engine are generating the phone calls. You could even learn that keyword one generates phone calls in Seattle, but keyword two is more likely to generate phone calls in Miami.

If you generate 100 clicks and one phone call in ten minutes, then any one of those keywords could generate the phone call. If in the next ten minutes, you have another hundred clicks, but only two of them are from the first set and you have another phone call, then there’s a higher chance that one of those two keywords generated a phone call, but of course, this is not enough data to make a decision. This method does require a significant amount of phone calls to be able to confidently understand what keywords are leading to calls.

Pros: One of the advantages of this method is that it enables aggregators to affordably understand which keywords are leading to phone calls. Another is that you do not need a significant number of call tracking numbers, which makes this method more affordable than other solutions.

Cons: This method requires a lot of phone calls to gain confidence that a particular keyword is generating phone calls. If your PPC advertising does not generate a significant amount of phone calls, this is not the method for you. In addition, as you are creating confidence numbers, but without a one keyword to one phone call ratio, your data will not be completely accurate. The goal is to understand that when a particular keyword is clicked, you receive a phone call a certain percentage of the time so you can set bids based upon returns.

Note also that this method does not work well with long tail keywords. Long tail words might not have enough traffic for you to be confident in their ability to generate a phone call. In these cases, you might track at the ad group level instead of the keyword level.

Phone number pools

The last method is to keep a pool of phone numbers. This pool is smaller than the total number of items you wish to track. When a click occurs, you dynamically assign that keyword or ad group a phone number. That phone number will be associated to that keyword or ad group for a certain amount of minutes. After a while, that number goes back into the pool to be used by another keyword when necessary. All of the numbers in the same pool should be assigned to the same advertiser so even if a number expires in the tracking system, the number still rings to the correct advertiser. If you are an individual advertiser and like the idea of confidence stamping above, then this is a solution you should investigate.

Pros: Phone number pools work very similarly to the confidence factors (above) at the keyword level, but do not require nearly as much traffic. The more phone numbers you have, the less total traffic you need. This allows you to receive better data on long tail keywords, and have a higher confidence on low to medium trafficked accounts.

Cons: Whenever you use confidence factors, you will have false positives. If you have a lot of traffic and keywords, then you need to buy quite a few phone numbers. While this is not nearly as expensive as the unique phone call method, it can become pricey. Ask the provider what their error rate is in determining what keyword generated the phone call. Then ask their competitors what their error rate is. Hearing a 30% error rate is not uncommon. However, the larger the pool of numbers, the lower the error rate should become over time.

Helpful hints when buying call tracking numbers

Some companies will use one of the more expensive techniques, such as unique phone numbers or phone number pools, for a few months to gain enough data to make reliable bid decisions. Then they will turn off the call tracking for a while to save on expenses. After a while, they might then turn on call tracking again to see what has changed with their phone call leads.

When buying one or more numbers from a call tracking company, one of the most important questions to ask is how long a number been “cleansed” or not used by another party. Do a search for your number and see if that number appears on many web pages. A number that keeps receiving old calls is often called a “dirty number” and can ruin your tracking. Having a number out of commission for 90 days between users is fairly standard. If you receive a significant amount of bad calls from your call tracking number, do not hesitate to inform the call tracking company and request new numbers.

The other conversation to have is how additional minutes are pooled. When you buy a number, you usually pay a flat rate for that number per month and that number comes with X number of minutes each month. If you go over that limit, then you pay a per-minute overage charge. Some companies will keep those limits at the phone number level, meaning you might have one number go over and another under the limit but you still need to pay an overage charge.

A way to control your minutes is to create a pool. Instead of a single number having 30 minutes of talk time, having 20 numbers and 600 minutes of talk time, regardless of the number called, can help lower total costs.

Some other features that are common to see are call recording, API (if you want to hook up the call tracking to your system), call length, number of rings (useful for seeing how fast the phone is being answered), caller phone number (useful for finding patterns, such as telemarketers and blocking those numbers), and call block. Most features that are available on land lines are available with call tracking providers.

It is helpful to first determine a budget and a required feature list before evaluating vendors. As many vendors methods vary, having an idea of your preferred method, budget and features can help you evaluate who is a best fit for your needs.

Conclusion

Call tracking does not have to be an expensive or difficult proposition. At the least, you could buy a call tracking number (or use a Skype, Vonage, or Google Voice number) and use a different number per inventory source (one of adCenter, Yahoo, and AdWords) to understand if your PPC campaigns are generating phone calls. While this method will not give you enough data to set different bids by ad group or keyword, it will give you a high level picture of whether you are even receiving phone calls.

The more you want to track, the more expensive the tracking proposition becomes. If you are buying a large amount of numbers, do not be afraid to negotiate the call tracking costs.

Do not optimize your account for just total calls. You should enter the call information into your system to not just evaluate total calls, but also sales. If you receive a hundred calls a month and no sales, were those really good keywords or phone calls?

Once you have instituted call tracking, then you can use that information to set budgets and bids based upon returns. You can split test ad copy, ad groups, and different video or image ads to determine what is really working for your company. You can even use different numbers for your newspaper, yellow pages, billboards and other media buys.

If you conduct sales over the phone, you should use some level of call tracking to understand the returns you are receiving for your marketing spend. Do not bid blindly. Rather, bid based upon the profits and marketing goals your campaigns achieve.

The number of people who access the web from a mobile device is on the rise. However, those accessing the web from a mobile device with a full browser (such as an iPhone or Blackberry Storm) and those on a mobile browser (such as Blackberry Curve) see different sets of information, and, more importantly to marketers — see different advertisements on a Google search result.

Showcasing ads to mobile audiences is excellent for companies who are searched for while a user is out in the community without a computer, such as:

• Department stores (show driving directions)
• Towing companies
• Locksmiths

Or to reach users who are looking for mobile software:

• Mobile applications (such as iPhone apps)
• Ringtones

There are other company types that will find mobile advertising useful. As most users on a phone are not going to visit many pages on your website, think about what a user would want to accomplish while on a small device and direct your traffic to such a page.

After we discuss the options for reaching an audience, we will showcase an easy way to test mobile campaigns if your website is not mobile compatible.

AdWords has ways of reaching audiences regardless of their phone type. However, there are different techniques for putting your ads in front of a consumer based upon their mobile browser type.

Reaching users with full Internet browsers

To reach a user with a full Internet browser, the steps are quite simple. In the settings tab for your campaign, you have options to display your ad based upon the user accessing the Internet via:

• Desktop or laptop computer
• iPhones or other devices with full Internet browsers

(Click image to view full size at Flickr)

You can choose one or both options.

If you choose to display your ads to users with a full Internet browser, check to see how your ad renders in that browser. Some full mobile browsers do not render flash or java correctly. If your images, flash files, or script files are large, it might take quite some time for your website to render on this device.

If you use Google Analytics, you can segment your current audience metrics for those users coming from an iPhone.

While this will not show you everyone coming to your site from a full HTML browser, it can give you directional information about how iPhone users are interacting with your website. Examine your bounce rate. If the keywords are relevant to your website, and your bounce rate is high, it may be an indication of a slow loading site on a browser.

Reaching users with mobile browsers

To reach users that are not using iPhones, or other full HTML browsers, there is a different technique. Google has another ad type called ‘mobile ads’. Navigate to the ad tab within your ad group and click ‘New Ad’, followed by ‘Mobile Ad’.

Mobile ads are much smaller than text ads. They contain only two lines of eighteen characters. As the display screen on phones is smaller than a typical browser, so are the text ads.

The real decision about this ad types comes from deciding what options to give the user. You can give the user three options upon seeing your ad:

• Click the ad to visit your website
• Click the phone number to initiate a phone call
• Give the user both of the above options

You must have a mobile compatible site to send users to your website. If you do not have a mobile compliant site, your only option is to use the click-to-call feature of the ad.

In addition, you can choose which network you wish to display your ads on. For instance, if you have created a app that only works on T-Mobile phones, you would only want to display your ad to T-Mobile users. For most advertisers, feel free to show your ad to all carriers.

Mobile ads are only shown if the user meets your other campaign’s restrictions, such as geography or time of day. Therefore, you can create click to call ads for a locksmith that are only shown in the city that you service.

Creating mobile pages

Google use to offer mobile landing pages that you could create within your AdWords account. Those were retired a few months ago due to low usage, and they will not be missed as there were not formatting options.

If you want to test mobile advertising, but do not want to pay for the development for a full mobile site until you are certain you wish to utilize mobile advertising, there is a very easy way to create a test site using WordPress. WordPress is an easy-to-install blogging platform. However, you do not have to use it as a blog. You can install the CMS and create a couple landing pages to test out for your mobile ads instead of starting a full blog. In addition, WordPress supports many plug-ins. One of these plug-ins makes WordPress mobile compatible. Therefore all you need to do to test out mobile advertising is:

• Install WordPress
• Install the mobile compatible plug-in
• Write a few pages for mobile users
• Send traffic to those pages via AdWords

Here is a full instruction guide to creating WordPress mobile pages.

Previewing mobile search results & websites

Google has a mobile preview tool that will let you view mobile results from within your desktop browser. Navigate to the http://www.google.com/m/adpreview and conduct a search. You see results in several countries, and even by carrier in the US.

You can also do a search at Google mobile to see the organic results, but Google mobile rarely shows ads even for results that should display ads when you are on a desktop.

Google also has a mobile proxy tool that will adopt any page to be viewed on a mobile device. It is useful if you are on a phone and the page uses scripts that your device cannot render, which is its primary purpose. However, it can also be used to get an idea of what your site might look like on some mobile devices.

That tool can become tedious to use over time. An easier way to see what your site will look like in a mobile browser is to use the FireFox plug-in wmlbrowser that will simulate WAP browsing from your desktop.

Conversion tracking

The AdWords conversion tracking script supports both mobile and desktop browser usage. You may have different conversions based upon a user coming from a mobile device compared to a desktop device. Therefore, set your goals appropriately.

It can also be useful to have completely separate mobile versus desktop campaigns. As you might send traffic to different pages on your site for the same keywords based upon their device, or have different conversion options, use the AdWords editor to quickly duplicate your campaigns and then set one for desktops and one for mobile devices.

Conclusion

Mobile usage is climbing, although it has definitely not hit its peak. There will still be many growing pains for mobile advertising and mobile usage. In addition, the volume of clicks for most businesses is still low from mobile devices. However, conversion rates for phone calls and driving directions are often higher than desktop browsing for many industries. If you are locked out of your car, you do not comparison shop locksmiths. Most users find one locksmith and then place a phone call.

Testing mobile advertising is not a difficult proposition. It will take you only a few minutes to create mobile ads or change your campaign settings. If you are technically minded, it will take only an hour to install a new WordPress site and write the content for your new mobile landing pages.

As with most new advertising techniques, the hard part is choosing to try out the new medium and setting aside the time to actually accomplish the changes. Take an hour one day and test out mobile. The results might surprise you. Even if it doesn’t work well due to the low volume of clicks, leaving the campaign running will let you see the stats for when mobile adoption is starting to hit your industry so you can be in front of the curve in mobile adoption.

Google has recently made two big changes regarding trademark usage within AdWords. One is a shift in the number of countries where trademarked words are allowed as keywords. The other is how trademarked words are allowed in ad copy within the US.

Google’s trademark policy has been rehashed in several large media outlets. Unfortunately, most of them got it wrong. So to clarify, here is a rough overview of the current AdWords trademark policy for the US, UK, Ireland, and Canada:

• Any trademarked term can be used as a keyword
• Any trademarked term can be used in ad copy
• If the trademark holder asks Google to not allow others to use a trademarked term, then it will be disapproved from ad copy, but can still be used as a keyword
• A trademark holder can allow only certain companies to use their trademark in ads (but this is all or nothing, with no conditional usage)

Country expansion

Google’s policy is the offspring of many lawsuits, the most talked about one being the 2004 decision of Geico vs Google. Google was only on strong legal footing in North America, and had a completely different policy for the rest of the world. Last year, Google extended its policy to include UK and Ireland.

Earlier this year, Google extended the US policy to another 190 countries. There are some European countries not yet included. A full list of countries where you can use any trademark as a keyword can be found here.

If you offer goods or services with trademarked words in any of these countries, you might want to start adding those services and products as keywords to find new search inventory.

US trademark changes

Google recently announced changes to its US trademark policy. Those changes can be summed up quite easily:

• If a trademarked term is generic or descriptive and does not reference the holder, or the term is not trademarked for your industry, you can use a trademarked term in the ad copy (for instance, an orchard farmer can advertise that they sell apples)
• If you sell a trademarked product, resell the product or sell replacement parts for the product, you can use the trademarked term in your ad copy
• If you are an informational site that is non-biased, and does not sell or facilitate the sale or a product, you can use a trademark when sending traffic to a review or comparison of the product

Who does this change benefit the most?

If you sell trademarked products or offer trademarked services, this is a fantastic change for you. In the past, if you sold iPods, you were never able to advertise them by name. That’s changed now. Ecommerce sites should start re-writing ad copy today to take advantage of this change.

This change will also make life easier for franchises. At present, most franchise trademarks are held by the corporate office. If an individual franchise tries to use their own business names in ads, those are are usually rejected as the account does not have permission to use their own trademarked business name in the ad copy. Under this new rule, they will be allowed to use their business name without having to get the corporate office to fax paperwork to Google.

At points in time, keywords such as Botox or Invisalign have been difficult products to offer as you couldn’t use them in ad copy. It was very difficult for consumers to determine which dentists offered Invisalign without clicking on every ad to see if it was in the dentists included products list. For keywords such as these, Google would argue that the consumer benefits. However, those providing these services benefit even more as now they can call out those exact services in ad copy.

The way that Google interprets their own policy will determine what informational sites actually benefit from this change. This looks like a perfect change for a site like CNET.com. However, CNET does track all their offers, and may even get paid when they send converting traffic to another site (I don’t know). However, many merchants track the quality of traffic sent to them, and if none of the traffic converts they stop buying ads on those sites. On paper, this change looks good for review sites, newspapers who offer reviews, and so forth. However, there is still interpretation left to decide how well this will work in practice.

How will this change affect affiliates?

There are two types of affiliates who buy PPC:

• Those sending traffic to the merchant site
• Those sending traffic to their own site

Affiliates who direct link to a merchant’s site can gain tremendously by this change. Those affiliates can now go and buy a significant amount of trademarked keywords as their accounts should be treated just like the merchant buying traffic directly to their site.

Affiliates who send traffic to their own sites, however, may find the new rule change makes things more difficult. For example, say you are an affiliate who offers free insurance quotes and has a consumer fill out forms to receive the quote on your website. However, before the actual sale of the policy, you pass the consumer to the insurance company’s website. In this case, can you use trademarked insurance names in ad copy? Or more importantly, will the Google reps understand you business well enough to know if you should or should not be able to use these trademarked words? It’s not clear.

Many affiliate sites look just like a merchant site, and it is not until the actual purchase is about to occur that the searcher goes to the actual merchant. This is another one of those very ambiguous cases that could be decided either way.

Pre-sell affiliates, those with one page wonder sites where the only purpose of the page is to go to the merchant or buy a ClickBank product should not see any benefit from this change as they do not actually sell the product. However, if these sites get classified as informational, then they might be able to use trademarked keywords in ad copy.

Affiliates who use comparison pages could be the biggest winners. The question will be how Google interprets the phrase “facilitate the sale of a service or product.” If facilitate the sale only pertains to selling the product on a site you own, then comparison affiliates will be the largest affiliate winners. However, if Google determines that affiliates are offering biased reviews to get people to buy their products, then these affiliates will not make any gains. If some of these sites are disapproved for being biased, and others approved for being non-biased, this could into a customer service nightmare.

How suppliers and merchants might change agreements

Many merchants currently put restrictions into their affiliate agreements or terms of service that affiliates cannot use certain keywords. Amazon recently announced that they would no longer pay affiliates who send PPC traffic to Amazon’s site. Expect more merchants to add excluded keywords to their affiliate deals.

Another ripple effect will be from suppliers making changes to contracts that would disallow merchants from using certain words, or their product names in PPC ads or keywords. While Google won’t police these activities, do not be surprised if a few suppliers that also sell direct to consumers online add new clauses to their contracts.

Trademarked business names

If your business name is not a product name, then your competitors have no way of placing your actual business name into the ad copy yet.

When business names are placed into ad copy, that could be considered the start of customer confusion. This can happen when a customer explicitly looks for one business, but is fooled into viewing another business due to how ad copy is written.

While your competitors do not have an easy way to put your trademarked business name into ad copy, business review sites, such as Yelp, can do this. If business comparison and informational sites start writing ads that contain business names, a whole new wave of lawsuits will follow.

If these sites are sending traffic directly to a review page for just that site, then you might not pay much attention. If they are sending traffic to a review page that also contains other information, such as related companies, then you might consider buying ads on those sites, or seeing if there are ways to put your business information there so you can be shown alongside your competitor’s information.

Will trademarks be better enforced?

Ad copy is suppose to be spelled properly. Trademarks are supposed to be kept out of ad copy. Therefore, trademark misspellings should not occur. However, iPod has been one of the most creatively worked around words I’ve ever seen.

Here are common spellings for iPod in search ads:

• lpod (l instead of i)
• iP0d (zero instead of 0)
• iPodVideo (running words together to form new words)
• Pod
• lP0D (an l and zero, and all upper case when shouldn’t be spelled that way).

At present, while Google does a decent job of automatically blocking the exact use of a trademark, creative uses of words need to be policed by the trademark holder. With the loosening of restrictions, hopefully the allowed trademarks will be better enforced and not lead to a new wave of poor ad experiences.

What’s the end result?

Ecommerce sites selling products with trademarked names and businesses offering trademarked services are by far and away the clear winners with these changes.

Informational sites, once it is determined what is a non-biased, non-commercial, informational site will be the second largest winner.

If the above were the only companies affected, then I would argue that consumers are getting a better experience in searching for information and products on the web.

However, these new policies changes will affect more than those few groups of companies. It’s how these other trademarks are enforced, or allowed to be used in ad copy, that will ultimately decide if these changes enhance the consumer experience or just lead to more lawsuits and more confusion among searchers.

Disclaimer

Companies can still sue advertisers regardless of whether they follow Google’s policies. Suppliers and merchants can put restrictions into their contracts to restrict keyword buying. Just because you are following Google’s policies does not mean that you cannot be sued by another advertiser. As always, first check with your lawyer before going down the path of using trademarked terms in ad copy or as keywords.

Search page views make up fewer than 5% of all page views on the web. By default, that means that content network impressions are the other 95%+ of all page view on the web. So why are so many people still shunning the content networks in favor of the comparatively tiny exposure offered by search result pages?

If you are looking to…

• Increase your total conversions
• Increase your company’s reach
• Expand the number of people who enter the buying funnel

… then the natural extension to search campaigns is a good content campaign. However, content behaves much differently than search. There are a few simple rules to follow to create effective content campaigns.

The most talked about organizational practice is to separate your search and content activities into separate campaigns. However, that rule is restrictive when you find particular sites across the content network that are sending your website excellent traffic where you would like to receive more traffic from that exact same site. There is a better way to organize your content campaigns that lets you both find new sites to display your ad, and also ensure your ad is on the websites that have converted for you.

Proper account organization

The first step in content optimization is to create a separate campaign for search vs content. There are a few reasons why you want to separate these two types of campaigns:

• Set different budgets
• Use different keywords to trigger your ads
• Use different text ads
• Try out rich media ads (video, image, interactive ads)

Creating a separate content campaign can be accomplished in just a few keystrokes with the AdWords Editor. Copy and paste your search campaign. In the original campaign, change the campaign settings to Google (and possibly search partners) only. In the new campaign, change the settings to just content reach.

Personally, I like to create a third campaign where you are only placing your ads on sites that have converted. The methodology will become more apparent how to use this placement campaign later in this article.

Use different keywords to trigger your content ads

With search, if your keyword matches the search query, your ad could be shown.

This is not true on the content network, where limited keyword matching occurs. On the content network, Google looks at a maximum of 50 keywords in your ad group and attempts to determine the theme of your ad group. Because you are just attempting to use words that create a theme, then you may use very general words which do not convert for search, but help determine a theme for AdWords.

I like to use the game Taboo as an example. In the game Taboo, a player is given a word on a card. They need to use words that describe what is on their card in attempts to get other players to guess their word. Advertising on the content network is quite similar. In essence, you are giving Google a list of words so they can “guess” what your ad group is about.

Consider this keyword list:

• mobile phone
• cell phone
• bluetooth
• wi-fi
• app store
• apple
• at&t
• 3G
• iPhone

How long did it take you to realize this list was leading to the final word—iPhone?

Words like “wi-fi” or “bluetooth” are so ambiguous, you wouldn’t want to use them in a search campaign, at least on their own. However, those are two words that are likely to be found on a page that discusses the iPhone, so they actually have great value when it comes to targeting the content network.

In content, there are generally 5-20 words that together describe your product and are likely to be found on any page that discusses your products. Therefore, it’s very useful to use totally different words to trigger your content ads. It’s a completely different mindset than you use when crafting a search ad campaign.

Use placement reports to block undesirable sites

The placement performance report will give you valuable information about the sites and special categories where you ad was shown. This will quickly let you see what sites are not converting for your ad groups. Once you’ve determined what sites or categories are not performing, block them from displaying your ad in the future.

This is easy. In your tools menu, navigate to “Site and Category Exclusion,” choose a campaign, and then list the sites, topics, etc. where you don’t want your ad to show. Here is a step-by-step guide that explains how you would accomplish this for parked domains. The exact same steps apply for different categories or websites.

Once you know your winners, create a third “targeted placement” campaign

There are millions of web sites, and it’s impossible to manually seek out every site where you would like your ad to show. It is also difficult if you are running a standard content campaign to always have your ad shown on a site that is converting, because other advertisers are in the potential queue to show up on those sites as well.

Therefore, I like to use a third campaign—a placement campaign—to increase your visibility on the content sites that are working well for you. Placement campaigns are different than typical content campaigns.

With keyword content campaigns, you choose some keywords and Google tries to show your ads on related pages.

With placement campaigns, instead of choosing keywords, you are choosing individual sites (or even specific placements within a site) where you want your ad to show.

If you run a placement report and find sites where your keywords are converting, add that site to a placement campaign so your ad will always show there. Don’t trust to Google-chance.

Take control of your content placements

Technically, you can do this for an ad group within your existing content campaign. However, this ad group will use the same budget from your content campaign and may not always be shown. Because you know that this site does convert, don’t you always want your ad to show on that site again?

By using a placement campaign, you are setting a different budget for these sites that you know are converting. Then, you can even set bids by each of those placements based upon how much each click is worth to your company. This campaign should also have a much higher budget than the ubiquitous content campaign as it is sending converting traffic.

By setting up three campaigns this way, your content campaign is a exploratory campaign. You are looking to see what sites do well so you can add them to a placement campaign where you always want your ad to display. Or, you are looking for sites that are not doing well so you can block your ad from showing on those sites. Since the ROI of this discovery campaign can fluctuate significantly, it is useful to set your exploratory budget at a lower target than you might have available. This budget may or may not give you good returns. However, it will let you optimize the content network over time and find the sties that are converting.

This leads to my favorite account structure:

• Search campaign: High budget. Bid by ROI by keyword
• Content exploratory campaign: Low to medium budget. Find what sites are converting or not converting
• Placement campaign: High budget. Only show your ad on content sites that are converting

Since the point of your discovery campaign is to find new sites to either block or to add to a placement campaign, and you want to make sure you maintain ad serving control, when you add a new site to the placement targeting campaign, block that same site from your content exploratory campaign.

Why? if you do not block that site from your content exploratory campaign, Google will make the decision whether ro show your placement ad or your campaign ad on that site. Since you have already chosen to always show that ad in your placement campaign, you do not want your exploratory campaign to use up budget designed explicitly for finding new sites.

Every account is unique & enhanced content campaigns can help

Within each ad group that will be shown on the content network, there are three ways that your ad can be triggered:

1. The default settings mean your ad is placed based upon your ad group keywords.
2. In placement targeting, you do not choose keywords. Rather, you choose websites (or sections of websites) where you want your ads to appear.
3. In enhanced campaigns, you choose both a web site and keywords. Therefore your ad is only shown on a particular web site if the page’s content also matches your keywords

Since the user mindset is different on various types of content sites (consider reading a digital camera review versus reading the sports scores), the interaction of those sites with your ads is also unique. Additionally, different types of users visit different types of sites (consider someone reading the New York Times versus National Enquirer).

This means that a site that does well for you may not do well for another account. You should not automatically assume that if a site does not do well for one of your accounts (or ad groups) that it will not do well for another ad group.

If you see a particular site where you think that your ad should work, or a site with so much traffic you really want to test that site you can target that site specially within your content exploratory campaign. There will be times where you might not want to just block a site because it’s not converting. If there’s a possibility of many conversions, it can be useful to test out that site with different ad copy or different rich media ads.

If its a small website, then create an ad group with just that site’s placements. Write ads specific to that site or its user base. Then run reports to see if one ad copy is converting at a lower cost per conversion than other ad copies.

If its a large website, then you might not want your ad to be shown across the entire website. For instance, the New York Times is a huge website. If you only sell a product targeted to stock brokers, you might want to just be shown on the New York Times when the article is also about stock brokers. In this case, use the enhanced targeting option and combine both keywords and placements together.

You can also create demographic based campaigns to test out different websites.

Highly Unique Targeting

With the content network, you set your ad to reach:

• Users on the New York Times
• Between 6-9am
• On a Monday Morning
• If the user is in the business section
• And the article is about stock brokers
• If the user is in Chicago (or the article is about Chicago stock brokers)

The more you segment your audience, the smaller your audience becomes. That’s an important note as you see dwindling impressions. However, that should also increase your conversion rate.

Therefore, if you see a site you want your ad to be on, but you cannot seem to convert those users, you might not immediately block that site; but instead, write unique ads, use different keywords, use ad scheduling, etc to try and engage that audience.

The work flow process

Here’s a simple step-by-step workflow to optimize your content campaign:

1. Run a placement performance report
2. If a site is doing well:
   • Block it from your content campaign
   • Add the site to your placement campaign
3. If a site or category is not converting:
   • Block it from your content campaign

The content network is a nebulous place filled with diamonds, sapphires, quartz, and fake jewels.

We want our ad to show on the diamond sites; therefore, make sure they are in a campaign with a high enough budget that your ad is always on that site.

Sapphires and quartz might not send as much value as a diamond; however, if the value is above 0, then it is more useful to set a bid based upon the returns of that site and not just block it. By using placement targeting, you can set a different bid for each website.

Fake jewels might look promising, but under close examination, there is little value left for our ads. Don’t have your ad shown on these sites – block these sites from showing your ad so you can find new websites that will send you higher value.

By organizing your account to using three different campaign types: search, exploratory content, and placement targeting; it’s quite simple to continue exploring the web to find new sites that are converting across the content network and bring more total profit to your company.

The display URL is an often overlooked element of ad copy. It’s most common usage is to signal to a searcher where the ad click leads. However, the display URL is also a quality score factor and has different editorial requirements than the rest of your ad.

The display URL is also a prime candidate for ad testing. A good display URL can increase your click through rate, increase your conversion rate, and improve your quality score which in turn, causes your ad to show higher in the search results or lets you pay less for your current clicks.

I’ll first discuss the basics of the display URL, quality score impacts and then showcase ways of testing display URLs.

Display URL basics

An AdWords display URL is the 4th line of your ad text. This line of ad copy signals to the searcher what website they will land on after the click. The display URL does not have to actually be a page on your website; however, the root domain of the display URL and the destination URL (where the user actually goes) must be the same.

For instance, if your sent a user to www.example.com/apple/ipod/page23.html, the display URL could be any of these:

• www.example.com
• example.com
• www.example.com/apple
• example.com/iPod
• example.com/compare
• www.example.com/WeSellStuff
• iPod.example.com

Example.com is the root domain. The user is going to a page within that domain. However, you do not have to put the actual page in the display URL—you just need to make sure that the root destination and display domain are the same.

Trademark concerns

In the US, UK, Canada and Ireland, you can advertise on any keyword you desire. However, if that word is trademarked and the trademark holder has filed an exception request with Google asking that advertisers not use their trademarks, then you cannot use those words in ad copy (More about this restriction from Google).

This is the editorial requirement for ad copy. However, the display URL is not currently encompassed by this policy. At present, while you can not use a trademarked word in ad copy, you could use it in the display URL.

For example, Apple has trademarked iPod. If you were to write the headline:

80gig iPods for Sale

the ad would be disapproved for trademark concerns.

However, if you were to write the display URL:

example.com/iPod

the ad would not be disapproved for trademark issues.

Whenever you are thinking of using a trademark in your ad campaign, please consult an attorney if you have questions. This article should not be accepted as legal advice.

Editorial concerns

Google made a recent announcement that all destination URLs within the same ad group must now go to the same site. The actual destination pages can be different; however, the root domain must be the same.

This also means that all dispaly URLs within the same ad group use the same root domain. You could use these display URLs:

• example.com
• www.example.com/widgets
• widgets.example.com

as they all lead to the same root domain.

However, you could not use all of these display URLs within the same ad group:

• example.com
• widgets.example.com
• example.widgets.com (different root domain from the above two display URLs)
• widgets.com (different root domain from the above two display URLs)

The only two major editorial concerns when writing display URLs is that the root domain must be the same in both the display and destination URL and that all destination URLs in the same ad group must go to the same root domain.

Display URLs and quality score

One of the quality score factors is the display URL click through rate within that ad group. The display URL is treated separately from the rest of the ad copy when determining the CTR used in quality score. While display URLs are good to test for increasing CTR and ultimately quality score, they are much better to test for conversion rates.

Testing display URLs

The display URL signals to the user where they are going after the click. It is also part of ad copy, so can be synergistic with the ad copy in increasing CTR and conversion rates.

Should you use a /folder after the display URL?

If your site is small or tightly themed, a /folder is not always necessary. However, if you have a large site such as IBM.com and the user did a search for ‘blade server’, which will lead to a better CTR?

• IBM.com
• IBM.com/BladeServer

You probably have an answer by looking at the above options, but it is useful to test these two elements against each other.

Do you need the WWW?

For most searchers, we know that ads lead to websites. However, there are demographics where I’ve seen the ‘www’ increase CTR. This is usually an older crowd where the ‘www’ signals that the click will lead to a website.

This is becoming less of an issue these days, but remains worthy of testing.

Can you create synergies with the ad?

While using a /folder for a product name can be useful for showing someone where on your site they are going to visit, it can be useful to adding a key word from your ad copy into the display URL. In this instance, the ad reinforces a comparison page, which is very useful for consumers that are still early in the buying cycle.

Test different folders for the same product

If you start to run too many words together in a display URL, they start becoming harder to quickly scan and read.

Testing a few different one vs multiple word folder types can lead to finding more readable, and hopefully higher CTR, ads.

To Subdomain or folder?

One more test to consider is determining if a folder or a subdomain increases CTR and conversion rates.

Conclusion

The display URL is one of the least regulated aspects of your ad copy. There are a few editorial requirements that Google imposes upon you; however, beyond a few simple rules, you can have fun testing your display URL.

It is always important to remember that your ad copy is the only part of your AdWords account a searcher ever sees. A searcher has no idea what your keywords, geographic targeting, ad scheduling, or other campaign settings are. A searcher probably has not seen your website. The only part of your account that induces a searcher to click on your ad is the ad itself.

The display URL is part of that ad copy. It’s the part that tells a consumer where they are going after the click. Don’t regulate the display URL to a secondary consideration when writing ad copy. Think about both your ad copy and the consumer when writing display URLs.

A bad display URL can ruin great ad copy. A good display URL can increase CTR, conversion rates, and quality score. It doesn’t take much effort to test display URLs. Are you using a display URL that makes your AdWords spend more profitable?

The economy’s downturn has finally reached the online advertising world. Spends are dropping, advertisers and clients are reducing budgets, and many are confused as to the correct course of action.

What do chief marketing officers think of marketing during a down economy?

Marketing is more important than ever. 94 percent of CMOs believe that “a tough economic period is precisely the time when marketing plays a key role.”

Source: Epsilon, “Epsilon CMO Survey,” September, 2008.

When you stop marketing, you lose touch with your clients. You stop understanding what is important to your consumer base. Many believe that customer service and marketing is more important in a recession than at any other time. If you lose touch with our client base, how can you see the trends your clients are facing?

In addition, you also lose access to valuable data. If you completely stop marketing, how will you know when the economy is rebounding for your products or services? The economy will rebound for different companies at different times. If you are not paying attention to your company’s data, it will take you longer to realize that its time to start advertising again. Those who have continued to advertise will make smarter decisions for their paid campaigns.

Will you understand how your customers have changed? Are there new features, products, benefits, or messages that you need to embrace? It’s not always about click prices and keywords. Effective PPC embraces landing pages and ad copy that speaks to your customer’s desires. You need to continue testing your offers regardless of the economy to learn when your customer’s are purchasing your product for different benefits. Or, more importantly, when they stop engaging your product completely and it’s time to invent the new version.

Where do we start making changes?

All sales and keywords can be attributed to a section of the sales funnel. In a booming economy, companies spend time examining the top of the funnel. How can we find the most prospects and start them down the funnel path towards conversions?

In a down economy, the exact opposite should be true. You make most of your sales from the ‘shop’ and ‘buy’ sections of the funnel. Spend more time testing and executing properly on the bottom of the funnel. As you cap out your effective spend at the bottom of the funnel; move up towards reaching more prospects.

Several companies will completely disagree with the above advice for three reasons. And they are three good reasons that you need to keep in mind.

Reason 1: These companies work with larger brands that live in the top of the funnel. For example, most Coke or Pepsi commercials spend their time at the top of the funnel. It’s the convenience store that converts the soda buyer to an actual product sale.

Reason 2: B2B, long sales cycles, new products, and similar examples need to spend some time at the top of the funnel. Often B2B companies rely on phone sales to close the deal, and a whitepaper download to start the funnel process. Without some awareness, they will not make any sales. However, it can be useful to consider the whitepaper download (or other alternative conversion activities) as the actual conversion, set a cost per download target, and bid within the top of the funnel on a cost per acquisition (CPA) basis.

Reason 3: Shouldn’t you keep an eye on total sales and profits? Yes, most definitely. Whenever you change your PPC tactics, measure not only the changes in CPC, CTR, conversion rate, etc – but always measure a change to total profit and total sales. If you spend enough time at the bottom of the funnel, you can eventually hit a very nice CPA. However, if you only have one sale a month, a low CPA will not keep you in business.

Therefore, spend most of your time optimizing the bottom of the funnel; however, never forget to measure profit and to make sure your marketing efforts are supporting your business goals.

Change your focus

Consumers will look online more. This year has seen a 10% increase in people who research online before buying in the store.

Source: Vertis Communications, “Customer Focus 2008: Holiday Retail study,” October, 2008.

Focus on price & savings – The above quote focused on the holidays. However, not everyone operates within the retail industry. There are two ways to focus on price and savings within your PPC account.

The first is to put yourself in the consumer’s shoes. Regardless of the economy, if your TV breaks, you’re going to buy a new one. None of us in the United States have a choice — taxes are due in less than three months. In a booming economy, the more features the better. In a down economy, we focus less on mass amounts of features and more on staying within our budget.

You should test ad copy and landing pages that focus on savings, and ‘good enough’ qualities.

Focus on additional value – Can your company add more value than your competitors? Value can take many different shapes. Consider these value statements:

• “We saved $1300 on our taxes by using Jack & Jill’s Accounting Service”
• “Buy Jason’s Firewall, and receive our spyware protection software for free”
• “Buy 3 copies of Jason’s Firewall, and we will upgrade you to Jason’s Firewall Pro for no additional cost”
• “We saved $800 in marketing expenses by reading Search Engine Land’s Paid Search Column”
• “Prius owners save $2000 on gas each year due to our patented hybrid engine system”

There are many ways to add additional value to your products. From saving money, to free upgrades, to not having to buy more of another product. You don’t even have to change your margins to add additional value. The last two value statements above just showcase how the product adds value to a consumer’s life, and don’t actually change the product’s price or the company’s bottom line.

Add money-saving keywords – We’re seeing more searches that include the words ‘discount’, ‘coupon’, ‘sale’, etc. This is backed up by a comScore data showing a sharp increase in visits to coupon sites.

Consider these discount saving keywords:

• Crocs coupon code
• Zappos coupon
• Cheap shoes
• Inexpensive shoes
• Discount shoes
• Low cost shoes
• Shoe Sale

You may wish to do some additional keyword research focusing on how your potential customer’s are changing their search behavior during this economy to search for savings.

Account optimization

Conversion optimization – This statement should be said with every single PPC article written; “Always test your ad copy, landing pages, and keywords to maximize your business goals”.

This doesn’t change in a down economy. The reason to test is two-fold.

First, it helps you understand your customers. The better you understand them, the better you can market to them (changing keywords, ad copy, and landing pages), the higher your conversion rates become, the lower CPA, the higher your ROI.

Second, it helps you see customer’s evolve. Some people will see a worse economy in the next few months. Others will see a better economy. What will your customers see? If you are constantly testing, you’ll see when the feature laden ads that don’t mention additional value start to increase in conversion rates. This is a sign the economy isn’t affecting your products as much as it was previously. You need to watch these signs so you can change your ad copy and landing pages to match your customer’s desires.

Refine your ad exposure – Spend time researching negative keywords. It is just as important to keep your ad from showing on a non-converting keyword as it is to show your ad on converting keywords.

Measure conversions by ad copy. Since we bid by keyword (or sometimes ad group), its easy to get lost in only measuring conversions by keyword. Do not forget to measure by ad copy. If you find ads not performing, replace them.

Location targeting. Run a geographic report. If you have geographies where you are having poor results, you can either:

• Block your ad from showing in that geography
• Create a new campaign targeted to that geography (this includes ad copy and landing pages) to see if you can market effectively to that area

Ad Scheduling (otherwise known as day parting). Measure conversions by day of the week, and time of day. If there are clear patterns, change your bids based upon those response rates.

In a good economy, we might want more total ad exposure and not keep our ad from showing at time where we might break even or have a slightly positive ROI. In a down economy, refine your exposure so every time your ad is shown, you have the highest possible chance for a conversion. If your budget is decreasing, every penny spent counts more than before.

Engage Your Audience – Can you add new conversion activities?

Examine your current offerings. Do you offer only a buy or contact option to a search? Do you only offer a whitepaper download? Are there other conversion items you can add that will help engage your audience?

Consider…

• Free product demo webinars
• Free conference call advice
• Subscribe to our newsletter
• Follow us on Twitter
• Subscribe to our blog
• Visit our booth
• Tour our offices
• Free product videos
• Downloadable whitepapers

Do not confuse your audience!

While it can be useful to try multiple conversion actions; do not put them all on a single page. All you will end up doing is confusing them and not ending up with any conversion.

Try a newsletter and RSS subscription. Then market your webinar to that audience that subscribes to your updates.

Try follow us on Twitter, then offer a free conference call for Twitter followers.

Try a simple, ‘attend our free webinar’ and them send out a complimentary whitepaper two weeks later to those who haven’t contacted you.

Do not bombard your customers, or potential customers with advertisements.

Find alternate ways to engage your audience.

Marketing in a down economy

I’m not an economic expert, so I won’t attempt to speculate on how long the economy will be in a downturn. However, I can say that we can weather the storm by becoming more customer centric. In any economy, you should set bids by your business goals. Any advice that contradicts that PPC axiom should be examined with common sense.

However, in a down economy, we’re adjusting the normal best practices for PPC.

• You’re testing value statements in a world that lives by feature comparison and benefit buying.
• You’re spending more time examining discount keywords and ad copy.
• You’re sinking time into creating landing pages that will not be useful when the economy turns around.
• You might be adding negative keywords that will be deleted when you can increase your company’s exposure.

However, the most important aspect to marketing in a down economy?

Thinking more about how to engage our audience.

It’s your audience that will ultimately determine your success.