BUS155 Ch 6

Learning Objectives
After reading this chapter, students should be able to
• Identify the various types of search engine advertisements and their parts.
• Analyze the effectiveness of an existing search engine advertising campaign.
• Determine ways to improve an ad's position without increasing the bid amount.
• Create a categorized list of keywords on which to advertise on a search engine.
Before Google became one of the most profitable companies in history, it built a worldwide user base of people that were searching for information about anything and everything online. Back then there were no ads, and Google made very little money.

In October of 2000, AdWords, a method for companies to pay to advertise on Google, was launched. Here is the AdWords official press release. The widespread use of Google as one of the most popular search engines, plus the user-friendliness of the online ad-buying platform, caused AdWords to swiftly gain popularity from online advertisers and ad agencies. To this day. Google still makes more money from its search ads (now a part of its Google Ads platform and no longer called AdWords) than from all of its other revenue sources combined. Almost 90% of Google's revenue comes from ads purchased through the Google Ads platform.
All of the search engines today have adopted the same pay-per-click (PPC) advertising model used by Google's search ads. In this model, search engines charge advertisers each time their ad is clicked, which means the advertisers don't have to pay for their ad to show up in the sponsored section of search results. They only have to pay when searchers read their ad and decide to click through to the website.
This chapter is dedicated to PPC advertising on search engines, sometimes referred to as search engine marketing (SEM). This chapter discusses why search engines make tens of billions of dollars annually through this model, but more importantly, it offers actionable insights into how a website owner can effectively use this advertising platform. Search engine advertising drives more new website traffic than any other online advertising method, and no other advertising method enables a more precise targeting of ready-to-purchase customers. Additionally, a site owner can measure the effectiveness of search engine advertising more reliably than any other method of advertising. Because of these many benefits, no website should overlook search engine advertising.

The Story of Darrell and Bart

Darrell is an avid wakeboarder. When his well-used Wakeboard breaks one Saturday, he does not hesitate to begin his search for a replacement. That night he does a Google search for "wakeboards".

Darrell would be the perfect customer for Bart's Water Sports website. Bart's website sells water sports equipment of all varieties, such as Wetsuits, tow ropes, paddle boards, and, of course, wakeboards. Bart likes advertising on Google because he knows his advertisements are only being shown to people who are interested in the products he sells. Whenever someone searches for "wakeboards," Bart tries to make sure Google shows his advertisement on the search engine results page (SERP).

When Darrell searched for "wakeboards," Google showed this page (Figure 6.1.1). The fourth advertisement was for Bart's Water Sports. Bart had written the ad long before. He offered a $40 gift card because it was something different than the other wakeboard advertisers were offering, and he hoped it would entice a good share of searchers to click on his ad and consider buying one of the many wakeboards he had in stock. Bart did not have to pay Google anything for his ad to show up on the SERP. He only pays when someone clicks on his ad.

Fortunately for Bart, Darrell does click on Bart's ad. For whatever reason, the offer of a $40 gift card was attractive to Darrell, so he saw Bart's wakeboards page (Figure 6.1.2). When Darrell clicked on Bart's ad, Bart was charged $.88 by Google. Bart was happy to pay this charge because he knew he had the chance to earn a paying customer. After browsing all of the wakeboards Bart had to offer, Darrell purchased a wakeboard for $340, so the $.88 was well-spent.

In well-managed search engine campaigns, everyone can come out ahead. In this particular example, Darrell came out ahead because he found a wakeboard he liked. Bart came out ahead because he made a $340 wakeboard sale and only spent $0.88 to get Darrell to his website. The search engine came out ahead because it earned $0.88 for connecting Darrell and Bart.
Not everyone who advertises on search engines comes out ahead like Bart. The goal of this chapter is to provide the skills necessary to make your website a winner in the search advertising realm.

The power of search engine advertising boils down to one simple fact-companies advertising on search engines only have to pay for advertisements going to people who have professed a real and current interest in the product offered by the advertiser. Searchers profess an interest every time they do a search. If the interest professed by a searcher aligns with the product or service offered by a company, the company wants to advertise to that person.

In the story of Darrell and Bart, Bart advertised to Darrell because Darrell professed an interest in purchasing a wakeboard when he searched for "Wakeboards." But what if Darrell had instead searched "are wakeboards dangerous?' In that case, Bart would probably not have wanted to advertise to Darrell, because his search would have indicated that he was not really interested in purchasing a wakeboard.

This example illustrates the importance of choosing ad placements carefully. The most important step in conducting good search engine advertising is to select which searches to target for ad exposure. Advertisers select when to show their ad based on (1) the search phrase entered by the searcher and (2) other characteristics of the search, such as the searcher's location, a device used (mobile, tablet, or desktop), time of day, etc.

A company wishing to advertise on a search engine provides a list of keywords to the search engine. (The term "keyword" can refer to a single word or an entire phrase of multiple words. Wakeboards would be considered a keyword, and the phrase wakeboards for sale would also be called a keyword.) The company's ads will only show up on the SERP when the search phrase entered by the searcher matches one of the keywords on the company's list. But the searched phrase does not have to be a letter-for-letter match of a keyword to activate an ad. It would be impossible for an advertiser to specify every possible search on which to advertise. For example, Bart wants to advertise to anyone interested in purchasing a wakeboard, but he couldn't possibly anticipate every potential misspelling and variation on the search, including searches like wake board (with a space), wakebords (misspelled), wakeboards online, wakeboards for sale, and any combination of these variations and misspellings. As a result, Bart
would miss out on valuable advertising opportunities if the search phrase always had to match one of Bart's keywords perfectly.

On the other hand, if Bart instructs the search engine to show his ad anytime someone searches for wakeboards or any misspelling or variant of it, he would advertise to people searching for things like are wakeboards dangerous, which would waste his money. (Even though Bart only pays for clicks, not exposures, some of these searchers would likely click on Bart's ad because many searchers do not carefully read the search results before clicking on them.) For this reason, search engines provide a variety of match types for advertisers to specify the searches they wish to target.

Exact Match.
The name of this match type is a bit misleading. "Exact match" implies that the searched for phrase has to be a letter-for-letter match to one of the keywords in the advertiser's list, but in reality, a search phrase will still match an exact match keyword if the search is a close variant of the keyword (as explained by Google here). A close variant includes misspellings and even abbreviations. To use exact match for a keyword, the advertiser places brackets around [a keyword]. The fundamental advantage of using exact match is that the advertiser will waste little to no money advertising to searchers that are not interested in the advertiser's products. The disadvantage is that the advertiser is likely to miss out on valuable advertising placements when searchers search for the advertiser's product using an unanticipated variation.

Broad Match.
On the opposite end of the match spectrum, an advertiser can utilize broad match by listing a keyword without any modifying characters. Broad match directs Google to show the ad to any search that broadly' matches that keyword. The primary advantage of using broad match keywords is, well, breadth. Advertisers are not likely to miss out on valuable advertising opportunities because their ad will show up for anyone doing a search relevant to their products. The main disadvantage of broad match keywords is that ads are likely to show up for searches in which the searcher had no intention of purchasing the advertiser's products. For example, someone who searches for a wakeboard fail video would probably see Bart's ad if Bart used broad match.

Phrase Match.
One step above exact match is phrase match. To use phrase match, an advertiser puts quotation marks around "a keyword", which directs the search engine to show an ad only when the searcher's query includes the quoted keyword, but the search can include additional words before or after. If Bart included "wakeboard sale" (including the quotation marks) on his keyword list, his ad would show up for the searches wakeboard sale best wakeboard sale, and wakeboard sale in Milwaukee, but not for the search wakeboard half price sale.

Modified Broad Match.
This match type was added as an option on Google's search ads platform in 2015 because advertisers wanted a middle ground between broad match (which is often too broad for advertisers' liking) and phrase match (which is often not broad enough). Placing a plus sign in front of a +keyword activates modified broad match. If Bart included "+wakeboard sale" (not including the quotation marks) in his list of keywords, Bart's ad would show up only for searches that included wakeboard or a close variant, but the term "sale" would allow for broader matches like discount or cheap. If Bart included "+wakeboard +sale" (plus signs in front of both words, not including the quotation marks), his ad would show up for searches using both words, close variants of both words, in any order, and including other words, like wakeboards half price sale and sale on wakebords (misspellings typically count as a close variant).

Negative Keywords
It can often be helpful to instruct the search engine where not to display an ad. For example, if a company sells men's athletic shoes, but it does not sell men's dress shoes, it should add dress as a negative keyword to ensure that no matter what search query a consumer enters, the ad will not be displayed if the word dress is used.

Other Selection Factors
Just because someone does a search that indicates an interest in your product does not mean you should show them an advertisement. If someone in Canada does a search for wakeboards, Bart would not want to advertise to them if he cannot ship above the border. Below we list other factors that can be used in targeting search ads.

Geography.
If your products only ship to the continental United States, you clearly should not show ads in Alaska and Hawaii. If you are a local service provider that serves only five zip codes, you should only advertise to those five zip codes. Search engines allow advertisers to specify the geographic regions they wish to target with their ads. Even if you sell products to all geographies, you may wish to create multiple ad campaigns, each targeting a smaller geography. This allows you to include geography-specific information in the ad itself ("Free shipping throughout Texas") or to direct people in different geographies to different landing pages. You may also find that certain geographies are more profitable than others, so you can allocate more ad budget to those geographies.

Device Type.
Search behavior and purchase behavior differ dramatically across users of desktops, mobile phones, and tablets, so search ads should also be adapted to these differences. If mobile ads prove less profitable than desktop ads, advertisers should pay less for ads on mobile phones (more on paying for search ads in the next section). Alternately, ads encouraging searchers to make a phone call might be more successful on mobile devices, so these ads should be targeted toward mobile devices instead of desktops or tablets.

Timing.
Many businesses find that searches vary throughout the day, not only in volume but also in quality. For example, an enterprise software company might find that serious potential customers do searches between regular business hours, so searches at night or on weekends are not profitable. The advertiser can restrict the ads to show only during business hours.
Search timing is especially important for ads on mobile devices. Many mobile searches are now-or-never searches. For example, if someone does a search for restaurants near me on a mobile device at 12:15 p.m., it is likely that she is looking for a place to eat lunch, so the likely conversion rate on this search is very high. If that same search is done on a desktop computer at 9 p.m., the likely conversion rate is much lower. A restaurant that caters to lunch eaters should be more aggressive in its mobile advertising during the lunch rush for mobile searchers near their location.

Language.
Advertisers should remember that not everyone within a particular geography always speaks the same language. Segmenting based on language can be an effective strategy for increasing advertising returns.

Keyword Research
Choosing the right keywords for an ad is critical to the success of any PPC ad campaign. The following are some of the most important steps of keyword research:
Search For Yourself. First, imagine being in the shoes of a potential customer. Sit down at the computer, think about what those customers would search, and start typing in what comes to mind. What keywords would they type in? Take special note of what Google Auto-Suggests, as shown in Figure 6.2.1.

Competitive Research.
While searching from the perspective of a shopper, take special note of who is advertising for those keywords. Using keyword research tools, input their URL to estimate what those competitors pay per click to advertise on those keywords. One can also gain insights into what other keywords they have selected as relevant places to advertise, and determine if competing on those same keywords is possible. If not, consider variations of those keywords where the competition may not be choosing to advertise. Could those be potentially lucrative keywords? NOTE:
SpyFu.com, SEMRush.com, and Keywordspy.com are tools that can be used for online advertising competitive research.

Google Keyword Planner.
The tools just mentioned are useful for creating a thorough keyword list, but they are costly. Advertisers with a lower budget should probably stick with Google's Keyword Planner tool, which is free. Figure 6.2.2 is a screenshot from Google's Keyword Planner tool:
Based on the information provided above, a company can make data-driven decisions about which keywords it wants to add to its ad campaign. Here are four initial things to consider for each keyword:

Searcher Intent.
Think about what the searcher wants and what your website provides. For example, if the Bart's Water Sports website does not include consumer reviews for each wakeboard it sells, then wakeboard reviews should likely be added to the negative keyword list. Otherwise, Bart will end up paying for clicks from a searcher who wants to see reviews before buying and thus is unlikely to convert on Bart's website.

Average Monthly Searches.
These data can be used to project how many impressions and clicks an ad can acquire in any given month. By clicking on the graph icon to the left of the average search volume number, one can see how many times each keyword was searched over the last 12 months. High search volume keywords will usually drive more impressions and clicks in general, but even keywords with low search volumes can often generate valuable traffic.

Level of Competition.
High-competition keywords mean that relative to other keywords, a higher number of advertisers bid on that keyword. Keywords with high search volume and low competition are ideal for getting the most traffic for the least amount of money. Keywords with low search volume and high competition tend to send less traffic and cost more money per click, but sometimes the high CPC is justified because these keywords convert at higher rates.

Suggested Bid.
Using historical data, Google looks at what other advertisers are paying per click to advertise on that keyword. Advertisers use this information to determine how much they can expect to pay per click to advertise on specific keywords. Because of the unknown variables of each advertiser, it is not the exact amount that each advertiser will pay, but it does provide a close estimate.

Branded Keywords
A common question asked by digital marketers is whether they should run search ads on queries in which they have earned the top organic listing. Most companies easily earn the top organic listing for searches that include their brand name, so an alternate question is whether a company should pay for advertisements on searches that include their own brand name. Consider a search for best buy tv (see Figure 6.2.3). The searcher clearly wants to purchase a TV from Best Buy (or at least wants to see the TVs available at Best Buy). Best Buy has chosen to purchase ads on this keyword, which means they might have to pay for this click when it would likely have been free if they chose not to advertise (because all of the organic listings are also for bestbuy.com). Shouldn't Best Buy stop running these ads and save its money?

Many digital marketers believe the answer to this question is no, for several reasons. First, if Best Buy does not advertise on its own brand name, competitors may start advertising more aggressively on its brand name and steal Best Buy's customers. Second, ads on a company's brand name are inexpensive (the reason for the lower price tag for brand searches has to do with the ad's quality score, which you will learn about in Section 4). Third, real estate on the SERP is valuable, and buying ads allows a company to take up more space on the SERP.

However, other digital marketers believe the answer this question is yes. Economists at eBay ran experiments on millions of keywords on which that eBay traditionally ran paid search ads. They found that search ads on branded keywords were ineffective, because when they stopped running the ads, searchers would either click on the organic link to eBay or type in eBay's URL directly. They also found that search ads for unbranded keywords (searches that did not include eBay) were also often ineffective because many of those searchers would have visited eBay anyway through the organic listing. These results stood in dramatic contrast to eBay's analytics, which showed enormous returns to search ad spending (i.e. analytics
showed that every dollar spent on search ads produced several dollars in revenue).

This situation provides yet another example of the difficulties of attribution (see Section 5 of Chapter 3). If a consumer clicks on an ad and purchases from the website, it seems straightforward that the money was well-spent, because the ad caused a sale. In reality, the ad may not have caused the sale at all, but instead cost the company money to acquire a sale that would have happened anyway. As we stated in Chapter 3, the only way to be confident of the causality of advertising is to conduct a carefully-controlled experiment. If a digital marketer is unsure as to whether paying for ads on branded keywords is effective, he or she should run an experiment.

Once an advertiser has determined where his or her ads should show up, the advertiser needs to have an actual ad to show to searchers. Creating ad copy for search ads is much more straightforward than creating ad copy for other media types, because the ads are strictly text, thus foregoing the need for expensive photography or video editing and scripting. But this does not mean that ad copy should not receive its due effort. The quality of a search ad's copy can dramatically affect its success.

Ads in search results continually evolve, but they generally have five key elements: (1) Final URL. (2) Headlines, (3) Path, (4) Description(s), and (5) Ad extensions. Figure 6.3.1 shows a text ad along with the text ad creation form that Google Ads provides.

Final URL.
The final URL is the URL of the landing page that customers will be taken to when they click on the ad. This full URL is not shown to searchers, because it may be too long to display on the screen. Even though searchers do not see the final URL, it can be considered the most important part of the ad, because if it is incorrect, the searcher will be taken to the wrong page of the website, or worse yet, may be sent to a broken, nonexistent page (where they will lose the visitor but still pay for the click).

Headlines.
Advertisers can create as many three headlines. Headlines can be up to 30 characters long. Google often changes the configuration of ads (e.g., it may put all three headlines on the same line or it may move the third headline to the second line) when its algorithms indicate that a particular configuration will lead to a higher click-through rate, so an advertiser can never be confident exactly how the ad will show up on the SERP.
Path.
The path, also called the display URL, is the simplified URL that is shown to searchers. This can simply be the domain name, or the advertiser can indicate to the searcher what kind of page the ad leads to by specifying further URL specifics. The State Farm ad from Figure 6.3.1 indicated to searchers that the ad would lead them to a CarInsurance Quote page rather than just to www.statefarm.com. The Path fields have a 15-character limit.
Description(s).
Advertisers can write one or two description lines, which can be up to 90 characters long. An advertiser would typically expect to see the description displayed below the headlines, but Google often changes the configuration of the description, for example by putting the description next to headline 3.

Ad Extensions.
The parts already discussed (the final URL, headlines, path, and description) make up a "typical" search ad. Interestingly, however, "typical" search ads have become more rare than common. Instead, most ads now are displayed with one or more extensions. The State Farm ad in Figure 5.3.1 was displayed with three ad extensions: (1) a review extension, (2) a consumer ratings extension, and (3) a sitelinks extension, in that order. Advertisers do not have the ability to control when these types of ads with ad extensions appear, but each advertiser does have the ability to opt in to each ad extension individually. Other ad extensions include the following: 

・Location extensions to display store addresses with a link to Google Maps
・Offer extensions to include printable coupons for offline purchases
・Shopping star review ad extensions
・Reservation extensions for travel/hospitality
・Call extensions to allow mobile searchers to call the store directly

These extensions are typically effective in helping an ad stand out and can boost an ad's click-through rate (CTR). A complete list of the various extensions offered by Google can be found here.

Dynamic Search Ads.
Dynamic search ads allow advertisers to turn over some control of their search ads to Google. Google will pull information about the advertiser's products from its website and show the advertiser's ads on searches Google deems are relevant. Google can dynamically change the ads based on what it observes to produce better results for the advertiser. These ads are only recommended for advertisers with enough budget to experiment with the ads first to determine whether they are profitable. The advertiser should also have a large volume of search ads. because the machine learning algorithms used to optimize these ads require extensive data to be effective.

Remarketing Lists for Search Ads.
Even a very good retail website will only achieve a 5% conversion rate from its paid search traffic. Among the many reasons why the other 95% did not convert is the possibility that they liked what they saw on the website, are primed to convert, but simply were not quite ready to make a purchase at that moment. Given that circumstance, an advertiser would like to make sure that its ad shows up prominently when this previous non-converter does another search for a product sold by the advertiser. Remarketing lists allow advertisers to create ad campaigns specifically for searchers who have previously visited their website. Knowing that a consumer has previously visited the advertiser's website recently and browsed handheld drills, for example, might make the advertiser want to pay more for prominent placement when this consumer searches for drills. Like dynamic search ads, remarketing lists for search ads should likely only be used if other search ads are optimized and the advertiser has budget remaining.

Writing Effective Copy
As with any kind of advertising, there is no guaranteed method for writing ad copy that will outsell one's competitors. But there are some guidelines that can help maximize the likelihood of success with search ads. When working on writing ad copy, web designers should remember to include three things in each ad:

1. Keywords for relevance
2. Unique Value Proposition (UVP)
3. Call to Action (CTA)

Keywords for Relevance.
Click-through rates tend to improve when the keyword being searched is in the ad. This is a first indication to the searcher that the website has exactly what the searcher is looking for. In addition, if the keyword is found in the description or in the path (the URL displayed to searchers), that keyword will be be bolded (but not if the keyword is in the headline). Bold type can further attract the searcher's attention. 

Unique Value Proposition (UVP).
Because a company's ad will be one of about 7 ads on the SERP (in addition to 10 organic search results), its ad must communicate something to the searcher to indicate that what the site has is superior to the other sites. The website's unique value proposition (UVP) is the single thing that sets the site apart from the competition. Here are some examples:

• Geico® could save you over $500.
• Progressive was named the #1 insurance website.
• US Auto Insurance Now has the cheapest car insurance in Utah (where the search was done)
State Farm is rated A++ by Insurance.com (this was an ad extension, so State Farm did not craft this UVP in its ad copy, but this ad extension nonetheless communicates a UVP). Each of the advertisers (competitors) has its own UVP. Advertisers should consider how their website's UVP compares to their competitors' UVPs. It is especially difficult to stand out when competing against well-known brands. US Auto Insurance is a great example of a company that is using its UVP to overcome this deficit in brand awareness.

Call to Action (CTA).
If a company's ad is relevant to the searcher, and the searcher likes the company's UVP, he will be favorably inclined to click on the ad. A CTA might be the last nudge that is needed to push the searcher over the edge and earn the advertiser a click. Geico, State Farm, and Progressive all include basically the same CTA: get a quote. Only US Auto Insurance is missing a CTA entirely. Their click-through rate might improve if they included a CTA. A strong CTA is considered good practice for search ads, but CTAs have declined in prominence over the past several years, which indicates that a CTA may no longer be crucial for success.

Testing Ad Copy
Fortunately, there is a way to test whether one's ad copy can be improved. By creating multiple versions of an ad, an advertiser can simultaneously run these various versions to see which ad attracts the most clicks. US Auto Insurance could, for example, create an additional version of its ad that included a CTA. Google would then randomly choose one version of the ad to show each time the ad was activated by a searcher. After several thousand searches, US Auto Insurance would be able to observe the CTR of each ad and determine which ad more effectively attracted clicks from searchers.

This method of testing, called A/B testing, can be used for any aspect of an ad, not just to test the inclusion/exclusion of a CTA. An advertiser can create multiple versions of each headline, multiple versions of the description, and multiple paths for an ad. The advertiser can run any combination of these different versions to test which ad is best. The ease with which an advertiser can test different versions of ad copy is one of the most important advantages of search advertising. Instead of relying on one's instincts to determine which ad is best, an advertiser can run a test and be confident in the accuracy of the results, because the results are generated by actual behavior. Create your first ads using the guidelines just provided; that is, use keywords, give a unique value proposition, and include a call to action. But after that, test multiple versions of your ad to continually increase your click-through rate.
When a company chooses to advertise on a billboard, it picks a billboard and pays a set rate per month to keep its ad on that billboard. While that rate could change at some point, the advertiser might go years paying the same price per month for the same as to appear on the same billboard. With paid search advertising, the price paid to advertise on any given keyword is dynamic-it could change from one minute to the next. This chapter explains how to buy ad space and how search engines determine where to show an ad. This chapter doesn't cover creating one's own advertising accounts with the search engines, but see ads.google.com and advertise.bingads.microsoft.com to learn more.

Ad Rank.
Search engines sell keyword ad space using an automated auction platform. In simplistic terms, advertisers place a bid on a keyword instruct the search engine the maximum amount they are willing to pay for a click from a searcher who has searched for that keyword), and the winning bid gets the best ad location for searches on that keyword. However, the bid from each advertiser is not the only factor that determines which ad "wins the auction" and ranks higher than the other ads. The Ad Rank (advertising position) of each advertiser is also determined by the ad's Quality Score, as illustrated in Figure 6.4.1.
Determing Ad Rank based on both bid amount and quality score ensures that search advertisements stay relevant to searchers, which increases the proportion of people clicking on the ads, which in turn means the search engine makes more money. Consider the following illustration as to why a search engine is better off determining Ad Rank from this combination of factors.

Assume an advertiser bids $10 per click for a keyword, but his ad doesn't get very many clicks. It is poorly written and every time people click on it, they leave the page quickly and go back to the Google SERP. Even though Google might make $10 per click, it is obvious the advertiser's website does not have what the user wanted. On the other hand, another advertiser may only be willing to bid $1, but her ad generates a high CTR and very few searchers who click her ad return to the Google SERP. That advertiser is helping Google achieve its goals of making money and giving searchers the results they are looking for. Here is the same scenario explained in diagram form:

Quality Scores.
Google uses a calculated metric called Quality Score along with bid amount to determine where a company's ad will appear. Google defines Quality Score as a measurement of how relevant your ads, keywords, and landing page are to a person seeing your ad. Higher Quality Scores can lead to lower prices and better ad positions."
Each time an ad has a chance to rank on a keyword, Quality Score is recalculated to see how it stacks up to the other competing ads. Three criteria affect Quality Scores:

1. Expected CTR. If an ad is well written and seems to generate more clicks compared to the other advertisers, Google is likely to reward it with a better ad position.
2. Ad relevance. The keywords an advertiser is bidding on should be used in the ad and on the landing page (the destination URL). This use of keywords shows Google that the ad and website are relevant to the keyword being searched.
3. Landing page experience. If people click a company's landing page and then quickly return to the SERP (bounce), the action sends a signal that the landing page did not, in fact, have what the searcher was looking for. If bounces occur frequently for those who click on an ad, this negative signal could affect the company's Ad Rank. When people come to a site, the company should deliver what its ad suggests.

CPC Bidding.
The maximum CPC bid is the amount the advertiser agrees to pay for a click. The actual CPC is the amount the advertiser ends up paying when a searcher clicks on his/her ad. These two amounts are typically different because the actual CPC is usually lower than the maximum CPC. Google explains it this way:

"If the advertiser immediately below you bids US$2.00, and if that advertiser's ad is the same quality as yours (and has equal-performing extensions and ad formats), you'd typically need to bid a penny more than US$2.00 to rank higher than that advertiser and still maintain your position and ad formats. With Google's paid search ads, that's the most you'll pay (about US$2.01), whether your bid is US$3.00, US$5.00, or more."

This auction model Google uses is called a second-price auction. In a second-price auction, the winner does not pay what he bid but instead pays a penny more than the bid of the next-highest bidder. In Google Ads, rather than paying the CPC amount he bid, an advertiser will pay one penny more than the bid of the advertiser below him. Google provides an additional explanation of the actual CPC.
Cost-per-Acquisition Bidding.
Search engines also allow advertisers to pay for each conversion rather than for each click. Under this bidding method, if the advertiser gets 100 clicks and 4 conversions, rather than pay for the 100 clicks, the advertiser only pays for the 4 conversions. (This method of advertising does not guarantee less expensive advertising. CPA bids have to be much higher than CPC bids or else the search engine will not display the ads.) To use this method of bidding, the advertiser must have sufficient history of conversions (Google recommends at least 30 conversions in the last 30 days), and conversion tracking must be set up. You can read more about this method of bidding here.

Clicks.
A click is counted each time a searcher clicks on an ad in the SERP. In the above example, this advertiser earned 2,101 clicks or 2,101 visits to his website.

CTR.
The click-through rate (CTR) is the percentage of impressions that yield a click. In the above example, the 60,367 impressions yielded 2,101 clicks for a CTR of 3.48%. CTRs vary by industry, advertiser, ad position, keyword, ad text, and so on. CTRs between 1% and 4% are typical, though an advertiser should be constantly striving to increase the CTR no matter what the current CTR is. A low CTR means few website visits, which in and of itself is bad, but a low CTR will also give the advertiser a low Quality Score, which will drive up the advertiser's cost per click.

CPC.
The average cost per click (CPC), in concert with the conversion rate, will determine whether the search ads are profitable. In the above example, this advertiser paid an average of $.83 per click. Because an advertiser's CPC is calculated dynamically, not all 2,101 clicks cost the advertiser the same amount. Some were more than $.83, whereas some were less, but on average, they cost $.83 each.

Cost.
This advertiser paid an average of $.83 per click for 2,101 clicks, yielding a total cost of $1,742 for this keyword. As the next section discusses, an advertiser can set a daily budget for search campaigns. The keywords that lead to the most sales with the best margins deserve the most budget. If an advertiser is reaching his daily budget each day, keywords with lower profit margins should not be allocated a budget that could go to better performing keywords.

Conversions.
Search advertisers should not lose sight of the fact that the purpose of a campaign is to generate conversions and profits, not just impressions and clicks. Each keyword an advertiser bids on should produce conversions. Multiply the profit per sale by the total number of conversions. That number should surpass the total cost to advertise on that keyword. If not, advertising on that keyword is not profitable. In the above example, the advertiser generated 140 conversions. If those 140 conversions generated an average of $20 profit each, he would have earned $2,800 in profit from $1,742 in ad spending, making this a profitable campaign. (Note that conversions are not tracked in Google Ads unless (1) goals were properly set up in the website's analytics package and (2) the analytics package was properly connected to the search advertising account.)

Conversion Rate.
The conversion rate is the percentage of visitors who convert. In the above example, 140 of the 2,101 visitors converted, leading to a conversion rate of 6.66%. Conversion rates vary by industry, but if an advertiser suspects his conversion rate is lower than the industry average, he should consider the following actions:
• Make sure the ad and landing page are relevant to the keyword being searched.
• Make sure the ad and landing have a compelling offer that is competitive with the other advertisers.
• Make sure keywords using broad match or modified broad match are not attracting traffic from irrelevant searches. If irrelevant searches are coming into the site, consider adding negative keywords or using more precise targeting.
・If the keyword is still not producing conversions after the above interventions, pause the keyword.

Return on Advertising Spend (ROAS).
Return on ad spend is not shown in the above report, but it is an important metric. The formula for calculating ROAS is:

(Total Profit/Total Adspend) x 100 = ROAS

Any company doing search engine advertising should monitor its ROAS. If it made $30,000 in profit selling cars, and it spent $10,000 on ads, the ROAS would be 300%. (Alternatively, ROAS can be calculated without multiplying by 100, which would give the number 3, meaning the advertiser earned $3 in profit for every $1 spent on advertising.)

Cost per Acquisition (CPA).
An acquisition is typically considered a completed sale for a retail site, or a lead generated for a lead generation site. CPA determines how much it costs, on average, to reach a site's acquisition goal. It is the total amount of money spent on advertising divided by the total number of sales completed (or leads generated). If an advertiser spends $100 and gets 10 sales, its CPA is $10. Search marketers often set a "target CPA," which is the amount they can spend on search ads for one conversion and remain profitable. If the CPA is so high that revenue from a sale is lower than the combined cost of producing and advertising a product, then search ads will not be profitable.

Search query Reports
During ad campaigns and keyword bidding, running a Search query Report shows all of the searches that resulted in an ad click. A wedding videographer in California ran an ad campaign and opened up his ads to Broad Match to expand his reach. Figure 6.5.2 displays the Search query report:

This report is valuable because it shows where Google chose to show an ad. As in this example, keywords will often show up that don't make sense, and an advertiser would rather not let Google show his ad on that keyword again. If a company's site is intended to generate leads for its wedding videography service, and the company doesn't sell "disposable wedding cameras," it may be wasting money paying for clicks from someone who searched for that unrelated phrase. The Search query Report is thus very useful for finding keywords that should be added to the negative keyword list.

Mobile Performance
Customer search behavior and purchase behavior often vary dramatically across devices, so it is important to break out all search reports separately for desktop, tablet, and mobile devices. In general, most digital marketers find that their ads perform worse on mobile devices than desktop computers (though this, of course, is not always the case). There may be many reasons for inferior performance on mobile devices, and the correct response to low-performing mobile ads depends on the reason for the low performance, so it is important to make a proper diagnosis.

Poor Mobile Optimization.
If your website is not optimized for mobile usage, you can expect a lower conversion rate from mobile users. Keep in mind that just because your website is mobile optimized from a technical standpoint does not mean it is necessarily optimized for mobile users. For example, even if all webpage content renders properly on a mobile device, the main call to action and purchase button may show up below the fold. This would decrease conversions despite the site being technically mobile optimized. Now that the majority of searches occur on mobile devices, all websites should ensure that the mobile experience of all webpages is optimal.

Low Purchase Intent.
People tend to do searches on whichever device is handy at the time they think of doing the search. Most of the time, a person's mobile device is the most handy, so people doing product research or searching for basic information often conduct such searches on mobile devices. Even an amazing mobile website is not likely to produce a high conversion rate from searchers who had no intention of making a purchase in the first place. In this case, it makes sense to adjust mobile bids to be lower so less money is wasted on mobile searchers who are unlikely to convert.

Alternate Mobile Conversion.
Mobile search ads may be profitable even when they yield a conversion rate of 0%. Someone who does product research through a search on their mobile phone may have no intention of converting on their phone. But if they find compelling information, they might later move to their desktop computer to complete a transaction, or they may locate the nearest store to make the purchase in person. In this case, it was the mobile search that led to the purchase, but the mobile search report would not show the profit that the mobile searches were producing. For a time it seemed that sophisticated tracking algorithms would be able to connect cross-device purchases and mobile-to-store.