Some time back, search engine marketing (SEM) included optimizing your website for high ranking in search engine results. This process is now called search engine optimization, or SEO, and I address it in a separate presentation. Now, SEM generally refers to paid advertising in search engines.

Search Engine Advertising Can Be a Highly Efficient, Easy-to-Measure Tool

SEM allows you to target the exact prospects that are best for your business. It’s not rocket science. You can create and place your own search engine ads. Setting up your first simple ad may take just a few minutes. You simply go to the Internet site where you want to advertise and follow their instructions and guidelines.

But It’s No Silver Bullet

However, to get really profitable results, you will more likely need to take time—just like in every other form of advertising—including testing, redoing, and testing again. I cannot emphasize this enough. It is also possible that search engine advertising will not work for you.

Search engine advertising is generally purchased on a price-per-click basis, or PPC. You pay every time someone clicks on your ad. Because your ad is only served up when someone does a search for your keywords, or at least something close to your keywords (you can usually specify which), you increase the chances of advertising to someone who is highly likely to be interested in your product or service.

One way to get going is to see what your competition is doing, and then ask yourself the following:

  • What phrases or words are they purchasing?
  • What headlines and ad copy are they using?
  • What does their landing page look like?
  • What copy do they use on their landing page?
  • What sales points do they emphasize?
  • What special offers do they use to try to close the sale?

However, I wouldn’t get too carried away with what my competition is doing. Instead, I would tend to think very carefully about each step myself, testing as much as possible. I would also think through how my business strategy and product or service attributes, particularly the consumer benefits, are different than those of my competitions. And if I have developed a unique selling proposition—the marketing phrase that really highlights in a succinct and memorable way my product differentiation—I would test including that in my ad copy or at least on the landing page.

Search Engine Advertising Is Perfect for Testing Because It Is So Measurable

Test carefully, test broadly, and carefully track your results. I would test the following, noting that to really test something you should change only one variable at a time:

  • Test different search phrases or words. Generally, you want the least competitive—and hence the least expensive but most profitable—words that get people to click. The more specific the wording, the more likely the target audience will be buyers and not simply browsers. So think which words a serious potential buyer of your product will be most likely to enter.
  • Test different search engines.
  • Test frequency.
  • Test timing—maybe ads perform better in summer, when they are less expensive, or maybe they perform better at the holiday time despite being more expensive.
  • Test headlines. Headlines should basically grab attention of your target audience.
  • Test copy. Copy should be geared toward getting the target audience to your landing page, not necessarily trying to “sell” them. The landing page of your website should do that.
  • Test landing pages. It is very easy to get carried away with testing search phrases and words too much but not testing your landing pages enough.
  • Test any pages after the landing pages.
  • Test the offer. On the Internet especially, you need to consider an offer to close the sale right now. Remember, your competition is just one click away.
  • Test different closing methods. Maybe you want to close the sale completely online. Maybe you would do better to capture an email address and then follow up. Or maybe you would do better with giving people a phone number and having them call you.

Try to measure separately the cost per inquiry and the cost per sale. Generally, people on an Internet search engine are comparative shoppers and will likely be considering and weighing multiple products or service providers. Thus, the true cost per sale, though harder to measure than the cost per inquiry, is much more important. Some keywords (often the more general ones) and some ad copy might be much more powerful at generating traffic; other keywords (usually more specific ones) might be much more likely to generate sales.

When you measure the cost of customer acquisition, ask yourself whether you are generating future, continued, repeat purchases. Consider, too, whether your satisfied customers generate word-of-mouth sales that lead to further sales with additional customers.

One of the traps you need to be careful not to fall into—and it is just about impossible to measure—is whether the target customer would have come to your website anyway, even if you hadn’t advertised. For example, buying the keyword for your own company name is likely to have very high click-through and appear to be highly profitable, but it could be that virtually all the people would have checked out your website anyway and that buying the keyword was actually a big waste of money, even though it tests well by simply measuring sales per click. This is a much more serious issue if your company is already very well known to potential customers, but all companies need to think about this issue that taints measuring on a strictly cost-per-sale cost-per-click basis. The big way to avoid this trap is to make sure you have a big, fat margin of safety in your payback from search engine ads.

Over time, refresh creative copy. Give some keywords a rest. Keep testing to find less competitive and hence less expensive keywords with good results.

You may want to test what is called dynamic search ads. These are ads that use search phrases and words developed by a search engine that continually crawls your website and updates your ad words.

As search engines get more sophisticated, there are additional parameters you may want to specify in your ads. For example, a geographical specification such as a particular city or town, or within 10 miles of a specific address, may be a very powerful tool for a local facing business.

Unless a user specifies, search engines generally use the input phrase to broadly match what the consumer is looking for. So, if the user’s phrase is “red toy boats,” the search results may also include “red boats.” Thus, for your ads, you may want to specify results that only include your phrase exactly.

You could also negatively define your search parameters. For example, you could specify that searches for “red toy boats” would see your ad, but not offer results for searches that include the word “motorized.”

As you can see, developing a profitable search engine advertising campaign is probably going to take a lot of thought and a lot of testing and a lot of tinkering. It could be highly profitable for your business, but like any advertising, it is not a slam-dunk.

Takeaways You Can Use

  • Testing is cheap with SEM, so test everything!
  • Research what the competition is doing.
  • Cost per sale is the best measure.