Accelerated Keyword and Engine Testing, Part 1

Keyword and engine expansion are fact of search engine marketing (SEM) life. Most marketers start campaigns with the obvious keywords and search engines. The good news is keywords that were obvious to the marketer were probably obvious to prospects and customers, too. Most search volume in an industry category is in those no-brainer keywords.

Yet keyword expansion is an ongoing process. Adding more engines to the marketing mix can incrementally increase revenue and profit. Adding a second-tier engine or an additional 5,000 or 45,000 keywords to a campaign won’t suddenly bestow you with enough power to kill your competition. It’s still a good idea to revisit expansion from time to time.

Success is never guaranteed. New keywords or search engines may not meet your success objectives (ROI or otherwise). In paid-placement SEM, you can limit the costs of a bad media decision (poor keyword; bad engine; or poor ad copy or landing page) because the auction allows near-immediate control. Media and testing costs exist in SEM as well.

Direct marketers know a certain percentage of a media budget must be allocated each month to test new media or new creative. These marketers also know 30 to 70 percent of all new media tested fails. Testing new ad creative also has a failure rate.

Each new media buy is a gamble, an educated guess the media may work. A budget allocated for trying new things is necessary to grow a campaign and a business. All your current keywords have bids that are inching up as the marketplace gets more aggressive and competitors become more skilled at running their businesses. I prefer to think of failed tests as learning experiences.

Every test you conduct with a new batch of keywords or a new engine is an experiment that yields extremely valuable data, even if the test fails. Can you expect a good return on investment (ROI) from a budget allocated for testing? No.

Though it’s possible new keywords and engines will succeed right away, that would be extremely unusual if you had tested aggressively enough. At least some media tests should be groundbreaking. They should try something completely new or something old in a completely new way. When a test finds a successful new way to use keywords, engines, or creative, the investment pays dramatic dividends from that point forward. This is why smart marketers always leave a percentage of their budget for testing.

Why You Must Bear Some Costs

You’re testing a new keyword in Google with an existing landing page. To eliminate a keyword/engine combination as a viable part of a campaign moving forward, you must give the keyword/engine a chance to perform.

Should a new keyword be tested at the top (high CPCs (define)) or at a lower position at lower cost? There’s no one answer, but there’s a big benefit to starting near the top when testing new keywords, engines, or creative.

Higher positions get more click volume, so you can test more quickly. Placing ads lower on the page (at a lower CPC) increases the chances of hitting your ROI target (due to a lower cost), but there may be a dramatic decrease in click volume at lower positions.

Factors that should influence your decision to test aggressively (high CPC and position) or conservatively (lower CPC and position):

• Based on similar existing keywords with similar creative landing pages and ad copy, what’s your conversion rate to positive behaviors?
• Given that conversion rate, could you conceivably afford the top positions?
• Can the keyword or engine deliver a huge volume of clicks even at a lower position? Some keywords are so heavily searched, they deliver volume at lower positions. It’s unusual for those keywords to be new to your campaign; however, you may be testing new creative against those keywords.
• If the test is successful, how much positive outcome can the new keywords, creative, or engines have on your business?
• How quickly do you want the test results?

Deciding how aggressively to position a campaign for testing purposes isn’t easy. Starting high answers conversion and ROI questions quickly. Starting low may dramatically extend the test period.

In part two: Proven strategies to help effectively plan and execute keyword expansions.

Meet Kevin at Search Engine Strategies in Toronto, Canada, May 4-5, 2005.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.