The Internet is a vast maze of content, products and entertainment—you name it, and it has a home online.
Now that just about every business entity has at least one place where their business “resides” online, the challenges of getting found by your target audience are needle in haystack-like stats.
Enter search engine marketing, and more specifically: paid search. According to IAB, Internet marketers spent $42.8 billion on digital advertising in 2014, of which paid search accounted for 50 percent, or $24.6 billion.
And understandably so—ninety-three percent of online experiences begin with a search engine. In order to get the most out of your paid search tactics, A/B testing is an essential tactic.
Related Infographic: 10 Interview Questions to Ask a Paid Search Director
The ABC’s of A/B Testing
According to Business.com’s Paid Search Manager Hoa Nguyen, businesses that don’t A/B test their paid search ads run the risk of losing their competitive edge when contending for valuable eyeballs.
A/B testing for paid search is when you simultaneously test two versions of your ad (typically with varying copy) in order to see which performs best. Once there is a “winner,” you run that ad with the remaining budget in order to maximize your spend and results.
Creating an A/B Test for Paid Search
In order to create an effective A/B test, you you’ll need to work with the following elements:
The nature of the test and the elements that you’re testing will largely depend on your goals. If it’s PPC you’re testing, for example, KissMetrics shares that you’ll want to test one or more of the following four things:
- The headline
- The body text
- The link
- The keywords the ad displays for
They recommend that the headline contain a maximum of 4-5 words, and always include the search keyword. The body text gives you room to expand upon what it is your business does or is offering.
The link is one of the most important elements, as this is where the conversion is (hopefully) happening. It is typically best to link to a landing or product page, depending on your goals.
Keywords play a key role in testing as well. It may seem simple and straightforward, but at Business.com, we deal with millions of keywords for paid search. These keywords are organized into ad groups and campaigns, and we test multiple copies for each keyword and ad group.
The clicks and conversions you receive largely depend on the phrases and keywords you use, so testing is the only way to know what will work best.
Related Article: 10 Best Practices for Paid Search (Part 1)
The Importance of Analysis & Reporting
After testing, the work is far from over. Your strategy down the road will depend on results from the past, so it’s important to analyze and report on your testing. After the testing and results are complete, the cycle continues. Another hypothesis should be formed and additional tests should be run.
Key performance indicators may vary based on the goals of each test. They can range from page engagement, time on site, bounce rate, ROI, click-through-rate, conversion rate and a variety of other factors. It is important to identify which KPIs are most important in your test so the outcome is measurable and significant.
Many types of software exist to not only help you create tests, but track and analyze as well. Many data management platforms offer this type of service and there are companies who focus primarily on A/B testing. Some of which include:
- Adobe Analytics
- Inside Vault
- Marin Software
Best Practices for A/B Testing
Before launching any SEM campaign, it's absolutely vital to that you identify the following:
- The purpose of your campaign: Are you looking for traffic volume, page views, site engagement, registrations or conversions?
- The metric(s) that will be used to gauge success. Most often they are revenue, ROI, gross profit, CPA, CPM and/or cost per registration.
- The numerical goal associated with your performance metric(s).
Kissmetrics offers the following best practices during testing:
- Test your ad variations simultaneously to minimize time-based factors that might skew your results.
- Only test one thing at a time so you can pinpoint what is affecting the success of your ads.
- Test early and test often.
- Pay attention to the hard data you collect, rather than your “gut instinct”.
- Make sure you let your tests run for long enough to collect enough results to get accurate results. You need at least a few thousand impressions in order to get any kind of accurate data. Remember, with PPC, you’re only paying when people click, not for impressions.