A/B testing in AdWords campaigns allows advertisers to continuously improve conversion rates on their PPC campaigns. Since accurately predicting user behavior is virtually impossible, campaigns seldom come close to maximum productivity when they are launched. However, with adept PPC split testing, more effective ads and landing pages are constructed as the campaign progresses.
Ad Text Split Testing
Split testing in AdWords or Bing Ads can focus on a great number of variables. Among the most important variables to test:
- Assurance. "100% guaranteed", "lifetime warranty", "no-hassle returns" and similar statements can produce dramatic changes in clickthroughs.
- Call-to-Action. Calls-to-action such as "Order now!" can encourage clickthroughs, but not always. By testing, it becomes unnecessary to rely on assumptions about user behavior and instead use actual data.
- Credibility Elements. Which will persuade more users to click, "BBB A+ Rating", "Since 1975" or something else? Testing is the only way to find out. Social proof — for instance, "Over 10,000 customers" — is often highly effective.
- Display URLs. Adding a subfolder to a domain, using a subdomain, or changing the punctuation of the URL can affect clickthroughs.
- Offers. Test, for example, "25% off" versus "$20 off". Other offer-related message variables include timing ("October only"), availability ("Quantities limited") and exclusivity ("New customers only").
- Questions versus Statements. Testing, for example, "Order now" versus "Are you ready to order?" can yield surprising results.
- Stating the Price. Testing will determine whether saying, for example, "Only $99" helps or hinders conversions.
Landing Page Split Testing
For AdWords, A/B testing of landing pages is just as important, if not more important, than testing ads. If the landing page is deficient, increasing clickthroughs will serve little purpose, because users will never convert. Landing page text, imagery and offers should always be relevant to the ad; whenever user expectations are not met, the landing page will fail. Many elements of a landing page must be tested. They include:
- Body Content. Typographical elements such as fonts and the use of bulleted lists can have a big impact on conversions. The content itself should also be tested for instance, does a conversational, humorous tone draw more conversions than a more formal, businesslike tone? A law firm might assume the latter approach is better, but by testing, the firm can be sure.
- Call-to-Action. Testing CTAs can be content-focused or design-focused. Changing the offer itself can obviously make a huge difference in conversions — but so can a change in button size, shape, color, or position. Also, advertisers should keep in mind that what seems like a subtle change can have a great impact on users, so no CTA element is too small to test over the span of a long PPC campaign. Also, consider testing the number of CTAs on the page. Too many options can confuse users and lead them to do nothing; on the other hand, having one or two secondary offers may pull in "tire kickers" who are not interested in making a commitment yet.
- Credibility Elements. Certain credibility elements carry more weight with users than others; "BBB A+ Rating", "Since 1975", and "Over 10,000 customers" all have merit, but just throwing them all on the landing page could be counterproductive, confusing users rather than reassuring them. In some cases, pages without credibility elements outperform pages with them — a great example of why testing, rather than intuition, should drive PPC campaigns.
- Headlines and Subheads. Headlines and subheads are read far more frequently than landing page body content. Thus, getting them right is very important. Appeal to different customer needs and problems. Try questions instead of statements. Create a sense of urgency. Use humor. These are a few ways headlines and subheads can be styled for testing purposes.
- Imagery. Stock photography usually goes over like a lead balloon on landing pages, but even the first attempt with well-executed, customized imagery may not be persuasive. Product images highlighting certain features may convert better than ones highlighting others; simple infographics describing a complex service may convert better than a collage of employee photos.
- Long Copy versus Short Copy. Theoretically, long copy is effective for complicated products and services, whereas short copy works better for ones that are easily understood. However, because there are many exceptions, testing is essential.
- Page Layout. Where text and imagery appear on the landing page influences conversions. Best practices exist for placing page elements — such as putting the phone number at the top of the page, on the right — but as always, users can defy expectations. Advertisers should also consider whether a responsive design is necessary. These days, the answer is probably yes, because so many people now use mobile devices for Internet access. Using a non-responsive layout could significantly reduce the number of potential leads and sales.
Methodical A/B testing in AdWords and Bing Ads is a way for advertisers to gain a considerable advantage over competitors that lack the patience, budget and expertise to do it or do it properly. These three items are required for an effective testing process: patience, because it can take months for test-driven campaign changes to reach a critical mass; budget, because testing requires time from PPC strategists, copywriters and Web developers; and expertise, because best practices for conversion rate optimization (CRO), user experience (UX), and the mechanics of Google AdWords, Bing Ads and third-party testing software are all complex in the extreme. Partnering with a PPC management agency adept at testing may be one of the most important factors in determining whether a campaign will succeed or fail.
Straight North PPC Marketing Services
Improve your PPC campaign today! Give us a call at 855-779-7675 or request a quote online.