Text Ad Optimization Q&A #3: What Factors Have The Greatest Influence in Testing?

This week, we’re celebrating the release of Text Ad Zoom with a 5 part Q&A series featuring the authors who wrote the articles highlighted in The Ultimate List of PPC Ad Testing Resources. Each day this week, we’ll have a different question and the answers. Previous, we asked:

    1. What are the biggest text ad testing mistakes?
    2. How do pick which ads to test first?

      Read below for today’s question and answers (in no particular order). Want to see Text Ad Zoom and all of the other great ClickEquations features in action? Request a demo.

      Text Ad Optimization Q&A #3: What Factors Have The Greatest Influence in Testing?

      Brad Geddes: The headline and display URL. I find that a strong headline can compensate for a weak description line 1. However, a strong description line 1 will not overcome a poor headline. I also think display URL is not tested enough. People like to know where they are going after the click, and the display URL tells the searcher where they will end up after the click. In fact, the instant previews that just rolled out for ads also shows that Google believes in this as well. Andrew Goodman: Fit and pain points. If you’re a particular kind of roof repair company then speed of response addresses your buyer’s concerns; same if you are overnighting fresh fish or meats. If you’re a store specializing in large shoe sizes, then selection and a return policy may be the key. Overall, you often just want to “drive it down the middle of the fairway,” so to speak, with obvious messages and minor adjustments. It’s as much staying away from any off-putting verbiage or symbolism, as it is convincing people of anything in that small space. Ad position is arguably “the greatest” influence in testing, so the fact that you have the budget to reach premium position (combined with attention to CTR’s and your other campaign elements, aimed at higher Quality Score) can’t be divorced from the ad testing exercise. If you’re habitually down in positions 6-8 you may see very different testing dynamics than what you see in 1-2. Jessica Niver: Adding time-limited offers/sales, pricing in ads, offering a free anything of value (brochure, tool, etc), prominently branding ad texts for non-branded ad groups (this has worked well for non-branded queries for better known brands). Chad Summerhill: Search query/ ad text alignment, calls to action, benefits, differentiating from the other competitor ads. Amy Hoffman: Consistency among keywords and landing pages.   Make sure that all of the ads tested are relevant to the audience, the landing page, and every search query that could trigger the ad.   Ads in question should deliver users to the same landing page, or the results could be skewed by landing page quality.  Delivering users to a quality landing page with a high conversion rate will most likely give the most useful data as opposed to delivering users to a landing page with a low conversion rate.  If you don’t have a high quality landing page, you won’t know if users aren’t converting because the landing page stinks or if it was because of the ad text. Erin Sellnow: Ads with offers in them always do the best – get a free guide, save 20%, etc. as the user can quickly see what they gain by clicking. Don’t over-promise, as that will have a negative impact when it comes to conversions or time on site, but be sure to say why your page/website is the best. Pete Hall: Landing pages. You can do so much with ads, but really ads exist to drive clicks. To convert those clicks you need to dial in the landing pages. Your ads need to compel users to click and then set the user expectation for the landing page. This is even more important now that previews are implemented for AdWords ads in Google SERPs. Ryan Healy: If there is a secret to writing winning PPC ads, I’d have to say it’s clarity. In case after case, the winning ads I see use clearer language than the losing ads. So if there’s a way to clarify your ad — and eliminate any kind of awkward or confusing words and phrases — then this is definitely something you want to test first. Tom Demers: The easy answers here in terms of the actual components of the ad are “the headline” and/or “the image” depending on the ad platform/format. That’s largely true, but in analyzing thousands of ad tests one of the biggest initial surprises for me was the impact relatively small changes like tweaking a single verb, including symbols, including (or excluding) price, etc. Facebook ads are a great example: prevailing best practice wisdom with Facebook ads is that “it’s all about the image,” and the image is important, but we’ve run tests where we changed nothing but the copy of the ad and the percentage differences across large sets of ads were well into double digits. For me the moral of the story is that different components of an ad aren’t the key factors – things like:

      • Being relevant or disruptive (depending on the context)
      • Thinking your way into the mind of the searcher
      • Having a lot experience seeing what works and doesn’t work in similar situations

      Are the key factors in setting up tests that show large improvements. Crosby Grant:The creativity and savvy of the advertiser makes a big difference.  Creativity for trying new things, savvy for paying attention to the competition, the market, the seasons and holidays, etc.  But maybe a more literal answer would be about the ads themselves.  Certainly including the user’s search term in the Headline is a big bandied about a bit.  More important though is to get at what they actually searched for, rather than the literal term they searched for.  As an example, if someone searches for “I need divorce help”, a headline of “Divorce Attorney” is perfectly appropriate, and may perform better than “I need divorce help” as a Headline.  Having a strong call to action in the descriptions cannot be emphasized strongly enough.  It simply makes a big difference.  If your call to action is turning people away then quite often that is actually a good thing; they most likely were not going to take that action anyways.  Once less click for you to pay for.  Your domain name has a big influence too.  While it is often fairly set in stone, a good bit of advice for any would-be-entrepreneur would be to test out different domain names prior to getting locked in.  Just spend a few grand on PPC with identical ads pointing to different domains and pick the one that sticks.  (If you have been around since before 2006 or so, you may recall when you could use any display URL you pleased – it did not have to match up to your domain.  Those days are long gone.) Rob Boyd: The largest factor is the setup. This would include creating a goal and developing/organizing your testing variables. If you don’t have a goal for testing how will you know what is successful? If you don’t develop and organize your testing variables properly, how will you know what action impacted the results? This is the easy stuff so you should be getting it right. We only have a certain degree of control in the outcome of a test so if your foundation isn’t proper you aren’t giving yourself an opportunity to succeed. The second largest factor in my opinion is the ad copy, be it title, body, or display URL. Plain and simple, you need effective ad copy for a successful ad. Greg Meyers: I think many advertisers put too much attention on the Text Ad, when they should first be reviewing the relevancy of the Search Terms that are driving the Text Ad. That is why I believe that the Adgroup and the search terms chosen for the Test have the biggest influence on success. Once that has been achieved, then testing different CTAs, Offers, Prices, becomes effective. Bonnie Schwartz: Headline is very important.  I see the biggest impact through refining my ad groups and getting my kw in the headline.  Oftentimes, it may have less to do with the messaging in your ad but whether or not you have the keyword in bold in your ad a few times. John Lee: For search ads, the biggest factors are the actual offer, the benefits touted in the ad and the call-to-action. Swapping these factors in and out can provide interesting, and hopefully positive results. For display ads it is more about how well are you distracting the user away from the website they are viewing. How crazy-distracting can you make your text ad without abandoning PPC ad writing fundamentals or crossing the line in terms of good taste? Jon Rognerud: Uppercase/lowercase, periods, different characters, numbers and call to action statements. Learn More About The Authors

      Acquisio Admin

      Acquisio Admin

      The First Machine Learning Marketing Platform
      Built to Scale Search for Local Resellers & Agencies

      Automate, optimize and track more campaigns, more profitably.