Skip to main content

Download the Marketing with Purpose playbook to get actionable tactics to start your inclusive advertising journey today

Microsoft Advertising Blog
Page 1 of results

The Art and Science of Ad Copy Testing: Follow-up Q&A

The Art and Science of Ad Copy Testing: Follow-up Q&A

There are very few things we can do to actively influence our search audience. We don’t have music, dancing ponies or massive graphics; all we have is our ad copy. What we can do is influence our audience with smart, effective ad copy. In this webcast, part of the Art and Science series, we first take a little time to understand the psychology of decision-making. Why do we make the choices we make? Are we more influenced by relief from a pain point or by a rub on our ego? Once we understand that we’re driven by our most basic Reptilian Brain, we can clearly see how to use this to motivate searchers.

In this webcast we focus on specific angles for ad copy to influence searchers, from pain points to lively verbs to FOMO and offering partner products. We wrap it up with more specifics – on brainstorming, keeping the process fun and how to go crazy (in a good way). We’re joined in this webcast by Purna Virji, Microsoft Senior PPC Training Manager, Luke Alley, PPC Director at Avalaunch Media and Erin Sagin, PPC Evangelist at WordStream.

Today we’re answering the questions we got during the webcast, but didn’t have time to answer. If you have new questions that aren’t answered here, ask in the comments below or take it social with #AskBingAds.

 Watch the webcast here

Register for the next webcast here

Q. Does using a few different emotions confuse the message to your customer?

A. It can; it can also make your message stronger when done well. For example, “Get rid of the gray and look younger” starts with a pain point (gray hair!) and finishes with ego (look younger!). It’s best not to shove too many different tactics into one ad in order to keep your copy clear. Experiment with the different emotions and test which gets the best response.

Q. These tips sound great for consumer goods, but how do I use an emotional appeal for a B2B product?

A. This is where it’s critical to think from the viewpoint of your target audience. Even B2B customers have pain points, have ego and fear involved in their buying decisions and will respond to a compare/contrast story.

Q. Would you want to be subtler or blatant in your appeals? For example, “You need these dresses to look good” versus “You will look amazing in these dresses.”

A. Test both and see what resonates with your searchers! The first example hits a pain point (“I don’t look good”) while the second hits ego (“I will look amazing”). It’s hard to predict which will appeal more to your searcher, which is why ad copy testing is such fun.

Q. Should I include sale prices in my ad copy?

A.  There are a few different considerations – think about price vs. sale price. If you have a sale, you will want to let people know about it. If it’s just price in general, include the price in the ad to qualify the click – this means that people who are clicking on the ad have likely already decided the price is ok, which turns them into a probable conversion. Whether it’s a luxury good or a discount good, like a cheap hotel, the more you can qualify that offering, the better you can do with the clicks.

Q. Is there a way to automate a countdown (“Only 4 hours left in the sale!”) in copy or is that manual?

A. Yes there is. During the webcast we posted a “How to Countdown for Ad Copy in Editor” booklet. To access the webcast console where this is stored, go back to your webcast on-demand link or register here. (You have to register to get into the console.)

Q. Can you automate bid increases with the countdown ads?

A. Yes you can, and this is an excellent strategy.

Q. How accurate do you think demographic targeting is?

A. Our demographic data is as accurate as the signals we pull from. This means the more signals we have, the more robust our picture of the searcher is. For Bing, the number of signals is pretty significant because we can add other Microsoft data sources, such as Windows 10 users and opted-in Internet Explorer browser sessions. All told, we’re pretty confident that when a Bing Ads advertiser makes a demographic targeting selection, we’re delivering ads to that specific audience.

Q. How much on average does a business need to spend in clicks per month to get effective action from traffic?

A. This depends on the business; if it’s in an incredibly competitive market (say, selfie sticks), it will cost more because other selfie stick sellers are bidding on the same keywords, driving up the price of a click. If the business is for walking tours of Columbus, Ohio, you’ll see very little competition on those relevant keywords which means you’ll get more impressions and clicks on a smaller budget.

The big idea with your cost-per-click budget is to get the maximum number of clicks for the least amount of spend. Smart, on-target ad copy helps achieve this, which means you can spend less per click, show up lower on the search results page, and still get clicks because your ad copy attracts searchers.

Q. How would you apply these principles to ads for non-profits – for example, an educational program about anti-bullying?

A. Who is your target audience? What is their goal? What is preventing them from achieving this goal? How does your educational program help them? What are the other anti-bullying educational programs saying? What’s unique about your educational program? Does this program have special certification or testimonials that make it better than others? This is how you apply these principles. You start by asking these questions (and more) and learning from the answers.

Q. Are there any trends in Google users versus Bing users in terms of psychographics?

A. We can’t speak to the psychographics of the Google user, but we can tell you all this awesome stuff about the Bing user.

Q. Do you ever find that click-thru rate (CTR) isn’t the best measurement? That you could get great CTR but fewer conversions and higher bounce because the ads are actually misleading? Whereas more direct messaging might be more successful?

A. There’s a lot to unpack in this question. First, no matter what your ad copy strategy, we recommend against misleading. You can use the principles in this webcast all day long and never be misleading. Sometimes clever ad copy will attract clicks but not conversions – this is the argument for testing ad copy until you land on the messaging that gives you a high CTR as well as a high conversion rate.

Also, a much more valuable metric than CTR is conversions per impression. This is ultimately what will impact your return on ad spend (ROAS).

Q. How many ads should I test?

A. We have two answers for you, as we’re not completely certain what this question is asking: How many ads should you test as a general practice, or how many ads should you test in each adgroup using these new ad copy tips?

First answer: All of them. You should test all of your ads every time. Start with an existing ad that has proved to be successful; use this as your benchmark. Then start running your new ad copy against that benchmark ad in structured waves, so you only ever have one new ad in test against the benchmark. Depending on your volume of clicks, you might run the test for two weeks or a month. As you find new ads that perform better than the benchmark ad, rotate those in as the new benchmark.

Second answer: It depends on the amount of clicks/data you’re able to compile in the adgroup. High volume adgroups may deserve to have 3 ads testing against each other. For most other adgroups, you’ll want to just do a simple 2-ad test so you can compile enough data to make a decision sooner. Desktop and mobile should be tested independently of each other so in essence you will double the amount of ads tested in each adgroup.

Thank you to our attendees for submitting questions. We appreciate your engagement and your interest in sharpening your ad copy brain.

Watch the webcast here

Register for the next webcast here