The power of testing: How ad copy insights drive better campaigns

Whether you’re with a startup trying to scale, or a brand that’s a household name worldwide, launching a new ad campaign is a gamble if you don’t do your research. One of the best ways to make launching less risky is by testing your concepts first. Once you have a concept, getting even more granular by testing your advertising copy and creative elements makes your chances of success even better. 

Testing advertising copy drives measurable results and reduces media waste, which translates to better use of resources, improved brand reputation, and ultimately, increased profits. 

In this post, I’ll discuss: 

• Types of ad copy testing

• Methods of ad copy testing

• What to test

• How to optimize ad copy using testing insights

• Examples from CPG and QSR

• Avoiding common pitfalls

• Tools for ad copy testing

• The future of ad copy testing

The world of advertising is littered with examples of advertising that missed the mark, misunderstood the cultural moment or didn’t speak to the intended audience. Testing ad copy is critical, both to avoid becoming an example of what not to do, and to stand out in an oversaturated, short-attention landscape. 

Advertisers have been working on the best ways to test ads for a long time. To better understand, PACT, which stands for “positioning advertising copy testing,” was introduced by 21 of the largest ad agencies in 1982. At that time, the idea was to improve the research standards used to create and test ads so that the agencies could develop better ads for their clients. At that point, the advertising options were TV, radio and print. Compare that to now, when brands have many more options, including: 

• Social media

• Influencer partnerships

• Podcast ads

• Digital ads (including SEM, banner ads, PPC and more)

• In-app ads

• Television  (including streaming services)

With so many choices, making data-driven decisions where possible can narrow the uncertainty and increase the chances of success. Ad copy testing allows you to compare how different versions of the same ad affect your audience, what consumers are looking for and gives you insights into how your customers think as they make purchasing decisions. 

You have almost as many choices about how to test your ads as you have options for distributing them. 

The type of ad you’re testing also matters—testing banner ads is quite different from testing ads that will play on television. 

The second crucial factor when deciding what kind of test will yield the most reliable results is a clear objective. Is the ad supposed to convince people to try a new product, or simply to increase brand awareness? 

Let’s get into some of the different types.

A/B testing vs multivariate testing 

Two common types of ad copy testing are A/B testing and multivariate testing. Both can be very useful in testing versions of things like email subject lines, landing pages or calls to action (CTAs). The main difference between the two types of tests is the number of elements that are different. 

An A/B test is pretty straightforward: Does version A work better or does version B? Multivariate testing, as the name suggests, involves multiple variables. For example, if you have two email subject lines an A/B test can help you decide which one to use. But what if you have two versions of a landing page, each with multiple unique elements? An A/B test could still identify the one that works best, but wouldn’t help you understand whether it was the graphics, font or wording that made a difference. 

Here’s an example of an ad where an A/B test could be used to reveal which image works best: 

And here are two versions of Airbnb landing pages designed to convince people to be hosts. Virtually every element of the designs are different, including the colors, format, images and so on. A multivariate test is the better option for choosing between these two. 

PPC ad copy testing vs brand-level ad testing 

Pay-per-click (PPC) ads have the very specific goal of getting a consumer to click. Brand-level ads are more nebulous, and the goal is to influence how a segment of people feels about a brand. As you might imagine, there are different kinds of tests for these two different types of ads. Again, the objective of the ad guides the best ways to test it. If the goal is to get a consumer to click, the approach is significantly different than if the goal is to change how someone thinks of your brand. 

“As digital marketing budgets rise and new channels emerge (from podcasts to social media), brands face the mounting challenge of proving not just how many clicks their ads get, but how their ads change brand perception.”

- Gautam Gupta, MarketStar

Static vs dynamic ad formats

Consider the difference between an ad on the side of a city bus and a social media ad shown to you because you searched for something like “personalized stationary.” 

The ad on the bus remains the same for weeks, and you probably see it regularly, along with everyone else who sees the bus. This type of ad is a static ad. More commonly online or on social media, these ads are typically ones that don’t involve video or animation.

Ads that are personalized or change based on your search, though, is a dynamic ad. That means it changes based on your particular details—your browsing habits, preferences, location and other demographic details. It might show an image of stationary with your name on it. The idea behind a dynamic ad is personalization, showing you the right message at the right time to motivate you to take some action. 

The two formats are used in different ways, and which type or combination works best, depends on your audience, campaign goals, budget and platform. Testing can help you narrow those options and find what’s most likely to work.

In addition to knowing the objective of your ad, you need to have a clear goal for the test itself when you evaluate the available methods for testing. For example, you may be looking for more quantitative data to run a PPC campaign or, in a brand lift campaign, you may need qualitative data. In some ways, the type of data you need depends on whether you’re conducting a performance marketing campaign, such as PPC, social media, search engine optimization (SEO) or a brand marketing campaign which includes consistency across all channels, building customer trust and relationships. Very generally, performance marketing requires quantitative data and brand marketing requires qualitative data. 

Monadic, sequential monadic, eye tracking and scroll maps provide quantitative data. This kind of data is foundational for performance marketing, which includes things like PPC, search engine marketing (SEM) and paid search. 

Surveys with open ended questions, one-on-one interviews, consumer panels and focus groups are a few of the most common ways to gather qualitative data. These methods offer insights on why people respond in the ways they do. 

Market research tools and methodologies

A/B testing and multivariate testing are ways to evaluate single or multiple elements of your ads, and monadic and sequential monadic tests do the same, but with more context. 

Monadic example: A clothing company might say, “We are introducing a new style of jacket,” then the audience would see an image of the jacket. After that the audience responds to a couple of questions, such as “How likely are you to buy this jacket?”

Sequential monadic example: Your audience sees at least two concepts, perhaps a new jacket and a new style of jeans, or even a jacket, style of jeans and a new bag. Regardless of the number of concepts or which ones the audience member sees, they are asked the same set of follow up questions.

Once you know why you’re testing, and you’ve selected the type of testing more likely to give you useful information, you need to figure out exactly what to test. Here are a few ideas: 

• Headlines

• CTAs

• Emotional vs rational appeal

• Visual hierarchy

• Language

• Tone

• Value proposition

• Offer framing

• Creative combinations (text+image+format)

Every element of your ad affects how people interpret it. Colors, branding, the exact wording, how the idea is framed, where the different elements appear, whether you’re appealing to your audience’s emotions or logic—it all matters and can be tested. 

Often, relatively simple A/B tests can provide a starting point, but more complex methods, like multivariate testing can zero in on which combinations work best.

All of the testing in the world won’t help unless the results are translated into creative decision-making. Not using the insights from tests to guide the creative direction of your ad or campaign is like ordering a beautiful meal, but leaving without eating. 

One of the early decisions when it comes to ad testing is identifying performance thresholds. When do you have enough information to proceed? Data about what’s likely to work best is critical to limiting media waste and driving success, but it is possible to reach a stage of analysis paralysis. 

That said, even the earliest sketches or idea of an ad can be tested with early stage testing to make sure you’re on the right track with your audience from the start. 

In order to get the most out of each test your team runs, apply the results across other marketing collateral. For example, if you test a landing page and discover a particular phrase performs especially well, consider using it in email and paid ads too.

“We ran a lot of surveys. We started with cheap surveys, any survey we could get that were not friends and family.”

- Gyve Safavi, SURI

In an interview with Shopify, Gyve Safavi and Mark Rushmore for SURI, described how they prepared to launch their product. SURI makes sustainable toothbrushes, and remarkably reached $30 million in sales in two years in a market largely dominated by two huge, well-known brands. They attribute some of their success to market research and testing. 

After conducting surveys and interviews, the SURI team began A/B testing. “We were doing fake ads to emails to just see what worked and how to refine the concept further,” says Gyve. He adds that even when the product was locked in, testing the ads helped them communicate better.

Haleon

Haleon, maker of the pain reliever Advil, needed insights into how their ads were performing in the US, according to Krista Kempski, Director, Consumer Insights. She says the brand wanted to “determine how to optimize our digital advertising and increase market share.” Their goal was to better understand how the format, creative elements and audiences performed. 

Some of the insights testing revealed that video ads worked much better than animated display ads, product-focused ads were clearly more appealing than brand-focused ads, and younger people responded more favorably than older audiences. Haleon used those insights to plan their next campaign—which features digital video and product-focused messaging. 

PepsiCo

Leading brands put millions of dollars at risk with big campaigns, and PepsiCo is a master of producing Super Bowl ads, one of the most expensive and widely-viewed advertising venues in the world. One of the keys to PepsiCo’s success with Super Bowl ads is their iterative approach to development, where they test, learn and optimize concepts before choosing a concept. 

Through this process, PepsiCo can invest in those pricey Super Bowl commercials with confidence. The work has paid off, too. As Stephan Gans, Chief Insights and Analytics Officer at PepsiCo shared: 

“Since partnering with Zappi, we have seen our creative effectiveness improve by almost a third across all our advertising.” 

Ad copy testing is a critical piece of developing a creative message that hits all the right notes, but teams make some common mistakes. Choosing the right test to gather the most useful information, balancing quantitative data and qualitative data and deciding when to stop testing are a few pitfalls I’ve already mentioned. 

Sample size errors and testing too many variables at once are also common errors to watch out for.

Modern technology offers an array of tools for testing, and working with a company that provides guidance along with good tools makes an enormous difference. Zappi offers an ad testing survey platform that leads to more effective digital advertising. Zappi helps you optimize your ads, make better ads, create ads that build your brand and continually iterate and improve. 

Zappi’s platform is also customizable so you can develop a process for optimization that suits your needs and get fast, flexible analysis in order to apply what you learn quickly. Learn more about what Zappi can do for your brand advertising.

Combining the already-powerful tools with AI-assisted ad testing and predictive creative scoring promise that ad copy testing will become increasingly valuable and allow creative teams to improve their advertising more accurately and effectively as time goes on. These cutting-edge tools offer advantages like real-time feedback loops and even adaptive ad copy. 

One area to watch as new technologies and techniques emerge is the regulatory landscape. Emerging laws and trends will certainly impact the technologies used for ad testing. According to Forbes, eight new laws being enacted in 2025 “will primarily focus on enhancing individuals’ control over their personal data.” 

Too much is at stake to launch an ad or a campaign without testing first. Careful planning regarding the types of tests, the analysis of results, when it’s time to stop testing and move forward and how the insights can be deployed is necessary to make sure your team optimizes your efforts and avoids pitfalls. Zappi is here to provide guidance, tools and resources to help.