Beyond "pass or fail": Mastering ad campaign optimization (not just testing)

The punchy headline. The perfectly-stated benefits. The clickable CTA.

As a brand manager, you're under serious pressure to create high-performing ads and hit your target metrics as quickly as possible.

Your A/B test will give you “definitive proof” on whether your ads are good ads or bad ads. But ads don’t just pass or fail. They should evolve.

Great ads, successful campaigns and loyal audiences go beyond the contrast of success or failure. 

That’s why it’s essential to ditch the binary. A culture of testing creates and builds a philosophy around  continuous improvement through ad optimization.

By focusing on optimization, you can switch the focus from creating winning ads as quickly as possible to refining your campaigns so that they continually perform better as you implement and learn. Unlike the gamble of one-off testing, optimization is a strategic, ongoing process that minimizes risk and maximizes ROI.

In this post, I’ll explore the key difference between ad testing and ad optimization and show you how to protect your confidence in the process. 

Your standard A/B test involves creating several different versions of an ad and seeing which performs best with your target audience — version A or B (or C, D or E). An A/B test can directly show you which variations have the biggest impact on your metrics. 

While an essential tool for understanding which types of copy, images, keywords and more resonate with and drive action from your audience, A/B tests can force you into a dichotomy: your ad either wins or loses. You can perfect your copy, apply all the best practices and use all the right keywords and still not “succeed.” 

The pressure to prove your hypothesis and deliver a winning ad can stir up anxiety and take a hit at your confidence. No matter how prepared you are, your ad can still “fail” — and everyone’s watching. 

That’s why many brand managers are moving away from ad testing and towards ad optimization. 

Ad optimization moves you away from narrow definitions of good and bad ads, winners and losers. It moves you towards a growth perspective — helping you to see the process of ad creation and optimization as a holistic, iterative process supported by data insights. Ad optimization gives you the space to experiment, play and explore different variations of your ads. 

With ad optimization, you’re focusing less on hitting the perfect ad as quickly as you can. Instead, you’re using data to continually improve, learn and adapt — refining your strategy and creative to best suit what your audience wants and needs from your brand. Optimization puts the focus on insight and delivering incremental gains via a long-term strategy. You’re here to learn. Not to win or fail. Not to prove your smarts. And not to deliver a huge metric spike on your first iteration. 

In what ways can you optimize your ads? Let’s take a look at some of the main ways below. 

Creative optimization

Visuals are a huge part of what makes an ad effective. The subtlest of changes can have a huge impact on the effectiveness of your ad. Take these stats: 

• 44% of people preferred warm-colored advertisements.

• 53% of marketers report that images are the most valuable ad element impacting their overall social media performance goals. 

• A colored border on a Facebook ad image can double an ad’s click-through rate. 

• Personalized CTAs are 202% more effective.

Ron Sel talks about how visuals can impact the overall feel of an ad: 

"Design elements such as typography, visual hierarchy, and color choices play significant roles in capturing attention and effectively communicating messages. 

For instance, sterling typography can increase readability and influence the emotional response of the audience. At the same time, a well-structured visual hierarchy ensures that the most important information catches the viewer’s eye first. 

You’ve got to consider color schemes too. They’re not just about aesthetics but psychology. Bright colors might capture attention, while muted tones convey luxury or calm.

Don’t overlook layout and spacing, either. A clean, well-organized ad makes it easier for your audience to absorb your message."

Here are the core elements to experiment with when it comes to visual creative optimization: 

1. Image and video variations: Video ads have a 7% lower click-through rate than ads with images alone.  While ad images that show people expressing positive emotions tend to perform the best. 

2. Color palettes: Color impacts up to 85% of consumers’ purchase decisions. Color and variations in color combinations can influence emotion and perception in several different ways. Polymath Johann Wolfgang von Goethe was the first to connect colors to specific emotions. He divided colors into two core categories: ‘plus’ and ‘minus’. Plus colors are those that create positive feelings in people such as yellow, red-yellow and yellow-red. Natural and pastel shades bring feelings of calm. In particular, the color green typically evokes feelings of calm. While ‘minus’ colors like blue, blue-red and red-blue evoke negative feelings.

3. Layouts: Many marketers recommend going for the F-pattern layout to mirror how users typically scroll: top left, across then down. Place your most important text and visual elements at the top left to make sure they get seen by viewers. 

I love this inventive use of color from Nora’s Breakup Pints by Rethink Agency: the cutting copy paired with pastel hues bring an element of surprise and delight.  

Copy is another essential part of an ad. You can test and optimize several different copy elements including: 

• Headlines: Experiment with benefits, humor, emotion, scarcity and urgency (e.g. limited number of seats or limited-time deals) and stats to find out which approaches are most persuasive for your audience. 

• Body copy: Most ads limit the amount of body copy you can share. Don’t waste your space. Lean into your most persuasive, easy-to-digest benefits, add social proof in the form of stats and short customer quotes and include your keywords. 

• The call-to-action: Recent research shows that first-person CTAs can perform up to 90% better. Experiment with first and second-person CTAs and different word choices to see what hits with your audience. 

You can also test your ad copy with consumers via user testing for: 

• Message clarity: How clear is this version of your messaging? Is your audience taking away what they need to take away and acting when they need to act?

• Persuasiveness: Which persuasive techniques and formulas can you experiment with and optimize? For example, you could experiment with both the PAS and the AIDA formulas to see which has the biggest impact on your metrics. PAS stands for pain, agitation and solution. You start with their pain and then poke it (exploring the ways it impacts them both practically and emotionally). You then provide your solution in the form of your product or service. The AIDA formula stands for attention, interest, desire and action. The first step is to provide an attention-grabbing hook like an interesting stat or open-ended question. The second stage is to capture interest by exploring your product or service’s core benefits. The final step is to use your copy to drive action: you may use tactics like scarcity (such as via a limited number of places or a time-sensitive special deal) coupled with your CTA. 

• Memorability: How easily can your audience recall your core messaging?  

• Credibility: 92% of people say they’d trust a recommendation from a peer. What performance stats, social proof, and measurable benefits can you share? 

Subtle changes can make a huge impact. Think: adding a different frame to your image, bringing a warmer tone to your color palette or swapping out your CTA for a call-to-value (that insinuates the benefit your prospect will get from clicking — “save more time,” vs “learn more.”). 

You can also optimize your ads by experimenting with different combinations. Switching out your images, text, headline and CTA combinations and testing them against each other to which ones resonate best with your audience.

Audience targeting optimization

Demographics tell you who your audience is on a basic level. Demographics to target include age, gender, socioeconomic background, race, ethnicity, income and education level. You can test demographics in isolation or layer them such as, “College-educated women aged 30-40.”

You can also target people based on their behaviors. The more detailed picture you have of your audience the better. Collect and centralize data from multiple sources including on-site and in-app behavior, purchase behavior and interactions with past ads and centralize your insights to make it easier to optimize your ads. 

You can also target your audience based on their interests. Maybe they share a love of dancing, reading or going to stand-up shows. Getting clear on your target audience’s interests is helpful for ad targeting, creative choices and ad placement optimization. 

Say you’re placing ads with Google Display Network, you can choose to show your ads in specific contexts relevant to your audience’s interests such as “outdoor lifestyle,” or “ice skating.” 

You can also reach more potential customers by focusing on lookalike audiences. Lookalike audiences are audiences that share similar characteristics, behaviors or interests to your most high-value segments — known as your seed audience. The more detailed your segments are, the more data you’ll have on prospective best-fit customers. 

You can try targeting prospects based on how similar they are to your seed audience:

• 1% lookalike audience: This highly targeted audience is the audience that is most similar to your seed audience. 

• 10% lookalike audience: This audience may share some of your seed audience’s characteristics, giving you greater reach. 

Amelia Crisp at Talking Stick Digital recommends experimenting with different percentages to find your sweet spot:  

“Experiment with different lookalike audience percentages to find the sweet spot that balances reach and relevance for your specific campaign goals. We recommend testing with the 2% first, these people are very similar to your desired audience, however, is a little broader to ensure that you are showing your ads to a decent amount of people without facing any limitations.”

You can also test different retargeting methods, targeting leads who’ve previously engaged with your ads whether through views or clicks. Most platforms including Facebook, Instagram and Google offer retargeting options, making it easy to track and retarget people who’ve viewed or previously engaged when your ad. 

To retarget strategically, segment and target your audience based on the types and level of interaction they’ve had with your ads. For example, you could segment your audience based on their dwell time or how long they spend watching your videos.  

Placement optimization 

The first part of platform optimization is platform selection. 

To choose your platforms, take a look at which channels you currently get the most engagement on and review third-party data on where your audience hangs out. Once you have this essential information, you can begin to experiment with and compare performance on different channels. 

Social media, search engines and display networks are all places you can run your ads. Here are the most recent stats for the most popular ad platforms:

• Instagram: Instagram ads reach 96% of users every month. It boasts a 93% brand awareness in the U.S. 50.6% of users are men. 

• Facebook: In 2025, Facebook is the third most-visited website, ranking just behind Google and YouTube. Interestingly, Facebook (60%) comes in second by one percentage point to Instagram (61%) as the most popular social media platform for discovering a new product. The platform is still a frequent for millennials with 69% using it over other social media platforms.

• TikTok: 67% of TikTok users say TikTok inspires them to shop even when they’re not planning to. While TikTok has a diverse audience, it continues to skew younger. Right now, 36.2% of users are aged 25–34, while 33.9% of users are ages 35–44. 

• Google Ads: Google is a great place to position ads, no matter your audience. Paid search ads show your products at the top of Google’s search results, making them ideal for both viability and improving your click-through rates. 65% of people click on Google ads from Google search when they’re looking to buy something. While Google Shopping ads make up 76.4% of all retail search ad spend. Because of this, they generate 85.3% of all clicks on AdWords campaigns. 

• Website ads: You can also place your ads on both your site and on other third-party sites across the web. Say you own a beer brand, it makes sense to place your ad on sites like online grocery retailers and food and drink news sites. Context and audience targeting matters. Choose third-party sites that match your audience’s interests and product categories. 

You should also optimize ad placement within platforms, tracking ad clicks and views and optimizing based on where your ads see the most impact. Take Facebook, where you have the option to run ads on the newsfeed, via messenger or through Instagram. 

When creating your ads, it’s also important to optimize for viewability. Where you place your ad has a big impact on who sees it and how easy it is to read. The best ad placements can help reduce ad blindness and increase your reach. Take website ads, Brock Munro at Publift says: “Generally, horizontal ads on the top or bottom of the page perform best for mobile, while vertical ad placements on the left or right of the screen perform best on desktop.”

Bid and budget optimization

It’s also important to optimize your bids and budget. To optimize your ads based on performance, track your data to see which locations, days, times and devices perform best and then increase or decrease bids accordingly. 

Adjust your strategy as you get more insights from your data and allocate more of your budget to your highest-performing campaigns. 

Landing page optimization

No matter how well crafted your ad is, it won’t perform well with a poorly-planned landing page. 

Nothing will make your visitors bounce faster than a page that has zero relevancy to your ad. As such, it’s essential to make sure your landing page is relevant to the ad. 

Mirror the design and copy choices you made for your ad on your landing page. Any benefits you wrote about, keywords you used, offers you shared and promises you made need to be mirrored on your page. 

Also make sure your visitors see the most relevant, important messages first. 

Ensuring your messages match their stage of awareness (moving them through awareness to consideration to action) is key for moving them along the sales funnel. To improve conversion rates on your landing page:

1. Use social proof: This helps to persuade and convince visitors of your benefits and claims.

2. Experiment with one call to action: Copywriting best practices suggest that a single, first-person CTA performs best. Multiple calls to action can lead to analysis paralysis, reducing the likelihood that visitors will click.

3. Adapt as you go: In line with our ethos of data equals learning and growth rather than signifying success or failure, Bella Chandler at Workshop Digital says: “Creating a landing page is not a one-time task—it’s a process that requires ongoing refinement and adaptation. Iterative testing is the cornerstone of this process, allowing you to continuously optimize your page based on real user behavior. The study "Developing a New Model for Conversion Rate Optimization” by Pim Soonsawad underscores the importance of iterative testing, showing that ongoing adjustments based on usability and user interaction data are key to improving conversion rates over time. By regularly conducting tests on various elements of your landing page—such as headlines, CTAs, images, and layout—you can gather valuable data on what resonates most with your audience.” 

While an ethos of, “test, test, test,” can undermine confidence and put the focus on quick wins rather than learning and allowing a holistic understanding of your audience to evolve, A/B testing can still be a useful tool within your wider optimization strategy.

It can be essential for gathering data and insights that help you understand your audience and how best to target them. 

Without this data, you’ll effectively rely on guess work, generalized stats and third-party research that may not be relevant to your audience. If you can see A/B tests as just one tool in your toolkit to help  you understand your audience, creative and campaigns rather than relying on it to help you identify and categorize "winners,” and “losers,” then it can be an important tool to for giving you the data insights you need to create great ads.

Let’s take a look at how you can use A/B testing effectively:

1. Focus on testing specific, measurable variables. The first step is to isolate your variables. Which variable or set of variables will you test for your A test and your variation (B)? For example, you could test a first vs second-person CTA, two images or two headline lengths.  

2. Set clear hypotheses and track results. One of the main ways brand managers go wrong with A/B testing is failing to create a clear hypothesis. You need to know exactly what you want to measure and how you’ll do so before you test to guide your decisions going forward. An example of a clear, measurable hypothesis is: “Changing the length of the headline from five to seven words will increase click-through rates by 25%.”

3. Use insights to inform ongoing optimization efforts. Take your winning tests and implement them. Continue to test them against other ideas. The more insights you get, the more you can feed them back into your current and future campaigns. 

Avoiding common pitfalls 

Small sample sizes can undermine the reliability of your results. Short testing periods often don’t give you enough time to measure how the changes you’ve made impact your KPIs over time. Longer testing periods also give you time to account for external factors such as seasonal traffic spikes or local news stories that impact user search habits. The ideal test period for validating your results is two to four weeks. 

Biased data is another potential issue you need to look out for.

Return user bias is one of the most common biases. Oliver Paton says: “Return user bias is the most common bias that I’ve seen in tests. This is typically caused by changes in configuration of an experiment where these changes are made incorrectly. If one day one of the tests the splits in traffic are 90:10 And then on day two of the test changes are made to the configuration so the split is then made two 50:50 90% of the return visitors from day one will be coming into the variation. 

Because return users are more likely to convert, you have created a returned user bias to the A group.” 

To avoid this bias, he recommends that you never make changes to traffic splits once your tests are already up and running. 

When you run your test can also bias your results. For example, your ads may naturally get more traffic at high traffic times as day such as the evening. In addition, seasonal shifts will also spike or drop your traffic. It’s important to take peak times and quiet times into account when you run your tests and review your data.

Another common bias is the selection effect, this is when a brand manager selects a sample that’s not representative of their audience. In A/B testing, this bias can undermine a test when you generate traffic from different sources. 

Nick Osborne from Marketing Experiments, featured in Instapage, says: “We had radically redesigned their subscription offer process for the electronic version and were in the middle of testing when they launched a new text link ad campaign from their main website to the electronic product. This changed the mix of traffic arriving at the subscription offer process from one where virtually all traffic was coming from paid search engines to one where much traffic was arriving from a link internal to their website (highly pre-qualified traffic).

The average conversion rate increased overnight from 0.26% to over 2%. Had we not been monitoring closely, we might have concluded that the new process had achieved a 600%+ conversion rate increase.”

To help ensure your A/B test is valid, Conversion Rate Optimizer Peep Laja, recommends the following:

1. Test length should be 3 weeks minimum, 4 weeks ideally.

2. Calculate your sample size beforehand using several tools.

3. Conversions should hit between 250-400 for each variation you test.

4. Statistical significance should be 95% minimum. 

Let’s take a look at how to optimize your campaigns: 

1. Define clear goals and KPIs 

First up, identify your metrics. What are you trying to achieve? Pick one measurable goal such as, “Increase conversions by 15% in four weeks by adding an image of our company founder to the landing page.”

Here are some common KPIs to choose from:

• Conversion rate: The percentage of users that act on your CTA and fulfil the desired action (such as making a purchase). 

• Click-through rate: The percentage of people who click on your ad.

• Cost per click: How much you pay for each ad click.

• Cost per acquisition: How much it costs to acquire a new customer. 

• Landing page conversion rate: The percentage of visitors that click through your ad and convert on your landing page. 

2. Gather and analyze data

Now it’s time to gather and analyze your data. Use analytics tools to track campaign performance. Each platform will have its own analytics capabilities whether that’s Google Analytics or Facebook. 

Look for patterns and trends in your data, across keywords, audience segments, ads, landing pages and campaigns and see where you can optimize. 

3. Implement changes and monitor results

Time to implement. Make incremental adjustments based on your data insights. It’s important to make changes slowly, closely monitoring your results. By making gradual changes, it’s easier to monitor performance and isolate your variables, helping you to more easily identify what is and isn’t working well for your audience.  

Continue to use these insights to refine your campaigns, seeing your data-driven insights as one aspect of learning to make your campaigns more effective. 

4. The iterative process

Optimization shouldn’t be treated as a one-off or something you do occasionally, it should be baked into your process. To continue to refine your campaigns, put an optimization loop in place. Think: data gathering, analysis, take action, repeat.

5. Document and share learnings 

Create a knowledge base of what works and what doesn't. Make this accessible to everyone on your team. 

Creating a knowledge base both sets the precedent for building a culture of growth and learning, rather than one that emphasizes success and failure and helps provide your team with the immediate insights they need to optimize ads and campaigns.

Ad campaign optimization perfectly balances the need to make data-driven decisions based on what your audience wants, while allowing for the creative freedom and confidence that comes with building a culture of growth over one focused on success vs failure. 

Optimization is a strategic, data-driven process that minimizes risk. It’s a long game of continual growth rather than an approach that focuses on quick wins like testing does. 

As a brand manager, embracing a culture of continuous improvement will help you build your confidence, allowing you to deliver your best work, hit your goals and build a successful brand. Start now and build a culture of growth and learning, rather than one that presents strategic decisions as a risk that marketers on your team either win or lose.