4 expert tips for using generative AI to maximize native ad effectiveness

By Kim Feeney, Product Marketing Manager



Open Mic article

This content is produced by a publishing partner of Open Mic.

Open Mic is the self-publishing platform for the marketing industry, allowing members to publish news, opinion and insights on thedrum.com.

Find out more

April 16, 2024 | 6 min read

Generative AI is a powerful tool for maximizing native ad effectiveness and consumer engagement, says Taboola's product marketing manager, Kim Feeney. She sits down with Mark Coalson-Kiner, co-founder of Nexio by Xevio, to find out how his team of native advertising experts use generative AI to create and enhance their campaigns.

The advertising landscape is fiercely competitive. Attention spans are shrinking, and the pressure to drive consumers to action is immense. Creating engaging content is the cornerstone of any successful native advertising campaign, but it's a constant struggle. How do you craft fresh campaign ideas that stand out from the clutter, resonate with your audience, and drive conversions?

Generative AI tools, now widely available (75% of marketers say their organization is already using or considering using AI tools), can be a big part of the answer to this content creation conundrum. They can supercharge your workflow, churning out a stream of tailor-made headline and image variations in a fraction of the time and allowing you to test a broader range of messaging and creative approaches. But AI tools aren't a magic bullet. You need to know how to use them effectively.

We spoke with Mark Coalson-Kiner, co-founder of the native ad agency Nexio by Xevio, where the utilization of generative AI tools is widespread. Here, the agency shares its best practices for leveraging these tools to enhance native ad campaigns successfully.

1. AI is a tool, not a replacement

First, the elephant in every room: AI tools are just that – tools. Advertisers should not replace their creative team with AI tools.

At Nexio, AI tools are in the hands of the creative team. Nexio believes, 'It's important that the people prompting it are also those who've been writing copy, creating images, and designing for a while. Their experience allows them to know what is most likely to work.'

Think of generative AI tools as junior-level employees who, when given a detailed creative brief (a prompt), can do the heavy lifting and instantly come up with tons of ideas; however, just as a manager reviews their team's work, an experienced human needs to review the AI output and select the best options that should move to the next stage — refinement.

2. Know the limitations of AI

At this point, AI output can be formulaic and dry. It is, after all, a machine, and native speakers can often spot AI-generated content. Mark says he never expects AI output to be 100% spot on. Instead, he recommends using it as an idea generator or refining the content output to about 70% right. Then, a great copywriter can take it the rest of the way, replacing anything that sounds inauthentic with more natural language.

His team swears by the ‘rephrase’ tool within Taboola's GenAI Ad Maker. By highlighting a campaign title and clicking the ‘rephrase’ button, the copywriters can review an endless list of alternative phrasing to find the options with the most potential.

Mark underscored the importance of abiding by ad platform policies. He noted that a significant advantage of Taboola's GenAI Ad Maker is that all of the platform's content policies are built-in, so the team never has to worry, and the approval process is seamless. However, if creative teams use external tools to create ad content, Mark warns that they need to be hyper-vigilant when reviewing the content or risk having campaign items rejected if banned words were used mistakenly, for example.

3. The secret sauce of a prompt

Mastering prompts is one of the most challenging aspects of using AI tools. Single-sentence prompts don't elicit the best results from AI tools or produce relevant content that reflects a brand's USPs. Generic messages also don't resonate with audiences.

Since these tools can ingest significant amounts of information, Mark advises including as much detail as possible in your prompt. At a minimum, for landing pages, campaign titles, and descriptions, tell the tool who to act like, your target audience, and what you want it to do. For example:

You are a native ads expert who works in the healthcare industry. Create a landing page focused on benefits 'A,' 'B,' and 'C.' The target audience for the landing page is women ages 25-50.

For headlines specifically, Mark recommends adding examples – either a few fresh options or some from existing high-performing campaigns. Then, ask the tool to generate similar options, give it a different tone, adjust the length, etc.

When creating images from scratch, Mark suggests including the phrase 'create a photorealistic image of X'; otherwise, the results can look obviously AI-generated.

Be prepared to adjust and refine your prompts by adding more specific details. If the new content doesn't work, instruct the AI to go back to the previous version and try including different information. Don't get frustrated by the process; getting to that 70% benchmark often takes multiple iterations, and you’ll get a better feel for what works as you get more experience with prompting.

4. Get creative with images

One area where Mark has been able to maximize the boundless capabilities of AI is in image generation, where the technology can produce images that would be nearly impossible to create or, at the very least, cost-prohibitive for most brands.

While working on a campaign for a solar panel company, Nexio’s creative team had the idea to use an image of the panels on fire, something he would never suggest doing in real life just for the photo. Similarly, for a shoe brand, they rendered realistic images of the product in the desert, in the water, and again, on fire, testing these different backgrounds for the best performing formula.

An AI-generated image of two shoes on sand
Example of an AI-generated image

With AI's ability to change backgrounds, every brand can instantly create holiday-specific images without props and studios. They can alter locations without travel expenses or make culturally appropriate changes for campaigns in different regions. As long as you can describe what you want, the possibilities are limitless, and the cost is virtually non-existent.

Transform your ads with generative AI

As generative AI tools continue to emerge and evolve, brands concerned about their bottom line can learn to harness these tools to remain competitive. They are incredible for banishing creative blocks, coming up with new landing page directions, and using data-backed insights to optimize elements and drive performance at scale.

Simultaneously, generative AI tools produce fresh campaigns that combat user fatigue while eliminating countless hours of tedious work for creative teams, freeing them up to work on high-value tasks such as optimizing your media mix and brand strategy. That's a win-win for brands and consumers.


Industry insights

View all
Add your own content +