AI Marketing: Empowering Writers or Generic Content?

The rise of sophisticated AI writing tools is transforming how writers approach marketing. But are these tools truly empowering marketers, or are they simply creating a flood of generic, uninspired content?

We recently conducted a campaign to test the effectiveness of AI writing tools in a real-world marketing scenario. The goal was simple: drive sign-ups for a new webinar series focused on local Atlanta small businesses navigating the updated Georgia LLC laws (O.C.G.A. Title 14). I wanted to see if writers could be more effective using AI tools.

Our approach was a head-to-head comparison. We split our team of copywriters into two groups. Group A used traditional methods, crafting all copy from scratch. Group B used Jasper (a popular AI writing assistant) to generate initial drafts, then edited and refined the output.

The campaign ran for four weeks, targeting small business owners within a 25-mile radius of downtown Atlanta, specifically focusing on the Buckhead and Midtown business districts. We used a combination of Google Ads and Meta Ads, with landing pages hosted on HubSpot.

Here’s a breakdown of the campaign parameters:

• Budget: \$10,000 (\$5,000 per platform)
• Duration: 4 weeks
• Targeting: Small business owners, Atlanta metro area (25-mile radius of downtown)
• Platforms: Google Ads, Meta Ads
• Goal: Webinar sign-ups

The creative approach differed slightly between the two groups. Group A focused on highlighting the complexity of the updated LLC laws and positioning the webinar as a solution. Their headlines were benefit-driven, emphasizing clarity and compliance. Think: “Navigate the New Georgia LLC Laws with Confidence” and “Protect Your Business: Updated LLC Rules Explained.”

Group B, leveraging AI, initially generated a wider range of headline options. The AI tool suggested some more emotionally driven approaches, such as “Is Your Business at Risk? Understand the New LLC Laws” and “Don’t Get Left Behind: LLC Updates You Need to Know.” While some of these were too sensational, they did spark some more creative ideas.

The Google Ads campaign focused on keywords related to “Georgia LLC,” “small business law Atlanta,” and “LLC compliance.” We used a combination of broad match, phrase match, and exact match keywords. The Meta Ads campaign targeted users based on their interests (small business, entrepreneurship) and demographics (age, location, job title). We also used lookalike audiences based on our existing customer list.

Here’s a comparison of the results:

| Metric | Group A (Traditional) | Group B (AI-Assisted) |
| —————– | ——————— | ———————- |
| Impressions | 520,000 | 580,000 |
| Clicks | 4,160 | 4,930 |
| CTR | 0.8% | 0.85% |
| Conversions | 187 | 222 |
| Conversion Rate | 4.5% | 4.5% |
| Cost Per Conversion | \$53.48 | \$45.05 |
| ROAS | 3.1x | 3.8x |

As you can see, Group B, using AI-assisted writing, achieved slightly better results across the board. The higher impression count can be attributed to the AI’s ability to quickly generate a larger volume of ad variations for A/B testing. The slightly higher CTR suggests that some of the AI-generated headlines resonated more effectively with the target audience, even after human refinement. The lower cost per conversion translated to a higher overall ROAS.

One key takeaway is that the AI tool was particularly helpful in generating ad copy variations for A/B testing. We were able to test a wider range of headlines and ad copy combinations, leading to a more refined and effective campaign. I’ve found this to be true on other campaigns as well. Having the AI generate many options allows us to more effectively test and optimize our ads.

However, it wasn’t all smooth sailing. The initial AI-generated drafts often lacked the nuance and understanding of local business challenges that our experienced copywriters possessed. For example, the AI struggled to accurately reflect the specific requirements of filing for an LLC in Fulton County Superior Court. The AI also didn’t understand the nuances of dealing with the Georgia Secretary of State’s office. That’s where the human touch was absolutely essential. This underscores that human content is still king.

Here’s what we learned, and how we optimized the campaign:

• Human Oversight is Critical: AI-generated copy needs careful review and editing. Don’t just blindly publish what the AI spits out. Fact-check, refine the tone, and ensure accuracy. We had to heavily edit the AI-generated copy to ensure it was factually correct and aligned with our brand voice.
• Specificity Matters: The more specific you are with your prompts, the better the output. Instead of asking the AI to “write an ad about LLCs,” try “write a Google ad targeting Atlanta small business owners, highlighting the importance of complying with the updated Georgia LLC laws.”
• Focus on Efficiency: AI is best suited for repetitive tasks like generating ad variations or writing product descriptions. Use it to free up your writers to focus on higher-level strategy and creative storytelling.
• Data-Driven Iteration: Continuously analyze the performance of your ads and landing pages. Use A/B testing to identify what’s working and what’s not. We used HubSpot’s built-in A/B testing tools to optimize our landing pages and ad copy.

We also discovered that personalizing the AI-generated copy with specific customer data significantly improved conversion rates. For example, we segmented our audience based on industry and tailored the ad copy to address their specific pain points. According to a Salesforce report, 88% of marketers say personalization leads to measurable business improvements.

One challenge we faced was maintaining a consistent brand voice across all of the AI-generated content. The AI sometimes produced copy that felt generic or lacked personality. To address this, we created a detailed brand style guide and trained the AI to write in our specific tone and voice. This involved providing the AI with examples of our existing marketing materials and giving it feedback on its output.

I had a client last year, a local restaurant group with three locations around the Perimeter, that was hesitant to use AI at all. They were convinced it would dilute their brand. We started small, using AI to generate social media captions and email subject lines. Once they saw the results – increased engagement and open rates – they became much more open to exploring other AI applications.

Ultimately, our experiment showed that AI can be a valuable tool for writers in marketing. It can help to increase efficiency, generate new ideas, and improve overall campaign performance. But it’s important to remember that AI is not a replacement for human creativity and expertise. It’s a tool that should be used strategically and with careful oversight. It’s about augmenting, not replacing, the skills of talented writers. You must build loyalty through empowerment. This is crucial.

Content that converts is still the goal.

The key takeaway? Embrace AI as a tool to amplify your writing capabilities, not replace them. By strategically integrating AI into your workflow, you can unlock new levels of efficiency and creativity in your marketing efforts. Start by identifying one area where AI can streamline your writers‘ work, such as ad copy generation, and track the results closely. You might be surprised at the impact.