Wavelength AI: 1.8x ROAS for PR in 2026

In the competitive area of public relations, securing meaningful media coverage demands more than just sending out mass emails. It requires precision and personalization. Our recent campaign, using advanced AI for journalist outreach, aimed to demonstrate how a targeted approach, powered by tools like Wavelength, could significantly improve engagement and placement rates. This detailed teardown will explore the strategies, successes, and lessons learned from our six-week initiative. Can AI truly transform how we connect with the press?

Campaign Overview: The “Future of Work” Initiative

Our client, a Series B SaaS company specializing in asynchronous collaboration tools, wanted to position itself as a thought leader in the evolving future of work narrative. The objective was clear: secure features and interviews in top-tier business and technology publications to drive brand awareness and, indirectly, lead generation. We allocated a budget of $25,000 for a six-week sprint, commencing in late Q1 2026. This budget covered AI tool subscriptions, a dedicated PR specialist’s time for oversight and content refinement, and minor content creation costs for supporting assets.

The primary challenge was cutting through the noise. Journalists are inundated with pitches, and generic outreach often ends up in the trash. We needed to prove that a data-driven, hyper-personalized approach could yield superior results. Our key performance indicators (KPIs) included the number of earned media placements, total impressions generated, click-through rates (CTR) to client assets from published articles, and in the end, the cost per lead (CPL) attributed to this earned media.

Strategy: Data-Driven Personalization with Wavelength

Our strategy hinged on using AI to understand journalist preferences and tailor pitches with unprecedented accuracy. We integrated Wavelength, an AI platform designed for media intelligence and outreach, into our workflow. The platform’s core capabilities included:

• Advanced Media Monitoring: Tracking millions of articles daily to identify trending topics, journalist beats, and publication editorial calendars.
• Journalist Profiling: Analyzing a journalist’s past articles, social media activity, and professional affiliations to build a detailed interest profile. This went beyond simple beat matching, digging into nuances like preferred data sources, interview styles, and even the tone they typically adopt.
• Pitch Generation & Refinement: AI-assisted drafting of pitch emails, subject lines, and even suggested angles based on the journalist’s profile and the client’s news.
• Sentiment Analysis: Evaluating past coverage of similar companies or topics to identify potential areas of interest or skepticism among specific reporters.

The initial phase involved feeding Wavelength our client’s existing press kit, recent company announcements, and a complete list of target publications. We aimed for publications like Forbes, Fast Company, TechCrunch, and industry-specific outlets such as HR Dive. The platform then ingested this data, cross-referencing it with its vast database of journalist profiles. This step took approximately three days, during which we refined our core messaging points for the “Future of Work” narrative.

Creative Approach: Beyond the Press Release

We understood that a standard press release would not suffice. Our creative approach focused on providing journalists with unique angles and valuable assets. Instead of simply announcing a new feature, we framed it within a broader industry trend. For example, when pitching our client’s new asynchronous meeting integration, we didn’t lead with “New Feature Launch.” Instead, Wavelength helped us identify journalists who had recently covered topics like “meeting fatigue” or “hybrid work productivity challenges.” The AI suggested a pitch angle: “Solving the Virtual Meeting Overload: A Data-Backed Approach.”

Each pitch included:

• A personalized opening referencing a recent article by the journalist, demonstrating genuine familiarity with their work.
• A clear, concise hook explaining the relevance of our client’s story to their beat.
• Specific data points from our client’s internal research on remote work productivity.
• An offer for an exclusive interview with the CEO or a product demo tailored to the journalist’s interests.
• Relevant multimedia assets, such as infographics or short video explainers, hosted on a dedicated AI product media kit page.

One particularly effective creative element, suggested by Wavelength’s analysis of journalist consumption habits, was to include a “quote-ready” snippet from our CEO in the initial pitch. This allowed busy reporters to quickly grasp our key message and potentially use it directly, saving them time. We saw a noticeable uptick in initial positive responses from journalists when this element was included, often reducing the back-and-forth required to secure a quote.

Targeting: Micro-Segmentation for Maximum Impact

Traditional PR often segments media lists broadly (e.g., “tech reporters,” “business writers”). Our campaign pushed for micro-segmentation. Wavelength allowed us to create highly granular lists based on specific keywords, sentiment, and even the types of sources a journalist typically cited. For instance, instead of just targeting “HR tech reporters,” we created segments like:

• Journalists covering “employee well-being in remote settings”
• Reporters focused on “AI’s impact on team collaboration”
• Writers exploring “the four-day work week and productivity metrics”

This granular targeting meant smaller, more focused outreach batches. Instead of sending 300 identical emails, we sent 50 highly customized emails to a segment of 50 journalists, each with a unique angle derived from Wavelength’s profiling. This approach inherently reduced the volume of outreach but drastically increased its quality. We maintained a core list of approximately 700 journalists across various tiers, but outreach was always executed in these smaller, targeted waves.

What Worked: Precision and Personalization

The campaign yielded several positive outcomes directly attributable to our AI-driven approach. Our overall pitch open rate averaged 45%, significantly higher than the industry benchmark of 20-25% for unsolicited pitches, according to a 2025 HubSpot report on PR outreach effectiveness. The response rate, defined as any positive reply (request for more info, interview interest), hovered around 18%. This is where the personalization truly shone.

We secured 12 earned media placements in our target publications. These included features in Business Insider, Forbes Council (contributed article by CEO), and a segment on a popular tech podcast. The total estimated impressions across these placements exceeded 4.5 million. One placement in TechCrunch, specifically, drove an immediate spike in website traffic, resulting in 350 direct sign-ups for our client’s free trial within 48 hours of publication.

The cost per lead (CPL) directly attributable to earned media was calculated at approximately $71.43 (350 leads from the TechCrunch article / $25,000 budget, acknowledging that not all leads came from direct clicks but also brand awareness). Our overall ROAS for the campaign was 1.8x, a strong indicator that the investment in AI tools and specialized outreach generated tangible business value beyond just brand mentions. The campaign’s eMarketer-reported average CTR from earned media links is typically lower, around 0.5% to 1.5%, but our campaign saw an average CTR of 2.1% on links embedded in published articles, suggesting higher reader engagement with the content.

What Didn’t Work: Over-Reliance and Initial Setup

While successful, the campaign wasn’t without its challenges. The initial setup and training phase for Wavelength took longer than anticipated, about a week longer than the projected three days. This was partly due to the complexity of integrating our client’s diverse content assets and ensuring the AI accurately understood the nuances of their product. There’s a tendency, I’ve observed across many teams, to treat AI as a magic bullet. It’s not. It still requires significant human oversight and refinement, especially in the early stages.

Another learning point was the risk of “over-personalization” or rather, personalization that felt robotic. In a few instances, AI-generated pitch openings felt slightly off, referencing an article in a way that seemed too generic despite the specific mention. For example, a pitch started with “I saw your recent piece on hybrid work trends, which was insightful.” While factually correct, it lacked the human touch of elaborating why it was insightful. We quickly instituted a mandatory human review of all AI-generated pitch drafts before sending, focusing specifically on refining the opening two sentences and the call to action.

We also found that some journalists, particularly those with very niche beats, were less responsive to AI-assisted pitches if the core story wasn’t a perfect fit. No amount of personalization can overcome a fundamentally misaligned story. This underscored the importance of our own strategic judgment in selecting target journalists, even with the AI’s recommendations.

Optimization Steps Taken

Based on our findings, we implemented several optimizations mid-campaign:

1. Human Touchpoint Check: We added a mandatory human review stage for all AI-generated pitch drafts. This involved a PR specialist spending 5-10 minutes per pitch to refine tone, ensure genuine interest was conveyed, and add any specific human insights that the AI might miss. This included checking for any awkward phrasing or overly formal language.
2. A/B Testing Subject Lines: We continuously A/B tested different subject line variations, with Wavelength providing predictive analytics on which lines were likely to perform better based on historical data for each journalist. For example, “Data Reveals [Client] Solves Meeting Fatigue” versus “Exclusive: New Research on Asynchronous Work.” The latter, with its exclusivity angle, consistently outperformed the former by 10% in open rates.
3. Refined Data Input: We regularly updated Wavelength with feedback on which pitches were successful and which were not. This iterative learning process helped the AI improve its future recommendations. We also fed it transcripts of successful interviews to help it understand conversational nuances.
4. Tiered Outreach Strategy: For Tier 1 journalists (top-tier publications, high influence), we adopted a “hybrid” approach. AI would still generate the initial draft and research, but the final pitch was almost entirely rewritten by a senior PR professional to ensure maximum impact and a bespoke feel. For Tier 2 and 3, the AI-assisted drafts with minor human edits proved sufficient.

These adjustments, particularly the human review and continuous A/B testing, led to a 7% increase in positive responses in the latter half of the campaign compared to the first three weeks. It’s a subtle but critical point: AI augments, it does not replace, the strategic thinking of an experienced PR professional.

Conclusion

Our “Future of Work” campaign demonstrated that integrating advanced AI tools like Wavelength into journalist outreach can yield superior results, driving higher engagement and more impactful media placements. The key lies not in outsourcing the entire process to AI, but in using its analytical power to enhance personalization and efficiency, allowing PR professionals to focus on strategic storytelling and relationship building. Future campaigns should build on these insights, continuously refining the human-AI collaboration for even greater precision.