How Sahan Journal is Revolutionizing Sales Pitches with AI
In July 2023, the American Journalism Project launched its to explore the smart application of generative AI and other technology within local news. This Studio, made possible by OpenAI with additional support from the Patrick J. McGovern Foundation, is sharing regular updates of its learnings. Here, we examine insights from an experiment within our cohort to innovate sales operations.
In the evolving field of local journalism, innovation isn’t just about storytelling — it’s about business models, growth and community support. The team at , a nonprofit newsroom serving Minnesota’s immigrant communities and communities of color, is using AI to reimagine their approach to advertising and sponsorships, and within weeks of use, their tools have generated successful bids and expedited their process.
The Sales Challenge
The idea to use AI in sales surfaced after Cynthia Tu, data reporter and AI specialist, and Michael Tortorello, chief growth officer, saved hours using for an on charter school contracts. Historically, Sally Waterman, director of advertising and sponsorship, spent hours preparing and customizing proposals for potential advertisers or sponsors. This involved manually hunting for relevant articles, compiling audience data and creating compelling presentations.
“I always want to show potential clients how Sahan can deliver what they need,” Waterman said. “But because tailoring proposals took so much time, it limited the number of sales pitches I could make in a week or a month.”
The AI Solution: Two Specialized GPTs
The first thing Tu and Torotello did was update the media kit template. This gave them insight into Waterman’s needs. They then iterated on ideas to streamline access to the information Waterman used in the media kit. Since Waterman had not been a regular AI user, they wanted it to be simple, effective and integrated into their current workflows.
After multiple attempts and after several months, Tu partnered with Liam Andrew, technology lead at the American Journalism Project, to build a solution: two custom GPTs that worked in tandem to produce the data Waterman needed through simple prompts:
- Audience Insights GPT: This tool draws from audience reports to generate detailed information about segments of Sahan’s readership, providing potential advertisers with precise, data-driven reasons to place ads or sponsorships on their website.
- Article Discovery GPT: A second tool quickly identifies and pulls the most recent and impactful stories that demonstrate the value of Sahan’s journalism to the local community.
For the Sahan team, the ability to quickly and accurately draw from their audience behavior and demographics data set — thanks to the Audience Insights tool — represents a time-saving achievement and immediate value-add. And by pairing this new workflow with the Article Discovery tool, they can easily identify Sahan stories that support and complement those audience data points.
With help from the AI team, Waterman produced a one-page pitch for Metro Transit, entitled “Get on the Bus.” The Audience Insights tool helped Sahan glean relevant insights to share with the prospective client, such as:
- Sahan readers are more than three times more likely to engage with issues on improving transportation and mass transit issues.
- Ten percent of Sahan readers only use public transit — that’s two times more than the average online adult in Minnesota.
Alongside those data points, Sahan Journal included a recent Sahan article about efforts to boost safety on public transit. Using the Article Discovery tool, Tu said, “we are able to quickly extract three articles that talk about public transit and transportation within the Twin Cities and show the advertiser that what we’re doing fits with [their] interests.”
Early Results: Promising Success
Within a few weeks of implementation, the AI-powered media kit has already shown significant positive results. Out of five recorded uses, three resulted in successful ad placements or sponsorship deals; Waterman expects to hear responses this spring from the other two prospective clients.
Tu asked Waterman if she’d be willing to provide feedback in the form of a simple, post-use survey, and she agreed.
Tu said questions in the survey included, “Did you use, or reference the AI media kit, in your sales pitch?” as well as, “Did a client comment? Anything relevant to add?” and, “Did I find that useful?” At Waterman’s encouragement, the Sahan team also added a question to record the actual sales numbers from successful pitches.
“We want to see if it’s generating significant revenue,” Tu said. “And we also want to use this for future revisions of the product.”
Beyond Efficiency: Building Trust
In early testing with the advertising team, Tu and Tortorello spent an hour using the GPT to put together information for a sales pitch Waterman needed to present that afternoon.
“Once Sally saw how fast and high quality the results were, she said, ‘Oh, this is really cool,’” Tu recalled, “She didn’t just want to use it more. She wanted us to teach her exactly how to get the most out of the tool, so she could own the process moving forward.”
Our take: What makes Sahan Journal’s efforts promising is their transparent, strategic approach to AI integration. They’re not focused solely on efficiency, but on a thoughtful partnership to leverage AI’s benefits across their team verticals.
Next steps
As local newsrooms continue to seek sustainable revenue models, Sahan Journal’s experiment offers a glimpse into how AI can be a collaborative partner in journalism — not a replacement for human insight, but a powerful assistant. And with each incremental win, their team grows new confidence and curiosity about ways to use AI for greater capacity. After seeing the two custom GPTs in action, Sahan’s development director said she hopes to use the tools to enhance her communications with members, donors and other supporters.
The project demonstrates that with consistent effort and thoughtful implementation, AI can help mission-driven newsrooms do what they do best: connect communities, tell important stories and develop strong organizations.
“I know people in our industry are nervous about AI,” Waterman said. “But we need to experiment. I’m glad I’ve had the chance here to see whether these tools can help us to communicate better with clients.”
