Enhancing Newswire Products with AI and Machine Learning

How can Cision/PR Newswire enhance its products by incorporating artificial intelligence and machine learning?

Shipped an MVP of a generative AI press release builder — a new product built from scratch inside Cision Labs, PR Newswire’s first foray into AI-assisted content creation for its press release suite, built to close a capability gap against faster-moving competitors.

Fig1. PRNewswire AI & ML press release builder MVP.

The problem

PR Newswire’s core product — press release creation and distribution — hadn’t meaningfully changed since automating outreach in the 1950s. Competitors were shipping AI-assisted writing tools faster than Cision could match, creating real competitive pressure to move quickly without cutting corners on quality.

Fig2. High fidelity mockup of the initial form used to create the generative AI press release.

The process

Started with competitive analysis to find a differentiated direction rather than replicate what was already on the market, alongside Product Management’s requirements work to scope a realistic MVP. From there: low-fidelity wireframes and user flow diagrams to map the press-release creation journey, informed by research into adjacent writing and content tools to understand what users already expected from AI-assisted writing.

Moved to high-fidelity mockups built inside PR Newswire’s existing design system — new functionality had to feel native to the product, not bolted on. Built an interactive Figma prototype to let stakeholders experience the actual flow rather than review static screens, which shaped several rounds of refinement before the concept was locked.

Fig3. High fidelity mock up of the initial generative AI press release outline.

My role

Lead Product Designer on a cross-functional team with Product Management, Engineering, and Data Science. Owned competitive analysis, flows, wireframes, high-fidelity design, and the interactive prototype.

Fig5. A user flow diagram was developed in order to get a sense of how the user would move through the interface and what types of interactions may be possible.

The result

Delivered a working MVP that gives press-release authors a generative-AI-assisted starting point instead of a blank form — a new product category for PR Newswire’s suite, in a market segment competitors were already contesting. Specific gains in creation speed or output quality weren’t measured before handoff.

Fig6. High fidelity mockup of the generative AI press release editor interface.

What I’d do differently

Competitive analysis was treated as a one-time input rather than an ongoing practice. Given the market was moving as fast as it was, tracking competitors continuously — not just at kickoff — would have surfaced differentiation opportunities that a single analysis upfront couldn’t.