If you're still treating press releases as your AI visibility strategy, the data has bad news. Inithouse, a product studio running a portfolio that includes an AI citation tracking tool called Be Recommended, spent a year measuring how wire-distributed press releases perform across five major AI engines: ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews. The findings should make any marketing team rethink their content budget allocation.
The Decay Pattern Is Brutal
The numbers are consistent and damning. Citations for wire-distributed press releases peak within the first 30 days, drop to roughly half by month six, and fall below detection threshold somewhere around 18 months post-publication. Inithouse tracked citation persistence across hundreds of brands using Be Recommended's scoring system—which rates AI visibility on a scale from zero to 100—and the pattern held regardless of brand size or industry vertical. A typical press release gets strong initial pickup when it hits the wire, shows up in grounded answers, and occasionally gets paraphrased accurately by AI engines. Then it fades. Fast.
Why Press Releases Lose Ground
Two structural biases drive this decay. First, recency bias: AI engines weight newer sources more heavily because that's what makes sense for factual accuracy. A press release from 14 months ago is competing against blog posts, case studies, and data reports published last week. The fresh content wins every time. Second, primary-source preference. Inithouse measured Perplexity specifically favoring primary sources over syndicated content at roughly a three-to-one ratio. When your press release gets republished on dozens of wire service sites, none of those copies count as primary. They're echoes, not origins. A blog post published on your own domain with original data and first-hand experience? That's primary. That wins.
What Actually Builds Lasting AI Visibility
The brands maintaining stable scores in Inithouse's reports share traits that shouldn't surprise anyone who's been paying attention to how these systems work. They publish frequently on their own domain—blog posts, case studies, data reports with specific numbers, named customers, and clear claims. Products publishing at least two primary-source pieces per month stayed within five points of their AI visibility score over a six-month period. Products relying on press releases alone dropped 15 to 25 points in the same timeframe. They structure content for extraction: short paragraphs, clear claims in the first two sentences, data points that can be quoted verbatim. And they reference themselves by name consistently—something Inithouse noticed many brands fail at, using multiple name variations across their own content and confusing AI engines into treating them as separate entities.
The Measurement Shift
Inithouse shifted its framework from 'did the press release get cited?' to three metrics that actually predict sustained visibility. Primary-source citation ratio measures what percentage of AI citations come from a brand's own domain versus syndicated third-party content—higher ratios correlate with more stable long-term visibility. Content recency curve tracks how old the most-cited piece of content is for any given brand; evergreen visibility correlates with most-cited content published within the last 90 days. Answer-position frequency measures whether your brand appears in an AI engine's first answer paragraph or buried further down as a supporting reference, since both Perplexity and ChatGPT show measurable preference for direct answers over citations.
The Bottom Line
Press releases still have value—media pickup, investor signaling, SEO backlinks—but as an AI visibility strategy, they're dead on arrival after 18 months. Companies spending $3,000 to $15,000 annually on wire distribution are getting strong initial pickup and near-zero long-term grounding. If you want AI engines to recommend your brand consistently, invest in primary-source content published on your own domain. Make it data-specific, recent, and structured for extraction. The average brand scores around 31 on Inithouse's visibility scale; brands doing primary-source content right hit 80 or higher. That gap is almost entirely explained by content strategy, not budget size.