Join us in Dallas for LocalU
October 20, 2026 - Just $299
You know that moment of dread when you realize you’ve been completely wrong about something for years?
That’s exactly where we found ourselves recently. For quite a long time, we stood by the results we saw in years past, that press releases were an SEO ghost town, and we didn’t bother testing it again. They didn’t move the needle, and they certainly didn’t justify the effort.
We were wrong.
New testing has completely flipped our stance. We’re now seeing undeniable proof that press releases written about real news and statistics, when used strategically, can be a potent force for improving local, organic, and even AI overview rankings.
We’re sharing case studies from our research, including one that saw an 83% jump in traffic almost overnight.
We tested this tactic first on a realtor’s key landing page. Our expectation was that we wouldn’t see much impact, based on our bias from years past. The press release went out, linking back to their service page.
The results, however, were anything but flat.
In just 28 days following the press release, traffic to that target page jumped by 83 percent. This wasn’t a small fluctuation; this was a massive, statistically significant surge.
But the gains weren’t just in raw traffic. We also observed a noticeable, immediate improvement in their Local Pack rankings for several key city- and neighborhood- specific searches. The press release provided the immediate authority signal that Google’s algorithm needed to boost its local visibility.

Key Takeaway: A well-targeted press release can deliver an immediate and dramatic impact on both traffic and local search visibility.
Our second test, because one test does not make a strategy, was on a law firm and was more subtle, but equally compelling.
After the press release went live, the results didn’t skyrocket like the first case. Our immediate reaction was “Oh, the previous test was a one-time wonder.” We held steady and checked back in a few weeks later, and then it materialized, a slow, steady progression:

This felt like a compounding effect. The initial authority signal from the press release seemed to set the stage, and as Google processed the information and the resulting citations, the organic benefit slowly materialized over time.
This raised a critical question for us: are these ranking improvements an ongoing, cumulative benefit or a flash in the pan with no long-term impact?
The third test, on a local lawn care company, is what truly blew our previous assumptions out of the water.
After their press release went live, they saw the expected gains based on our first two tests. Their Local Pack rankings rose, and their organic rankings followed, indicating a strong, positive response.

But here is the part that changed everything:
The AI Overview for their main service keyword actually changed to quote directly from the press release.
Think about the significance of that. Google’s AI-generated summary, the new zero-click result everyone is trying to influence to gain visibility, was pulling language straight from the content we included in the press release.
Press releases are no longer about backlinks. They’re now about being a trusted and authoritative source that Google’s generative AI deems authoritative enough to quote and feature. This is a massive shift in how we view the utility of PR.
The New Equation: Press Releases = Potential AI Trust Signal
Five years ago, we tested press releases and found no impact on ranking. Nothing.
But today? That has completely changed. Well-written and structured press releases are influencing:
The only difference is time and a continuously evolving search algorithm.
The ultimate takeaway for every SEO professional and business owner is simple: Always be testing.
What didn’t work last year (or five years ago) might be a solid tactic today. SEO changes fast, and if you stop experimenting, you are guaranteed to miss opportunities like this one.

Which press release service did you use?
Hi Brandon – we use https://38digitalmarket.spp.io/r/NJ3R20 (partner link)
Tq you for good Information
Would you recommend writing a press release for an office move? I'm trying to get everything lined up with updated NAP in advance to minimize chances of GBP suspension when I change the address (or lay the groundwork for an easy reverification). Thanks!
Hi Adriena, I think an office move is a good reason for a press release. It's "news" for customers, plus having all of the address, name, etc included in the NEW PR is a good idea from a "consistency of new information" standpoint.
Which package to recommend from the PR company that you use? Seems they offer a couple.
Also, what anchor were used in these tests? URL? Exact Match?
Hey Joe – we use their base package here.
Thanks Carrie, and where are you sending the links, what anchors are being used?
Depends upon the goal – sometimes a keyword phrase, sometimes just the brand – linking to whatever page on the website might be relevant to the topic. If you spam up a bunch of Press Releases with gobs of anchor text keywords, you’re going to devalue what you’re doing.
6/5/2026 EDIT to add: After the most recent core updates, we are recommending to NOT use keyword rich anchor text in press releases at this time. Just a regular raw link to the page/site/domain you’re hoping to improve.
Thanks for posting this. It clarified a few things I was unsure about.
Do you have links to the press releases used in the tests? Thank you!
We can’t share those, but we use https://38digitalmarket.spp.io/r/NJ3R20 (partner link)
How was AI visibility tested?
We can see traffic from LLMs, like Gemini & ChatGPT, in Google Analytics (GA4)
What was the content of the actual press releases?
The content focused on notable changes to the business, like a new office, or awards they have won.
Is this article still valid after the most recent core update?
Hi Shaun,
Yes! But be careful not to include anchor text in the Press Release – just a nice HTML link to a page that complements the press release content. Press releases continue to deliver for our clients.