If Even Chainalysis Can Get Hacked, Can Google Really Trust Your Business Listing?

A Data Company Just Got Breached

Chainalysis sells trust for a living. The entire business model is built on one promise: we can verify where crypto money actually went, who actually controls it, and whether the data behind a transaction is clean. That's the product. So when the company disclosed in November 2025 that attackers had stolen credentials and authentication tokens affecting its entire customer base, it wasn't just a bad week for a crypto compliance vendor. It was a demonstration that even a company whose sole job is verifying data integrity can have its own data integrity compromised.

Why does this matter if you run a dental practice in Ohio or a landscaping company in Arizona? Because the systems recommending your business to customers right now, Google's AI Overviews, ChatGPT, Perplexity, don't pull from some magical understanding of who you are. They pull from third-party data sources: review platforms, business listing aggregators, schema markup on your site. Every one of those sources is a company, run by people, sitting on servers somewhere.

If Chainalysis, built from the ground up to be the authority on data trust, can get breached, the review platform or listing aggregator your AI visibility depends on isn't immune either. The question isn't whether these systems are trustworthy. It's whether you've done anything to protect yourself when they're not.

What AI Engines Actually Pull From

there's no crawler reading your homepage and deciding you sound trustworthy. ChatGPT, Perplexity, and Google's AI Overviews aren't ranking pages anymore. They're assembling answers from structured data they can verify, and they're citing their sources while they do it.

Google pulls heavily from Google Business Profiles because that data is first-party and it already controls the verification process. Perplexity has spent 2025 building out shopping and local data partnerships, including deals with outlets like Dow Jones and Reuters, specifically so its answers carry citations a user can check. ChatGPT's shopping and browsing features, expanded through late 2025, lean on the same pattern: Yelp listings, schema-marked websites, Google Business Profiles. Three different companies, three different products, and they all converged on the same sourcing strategy.

Search Engine Land and Moz both flagged this trend heading into 2026: consistent NAP data, Schema.org markup, and review volume are the factors that keep showing up as the top predictors of whether an AI system cites your business at all. Not your homepage copy. Not your meta description. The structured stuff sitting underneath it.

The NAP Problem Nobody Fixes

most businesses are failing the NAP test and don't know it, because nobody ever told them to check.

NAP stands for name, address, phone, and it sounds too simple to matter. It isn't. If your Google Business Profile lists "123 Main St Suite 4" and your website footer says "123 Main Street, Ste. 4" and your Yelp listing drops the suite number entirely, you've created three slightly different versions of your business. A human reading all three would figure it out in two seconds. An AI system weighing which listing to trust, or whether to cite you at all, treats that mismatch as a signal of unreliable data.

Pull up your own listings right now and check them against each other. Google Business Profile, Yelp, Facebook, Apple Maps, your own website footer and contact page, any directory you signed up for in 2019 and forgot about. Look for inconsistent suite numbers, old phone numbers you migrated away from, abbreviations that don't match (Street versus St, Avenue versus Ave), and your business name formatted differently across platforms (with or without an LLC, a comma, a location modifier).

Then layer in whether you've got Schema.org markup identifying that data on your site, and whether your review volume across platforms looks substantial enough to count as a trust signal. Most businesses have never audited any of this.

Why Keyword Tricks Won't Save You

a lot of business owners still think the fix here is more keywords, or a bigger ad budget, or finally getting that Google Ads campaign optimized. None of that touches AI recommendations. ChatGPT and Perplexity aren't crawling your landing page looking for "best plumber near me" repeated eleven times. They're checking whether your Google Business Profile, your Yelp listing, and your schema markup agree with each other and whether real customers have left enough reviews to back any of it up.

Paid ads buy you visibility in one specific channel. They do nothing for whether an AI system trusts your data enough to cite you in an answer. Keyword stuffing is even less useful. It was a workaround for how search engines used to rank pages, and AI systems don't rank pages anymore, they assemble answers from structured sources they can verify.

What actually moves the needle is boring by comparison. Verified reviews across multiple platforms. Schema.org markup that tells the AI exactly what it's looking at. Third-party mentions that corroborate what your own site says about you. None of it is flashy, and none of it fits into a one-time campaign. It's closer to bookkeeping than marketing, which is probably why most businesses have skipped it entirely.