ZoomInfo's pSEO Collapse: Anatomy of a 90% Traffic Loss

What happened to ZoomInfo's SEO traffic?

Between October 4 and October 20, 2023, ZoomInfo's Sistrix visibility index dropped from 9.6638 to 5.4531, a 16-day loss of -43.57%. By year-end, ZoomInfo sat at #94 on Sistrix's IndexWatch US 2023 losers list with a full-year visibility decline from 21.85 to 9.19, or -57.94%.

The damage was concentrated, not site-wide. ZoomInfo's homepage, pricing page, and branded queries kept ranking. The pages that died were the directory templates: millions of company profile URLs (/c/[company-name]/[id]) and people profile URLs (/p/[person-name]/[id]) that had been driving the bulk of non-branded organic traffic since 2020.

Gaetano DiNardi, who first surfaced the data publicly, framed the timeline cleanly: organic traffic took off after ZoomInfo's June 2020 IPO, peaked in mid-2023, and started its decline in October 2023. By Q2 2024, the directory-page traffic stream that built ZoomInfo's lead-generation engine was a fraction of its peak.

ZoomInfo's own 2024 Annual Report acknowledges the dependency, listing search engine algorithm changes as a material risk: ZoomInfo notes it relies on "the indexing of public-facing directory pages" for a significant portion of website traffic.

Which Google update caused ZoomInfo's organic decline?

The October 2023 Spam Update is the proximate cause. It rolled out October 4-20, 2023, overlapping with the October 2023 Core Update (October 5-19). Google's announcement explicitly named the targets: cloaking, hacked content, auto-generated content, and scraped content.

ZoomInfo's directory pages matched two of those four categories at scale. The company profile and people profile templates were programmatically generated from a database. Much of the data was scraped or aggregated from public sources -- press releases, SEC filings, social profiles, and corporate websites.

The damage compounded across subsequent updates. The March 2024 Core Update introduced Google's formal scaled content abuse policy, which targeted "creating many pages where the content makes little or no sense to a reader." By April 2024, the rollout had completed, and pSEO directories that survived October 2023 took a second hit.

Lily Ray of Amsive, writing in the Sistrix analyst network, characterized the October 2023 wave as one of the most aggressive spam-targeted rollouts of the year. ZoomInfo wasn't named individually in her analysis, but the pattern she described -- "sites that targeted all the keywords" losing 70%+ of traffic -- maps cleanly onto ZoomInfo's directory architecture.

What were the specific signals that flagged ZoomInfo's pages?

Five compounding signals tripped Google's classifiers. Each one alone might have been survivable. The combination was not.

1. Auto-generated content at scale. ZoomInfo had millions of pages stamped from two database templates: company profiles and people profiles. Google's scaled content abuse policy names this pattern explicitly.

2. Paywall on the value content. The contact data that justified the page's existence -- emails, direct phone numbers, decision-maker info -- was gated behind ZoomInfo signup. Users hit a wall after the click. Bounce rates spiked, dwell time collapsed.

3. Two-template uniformity. ZoomInfo used essentially two page molds across the entire directory. Google's doorway page guidance flags large groups of pages "that exist mainly to direct users somewhere else."

4. SERP saturation from copycats. Apollo, RocketReach, Lusha, and others scraped largely the same public records and shipped near-identical pages. Google's helpfulness systems suppress sites where there's no distinguishing value.

5. Engagement signals consistent with low value. DiNardi's public analysis hypothesized that high bounce rates and low time on page accelerated the decline. The paywall plus templated content guaranteed those metrics.

Why did profile pages survive on some sites but not ZoomInfo?

LinkedIn, Crunchbase, and Wikipedia all run profile-page directories that look superficially similar to ZoomInfo's. None of them collapsed in October 2023. The reason is content depth and signal density per page, not page type.

LinkedIn profiles are user-generated and user-maintained. Each profile carries posts, comments, reactions, recommendations, endorsements, job changes, and connection signals. Google sees fresh activity on most active profiles weekly. ZoomInfo's pages updated when the underlying database refreshed, with no per-page user signals.

Crunchbase pages are editorially curated with citations to funding announcements, news coverage, and primary-source documents. The platform layers human research on top of the database, and pages link out to the sources that justify each fact. ZoomInfo's pages summarized data without sourcing it.

Wikipedia entries carry external citations, talk-page discussion, edit history, and linkbacks from across the open web. Each article is a small node in a co-mention graph. ZoomInfo's directory pages had no equivalent off-site citation footprint.

The lesson: programmatic page templates aren't penalized because they're programmatic. They're penalized when the unit of content cannot stand on its own as helpful to a reader who didn't already know what they were looking for.

What can B2B SaaS learn from ZoomInfo's collapse?

Five lessons hold up across every post-2023 pSEO failure analysis, including other directory collapses we've documented.

1. Gate less. Signal more. If your most useful data point is hidden behind a signup wall, every visit produces a negative engagement signal. Surface enough genuine value above the wall that users want to read, scroll, and link.

2. Diversify your page templates. Two molds across millions of URLs is a doorway pattern. Build at least 4-6 distinct page types per pillar -- each with different sections, data sources, and intent matches.

3. Add a unique signal per page. Editorial summary, dated insights, customer review snippet, computed metric, fresh data point. Something that exists only on this page and would not appear on a competitor's scrape.

4. Monitor indexation rates weekly. DiNardi has noted clients with 8M discovered pages and only 650K indexed. When Google indexes less than 30% of your directory, the algorithm is already telling you most of the pages don't deserve to exist. Cut them before the next spam update does.

5. Never copy a 5-year-old pSEO playbook. ZoomInfo built its directory in 2018-2020 when it was novel. Apollo and RocketReach copied it in 2021-2022. By 2023, the pattern was a category-wide spam signature. If a strategy has been public for three years, the algorithmic countermeasure is already built. See will programmatic SEO get penalized for the full risk framework.

Signal	Google's Spam Policy Match	ZoomInfo's Pattern	Likely Penalty Weight
Auto-generated content at scale	Scaled content abuse	Millions of company + people profile pages from a single template	High
Thin content behind paywall	Low-value content, deceptive UX	Most contact data hidden until signup	High
Duplicate templates	Doorway pages	Two page molds repeated across the directory	Medium-High
Poor engagement signals	Unhelpful content (HCU overlap)	High bounce, low time on page reported	Medium
SERP saturation with copycats	Lack of unique value	Apollo + RocketReach scraped the same public records	Medium