Every platform merger is a data event.

On July 1, 2026, Glassdoor legally merged into Indeed, closing a consolidation that began when Recruit Holdings bought Glassdoor for $1.2 billion in 2018. The announcement emails describe continuity: same brand, same reviews, one login. Read as a product update, it is unremarkable. Read as a data event, it is the labor market losing an independent observation point — and almost nobody who depends on that data noticed.

Two datasets became one vantage point

Indeed and Glassdoor measured overlapping realities from different angles: one from job postings and applications, the other from employee reviews and self-reported salaries. Analysts triangulated between them precisely because they disagreed — disagreement between independent sources is where signal lives. Under a single operating entity, with unified accounts, shared terms, and one privacy policy, the two datasets share governance, incentives, and increasingly, plumbing. The nominal brands survive; the independence of the measurements does not.

The integration timeline shows how consolidation actually proceeds: an operational merger and layoffs announced mid-2025, mandatory Indeed sign-in for Glassdoor users by late 2025, account linking deadlines in spring 2026, and the legal merger last month. Each step was individually small. The cumulative effect is that a market once observed from two positions is now observed from one.

Consolidation is the norm, not the exception

This is not a story about one merger. Job platforms, property portals, and social networks all trend toward concentration, because aggregation economics reward it. Every acquisition quietly does three things to the data layer: it reduces the number of independent sources, it moves a dataset under new terms that can change without notice, and it creates a single point of failure — commercial, technical, and policy — where there used to be two.

For anyone building on market data, the lesson is structural: source diversity is not a nice-to-have; it is the asset. A pipeline that reads one platform is a pipeline that inherits that platform's every future decision.

Why we build multi-source by default

This thesis is load-bearing in our data intelligence venture. The AgentX catalog's flagship collectors are deliberately multi-source — the jobs orchestrator reads dozens of hiring platforms through one normalized schema, so a merger, an API shutdown, or a terms change on any single source degrades coverage instead of severing it. The same architecture backs our property and video collectors. BestCrawler's field guide on LinkedIn, Indeed, and Glassdoor job data — written when those were three meaningfully distinct vantage points — is now also a document of why we refuse to depend on any one of them.

The quiet part

Platform consolidation never announces itself as a data event. It arrives as a cheerful email about better products and single sign-on. The teams that notice are the ones whose infrastructure already assumes that any source can disappear, merge, or change its terms tomorrow — because on a long enough timeline, every source does.

← All notes