AfterQuery Customer Concentration Risk

AfterQuery is selling into buyers that can wake up one day and decide to build the same stack themselves. Frontier labs already run model training, evals, and internal tooling, so a vendor focused on human curated post training data is useful to them only as long as it is faster or better than doing it in house. That makes customer concentration more dangerous here than in ordinary enterprise software, because the customer can also become the replacement product.

• The pattern is visible across the market. Scale sells labeling, training, and deployment tooling as a full suite. Turing and Mercor use large expert networks to supply labor more flexibly. AfterQuery sits at the high quality, high touch end, with custom datasets, agent environments, and validation tooling rather than a broad marketplace.
  1 sacra 2 sacra 6 sacra 7 sacra
• The likely concentrated buyers are the frontier labs themselves. OpenAI and Anthropic are among the few companies spending heavily enough on post training and eval infrastructure to support a specialized vendor, and both have the technical depth and capital to internalize those workflows if they want tighter control over quality, speed, or cost.
  1 sacra 4 sacra 5 sacra
• The company already has meaningful scale, with an estimated $100M revenue run rate as of April 30, 2026. That shows real demand, but if a small number of labs account for a large share of that spend, losing even one flagship account would hit revenue harder than it would for a vendor with hundreds of mid sized customers.
  1 sacra

The path forward is to become harder to replace than a staffing vendor and easier to trust than a general marketplace. If AfterQuery can turn bespoke lab work into reusable eval frameworks, domain specific expert pools, and enterprise ready validation products, it can shift from being a temporary external team for frontier labs to a durable infrastructure layer used by both labs and enterprises.