Invisible Elastic Fractional Teams
Invisible
The key advantage is not cheaper labor, it is variable capacity packaged like software. Invisible can break a workflow into small steps, route those steps through its internal platform to a distributed workforce of 3,000 plus people, and add or remove coverage as demand changes, which fits spiky AI training and back office workloads better than the fixed teams and long contracts common in classic outsourcing.
-
Traditional BPOs usually sell dedicated headcount for a defined process over multi year contracts. Invisible instead sells task level execution, where work can be split into microtasks and assigned across specialists, which makes it easier to handle changing volumes and more judgment heavy work.
-
That model showed up first in AI training. Invisible routed rating, ranking, and annotation work through trained raters, charging labs roughly $30 to $45 per hour while paying raters about $15 to $20 per hour. The same operating model can also be applied to enterprise ops workflows that need human review.
-
The shift matters because Invisible is moving from being compared with call center outsourcers toward being compared with AI operations companies like Scale AI and talent networks like Mercor. In 2024, Invisible reached an estimated $134M in revenue, versus Scale AI at $1.5B ARR and Mercor at a $50M run rate.
Going forward, the winners in outsourced operations will look less like static staffing vendors and more like workflow engines that can mix software, AI agents, and human reviewers. Invisible is well positioned if it keeps turning fixed labor into elastic capacity for regulated, high judgment work where customers need speed, audit trails, and the ability to resize teams quickly.