What the work actually is

Mercor is building high-fidelity replicas of a marketing org's tool surface — GA4, Search Console, Semrush or Ahrefs, a CMS, CRM, project management, ads, support — populated with realistic documents, dashboards, personas and deliberately planted problems. Your job on the Organic Growth track is to convert work you have actually done into a task brief: what the agent is asked to do, what the planted issue is, what a strong practitioner concludes, and what a plausible-but-wrong answer looks like. That last part is the hard part. A task that only has one obvious answer measures nothing; a good brief has a trap that a competent-sounding response walks into — the canonical tag that explains the traffic drop that everyone blames on an algorithm update, the keyword cluster whose volume is real but whose intent is transactional and unwinnable.

You will also specify the artifacts: which dashboard shows which number, which Slack thread contains the contradicting claim, what date range the GSC export covers, which pages the CMS actually has. Then you write or review the reference answer and the criteria separating strong from weak, and later grade AI agent attempts against your own standard. Task types span diagnosis from messy or conflicting data, prioritization and tradeoffs, planning and execution, QA and reconciliation, triage and escalation, research, reporting, and multi-tool orchestration.

What the screen is looking for

  • Hands on keyboard, not hands on a team. They say it plainly: owned the work, not only managed people who did it. Expect follow-ups that go one layer past what a manager would know — sampling in GA4, property-level vs URL-prefix differences in GSC, how you'd reconcile Semrush estimates against GSC impressions.
  • AEO literacy, not just classic SEO. Answer-engine visibility is named in the scope. Vague gestures at "AI search" will not survive a probe.
  • Ability to articulate why. Much of the value is written reasoning. If you can diagnose but can't explain the diagnosis in a way a grader could apply consistently, the output is not usable.
  • Prior environment/rubric work is a strong preference, not a gate. Rubric Academy or Bootcamp certification is likewise preferred, not required.

Applicants complete a short multiple-choice knowledge screener specific to organic growth before human or AI review.

Logistics

Remote and largely asynchronous, contractor engagement, hourly at an observed $60–100 depending on depth and track. Volume varies by cohort — some contributors run 5–10 hours a week alongside client or in-house work, others take larger blocks during environment build pushes. Expect written deliverables with review cycles rather than live meetings, and expect your briefs to come back with questions about whether the planted issue is truly discoverable from the artifacts you specified.