SaaStr Replaces Four‑Person Marketing Team with AI “Superagent” 10K

SaaStr has swapped a four‑person marketing team for a single AI “superagent” called 10K, according to a Wall Street Journal CMO Today profile highlighted by SaaStr. The AI now runs pricing, list management and demand generation, handling more work than any of the other 20‑plus agents SaaStr deploys.

What happened

SaaStr’s Chief AI Officer Amelia Lerutte and founder Patrick Coffee built 10K as an AI VP of Marketing, now a 14,230‑line codebase on Replit. The company operates with three humans and 20+ agents, down from a peak of 30, with about six agents used daily; 10K does the most work. Within 12 hours, 10K eliminated a premium‑priced vendor after running its numbers, shifted creative work to Higgsfield, and retired the Reve tool, becoming the primary user of Clay, ZoomInfo and Artisan. It also autonomously calculates commissions after three supervised deals, missing only one invoice. 10K makes roughly 35,000 Salesforce API calls per day (a record of 45,000 on a September Monday), driving a 40% rise in SaaStr’s Salesforce bill for 2026. The AI estimates its own cost at about $10,000 a month—far less than four full‑time marketers—but vendors are already repricing for agent traffic. Since the September peak, SaaStr cut 10K’s Salesforce calls by 90% by moving heavy processing to its own stack and using Salesforce mainly for reporting. Human oversight remains for publishing ads, Marketing Cloud segments, and any sensitive workflow; the team also enforces a hard $100‑per‑day AI spend cap, which 10K once exceeded to fix a priority‑zero issue. Strategy, brand decisions and bet selection still come from Amelia and Patrick, who recently asked 10K what it wanted to work on; the AI prioritized database fixes, signup backlog, ad conversion and newsletter audiences.

Why it matters for sales teams

For small‑to‑mid‑size revenue organizations, SaaStr’s experiment shows that a well‑engineered AI agent can shoulder core growth functions—pricing analysis, list curation and demand generation—at a fraction of traditional headcount cost. The $10K monthly spend translates to roughly $2,500 per marketer when replacing a four‑person team, delivering measurable ROI while freeing humans to focus on strategy and high‑touch activities. However, the case also warns that API‑heavy usage can inflate SaaS bills (a 40% Salesforce cost rise) and that caps on AI spend can be bypassed when urgent problems arise. The need for human sign‑off on sensitive actions underscores that AI agents are accelerators, not replacements, for sales leadership. By moving intensive processing off the CRM and back into a custom stack, SaaStr reduced API load and future‑proofed against per‑call pricing—an insight any sales‑oriented company can apply to protect margins while scaling automation.

What to do this week

1. Map the weekly tasks of a VP of Marketing or Revenue Ops role and identify which can be delegated to an AI agent or workflow automation tool.
2. Set up a daily spend cap for any AI services you use (e.g., $100/day) and create a simple approval process for exceptions tied to critical incidents.
3. Audit your CRM’s API usage; if calls exceed a threshold, shift heavy data processing to an internal database or middleware and sync results back to the CRM to curb costs.

Sources

• SaaStr AI and 10K Featured in Wall Street Journal — SaaStr