Headcount Zero is about scaling a company with AI and no employees. Each week, Jordan breaks down one practical system to win customers, deliver better work, and protect margin.
Reading the feed…
Solo automation consultants waste hours writing custom proposals for projects that follow the same patterns. Jordan breaks down the 3-2-1 portfolio framework: three service tiers (Starter, Pro, Enterprise), two standardized add-ons, and one hard scope gate that lets you price any project in sixty seconds. Using real boundary language from nine automation shops, you'll learn to constrain workflows, channels, interactions, and support windows to make delivery predictable and margins sustainable. Includes the exact Notion Offer Catalog template with toggleable SLAs, red-and-yellow scope flags, an
Jordan breaks down how to package your security posture, questionnaire automation, and SSO capabilities into a named, priceable Enterprise Solo Pack that gets procurement teams to yes faster than larger agencies. Learn to build a Trust Center backed by CAIQ-Lite answers, price SSO as a scoped add-on, structure SLA tiers, and run the 10-minute demo that closes enterprise deals without anyone asking about your team size.
Solo operators expose dozens of public endpoints—webhooks, client portals, form handlers—but unlike companies with security teams, nobody's checking if they're properly configured. This episode walks through setting up OWASP ZAP's passive baseline scan in Docker, scheduled weekly with cron, to catch missing security headers, CORS misconfigurations, and cookie issues before they cost you a client or a deal. Jordan covers the exact Docker Compose setup, config file tuning with OUTOFSCOPE patterns, exit code automation for Slack alerts, and ticket creation in Linear or Jira. The scan is passive-o
Every proposal Jordan sends includes an SLA section promising 99.9% uptime and 20-minute update cadences. But for eight months, he had no way to prove it. This episode builds the complete SLA monitoring automation: your monitor fires, a draft status update appears for approval, Slack enforces update cadence, and credit calculations run automatically when incidents close. Learn to connect Statuspage's API to your monitoring stack, implement human-in-the-loop approval gates, and automate the credit math that turns incidents into trust signals. By the end, you'll have the infrastructure to operat
If you have SLA commitments in your contracts and you're still calculating credits manually — or worse, waiting for a client to ask before you even check — you are one bad month away from a trust conversation you can't win. This episode shows you how to automate the entire trust paperwork stack: credit math, ledger tracking, status page updates, and incident communications. Jordan walks through building an SLA automation pipeline that computes monthly uptime from your audit logs, applies vendor-grade credit ladders (like Twilio's 10% threshold and AWS S3's tiered system), writes monthly ledger
Jordan walks through the exact migration playbook he used to move a revenue-critical 12-step Zapier workflow to Make without losing a single lead. Learn the 14-day shadow-run methodology with correlation IDs and parity checks, the three Gate Zero checks that determine migration feasibility, and the 15-minute cutover process with a tested 60-second rollback. Includes current 2026 cost models comparing Zapier tasks vs Make credits vs n8n executions, SSO parity requirements across platforms, and webhook throughput limits. Jordan saved $400/month on this specific workflow while proving the new sys
Jordan challenges the default assumption that agents are the sophisticated choice for automation workflows. Using side-by-side testing data, he demonstrates that deterministic flows outperform agents on reliability and cost for 80% of production work. For the 20% where agents are justified, he shows exactly how to box them with tool allowlists, hard timeouts, per-run budgets, checkpoints, and approval gates. The episode includes a practical A/B testing harness to benchmark any agent against your deterministic baseline before going live, plus the framework updates from Microsoft Agent Framework
After losing a $14K/month deal because he couldn't answer a 48-page security questionnaire, Jordan built a system that turns vendor security reviews from deal-killers into 40-minute tasks. In this episode, he walks through building a public Trust Center in Notion, creating a reusable answer library mapped to CAIQ-Lite and SIG Lite standards, and automating questionnaire responses with a Make flow that handles 70% of questions automatically while routing the rest for human approval. Jordan covers the two industry-standard "lite" frameworks most mid-market buyers accept, shows exactly which arti
Most private RAG deployments fail not because of the model, but because of three fixable infrastructure problems: stale embeddings from unchanged source documents, unbounded corpora that contaminate results with irrelevant content, and no retrieval guardrails to catch bad matches before they reach the LLM. This episode walks through building a production Supabase pgvector RAG system that solves all three: automatic re-embedding when source documents change, bounded corpus design, and SQL-level retrieval gates that block the LLM call when similarity scores are too low or keyword overlap is insu
When a two-word prompt change can approve fourteen spam submissions in one weekend, you need more than eyeballing outputs. This episode builds a regression gate—a frozen test set, three types of automated checkers, and a pass-fail threshold that blocks deploys when something breaks. The whole system runs inside Promptfoo, triggered from Make or n8n, and takes about forty-five minutes to set up from scratch. You'll get the exact CLI commands, environment variables, and exit code logic to catch prompt regressions before your clients do.
Most solo operators assume they need to self-host LLMs when enterprise prospects mention "on-prem" requirements. But you can satisfy 90% of enterprise security questionnaires without touching a single GPU driver. Jordan walks through the four controls that replace self-hosting: BYO API keys per tenant, region-pinned endpoints, structured logging with correlation IDs, and a visible kill switch in the client portal. You'll get the exact provider settings for Azure, OpenAI, Vertex AI, and Anthropic, plus a one-page checklist you can send to procurement the same day you get the ask.
When clients text you at midnight asking to turn off automations, you've hit the client proof problem. This episode walks through building a minimal control surface—not a full portal—that gives clients on/off toggles, SLA status chips, and a panic switch using Airtable + Softr + Make. Total build time under a day, total cost under $140/month. You'll get the exact schema, webhook wiring, and role-based visibility setup, plus a Notion + Super fallback for read-only clients. No custom code, no complex auth—just proof and control for the two things clients lose sleep over: is it working, and can I
Solo automation consultants running Make, Zapier, or n8n in production face a hidden risk: every platform has version history, but each has gaps that will cost you time, money, or client relationships during an actual incident. This episode delivers semantic versioning for no-code workflows, Git-backed exports that work on free tiers, and practiced 60-second rollback drills for each platform. Jordan walks through the exact system that saved him from a costly HubSpot sync failure and shows you how to build rollback-on-demand into your client automation practice.
Learn to build a complete data lineage automation system using correlation IDs, append-only logs, and exportable evidence. Jordan covers three implementation stacks - Airtable with permissions and interfaces, Notion with integration-only writes, and Postgres/Supabase with hash-chained tamper evidence. Plus a Slack slash command that answers "what happened on this run" with timestamps and status codes in under three seconds. Turn invisible automation work into defensible, auditor-ready evidence that shifts client conversations from "trust me" to "here's the proof."
Solo operators running Make or n8n workflows that touch client data face a hidden risk: PII and secrets leaking through LLM calls, API requests, and workflow logs. While vendors offer inconsistent protection—Vertex AI has non-configurable filters, Bedrock Guardrails cost per request, and Anthropic relies on prompt patterns—you need a solution that works everywhere and proves it's working. This episode walks through building a detect-mask-log subflow that drops in before any external call. Using Microsoft Presidio or OpenAI's new Privacy Filter, you'll scan for emails, phone numbers, API keys,
Turn your compliance policy from a PDF promise into executable rules that gate every API and LLM call in your Make and n8n workflows. This episode shows you how to encode consent flags, data residency requirements, and tool allowlists as JSON, then evaluate them at runtime using JSONLogic or json-rules-engine. Jordan walks through the complete implementation—from policy file structure to Code step evaluators to fail-closed routing with Slack approval overrides. You'll see specific examples for OpenAI's regional endpoints, Bedrock's model ID prefixes, Vertex AI's region processing, and Azure's
When enterprise buyers require SSO for your portal builds, you can ship SAML and OIDC authentication in under a week using a hosted identity provider. Jordan walks through the exact setup for Softr portals using Auth0 or Clerk as the authentication layer, including the critical ACS URL configuration that trips up most first attempts, a complete test plan for both SP-initiated and IdP-initiated flows, and the fallback strategy that keeps deals alive when SSO breaks during pilots. Also covers what's different on Glide and Retool, plus the three scoping questions that prevent you from over-engine
Solo automation consultants face a growing secrets management problem: API keys scattered across Make connections, n8n credential stores, and personal vaults with no audit trail or rotation strategy. Jordan breaks down the 1Password vs Doppler decision for teams of one, showing how to implement per-client scoping, automated rotation reminders, and centralized access logging. Learn the two-tier rotation strategy that aligns with AWS Config defaults, how to use Make teams and n8n projects for blast radius control, and why the proxy pattern matters for LLM integrations. Includes the complete Solo
Solo automation consultants pursuing enterprise clients face a hidden deal-killer: vendor security questionnaires. When procurement sends you 261 questions about your encryption policy, subprocessor list, and AI data handling, you need artifacts, not explanations. This episode builds the complete trust surface that prevents deals from dying in procurement — a one-page Notion Trust Center, a living subprocessor register, a plain-language Data Processing Addendum, and a Make scenario that automates DPA signature and storage. Jordan walks through the exact structure that mirrors CAIQ, SIG, and SO
Jordan shares the exact guardrail system he built after a $230 Zapier overage nearly cost him $700 in margin. You'll get a unified cost ledger that tracks LLM tokens, Zapier tasks, and Make credits per client, with Slack alerts at 60/80/100% of budget. Plus auto-throttle rules using Delay After Queue, n8n concurrency caps, and Make's built-in backoff—and kill switches on all three platforms that shut down non-critical automations before they eat your margin. Includes the complete template pack with alert formulas and platform-specific toggle recipes.
Turn messy client paragraphs into validated, build-ready specs using LLM JSON schema workflows. Jordan walks through the complete system: generate structured specs with OpenAI/Claude/Gemini Structured Outputs, enforce them with JSON Schema validation before any build begins, and auto-create Notion checklists from validated specs. Covers implementation paths for Make (external validator via HTTP) and n8n (Code node with Ajv or HTTP to microservice), plus the Cloudflare Worker setup that makes cross-platform validation portable. No more rewriting specs halfway through builds—the validator catche
Build a production-grade AI email triage system that safely automates 80-90% of your inbox while protecting against costly misroutes. Jordan walks through the Extract → Classify → Route pattern with confidence thresholds, deterministic overrides for high-risk emails, and provider-specific rate limiting across Make, Zapier, and n8n. Learn to respect Gmail's 15,000 quota units per minute and Exchange Online's 30 messages per minute caps while implementing exponential backoff and human-review queues. Includes downloadable JSON schema, confidence rubric, and Slack alert templates.
How many of your automations are one rate-limit change away from breaking? Jordan breaks down the decision framework for choosing between Airtable, Notion, and Postgres (Supabase) based on workload shape—not hype. Learn the exact throughput ceilings, concurrency limits, and cost breakpoints that determine which tool is right for your automation stack. Includes real client examples and a calculator to run your own numbers.
Build a universal automation monitoring system that catches failures before your clients do. Jordan walks through creating a Notion run log database that every workflow in your stack writes to, plus a Slack alerting layer that respects rate limits and only pings you on actual failures, duration spikes, and consecutive error patterns. The complete build takes 90 minutes, works across Make, Zapier, and n8n, and ensures you'll never learn about a broken workflow from a client email again. Includes the exact database schema, Slack app configuration, and post-run webhook snippets for immediate impl
If you're running ten or more Zaps, scenarios, or n8n workflows across client accounts without production-safe controls, you're one API hiccup away from a weekend you can't get back. Jordan walks through the four essential controls every automation consultant needs: idempotency keys to prevent duplicate runs, explicit retry logic that works when platforms fail you, a dead letter queue for clean error recovery, and circuit breakers to contain cascading failures. You'll see exactly where to toggle these settings in Zapier, Make, and n8n, plus the SQL and snippets to implement them across your en
