Skip to content
All posts

How to Prepare Your ERG Program for AI in 2026

Maceo Owens
Written by Maceo Owens
Published 06/03/2026 · Updated 06/03/2026 · 4 min read

A 4-step order of operations for adding AI to your ERG program in 2026 — pulled from the video — so automation and agents actually work instead of multiplying chaos.

[[youtube:ZrJPwlcES04]]

Most ERG program managers will spend 2026 trying to bolt AI onto a program that isn't ready for it. Here's the short version: AI can only repeat what your system already defines. If your ERG operating system is vague, AI just makes the chaos faster.

This post is the direct answer to how to actually prepare — pulled from the video above.

The 4-step order of operations

Do these in order. Skip a step and the next one collapses.

1. Define your Minimum Viable Quarter (MVQ)

Write down the smallest set of things an ERG leader must do in a 90-day period for the ERG to count as "on track." Not aspirational — minimum. If you can't name it, no tool can enforce it.

Think of the MVQ as the floor beneath your ERG program structure. Without that floor, every quarter feels like starting from scratch. With it, you have something to automate against.

2. Document the 3Ps for every ERG

Purpose, People, Programming. One page per ERG. This becomes the source of truth that automation and AI will reference. The video calls this the "ERG operating system" — and it's the same concept behind an ERG blueprint. When each ERG has its 3Ps written down, you stop answering the same questions and start delegating to process.

If you're early in your program, start with the one-pager version before building the full operating system.

3. Automate the repeatable parts

Recurring reminders, meeting cadences, reporting templates, onboarding for new ERG leaders. Boring, deterministic workflows — not AI yet. The goal is to turn ERG roles into repeatable processes so continuity doesn't depend on who remembers what.

This is also where an ERG program manager's guide to better processes becomes useful: mapping what actually happens each quarter, then stripping out the repetitive steps.

4. Then layer AI

Once steps 1–3 exist, AI can draft event recaps, summarize member feedback, suggest programming based on the 3Ps doc, and flag ERGs drifting from their MVQ. It works because there's something concrete to reason against.

Two learnings from the video worth pulling out

1. Consistency is a system problem, not a people problem. When ERG leaders go quiet, the instinct is to recruit harder or train more. The video reframes it: inconsistency almost always means there's no operating system underneath the volunteers to be consistent with. Fix the system and the same humans suddenly look reliable.

This is the same insight behind why you don't scale mess: growth without structure just multiplies the chaos.

2. The five components of an ERG operating system. The video breaks the OS into five pieces: a defined purpose, a leadership structure, a programming rhythm, a measurement loop, and an escalation path to the business.

That escalation path is especially important — it connects to the work of getting middle managers to actually support ERGs. Without that support, the operating system has no teeth. If you're seeing resistance, read why middle managers won't support your ERGs first to diagnose the real blockers.

AI plugs into all five components — but only after each one exists in writing.

What to do this week

  • Pick one ERG and write its MVQ in a single paragraph.
  • Document its 3Ps on one page.
  • List every recurring task that currently lives in someone's head.

That list is your automation backlog. Once it's running cleanly, you've earned the right to add AI on top — and it'll actually work.


Keep going: