WhatsApp member concierge built on Make AI Agents

Source-checked Oct 11, 2026Case study

OPUS runs two Make AI Agents (one internal, one member-facing on WhatsApp) and 49 workflows in three layers: data capture, data augmentation and personal interactions. Members ask who to meet or which events fit them.

Source: Make, Make case study, 2025

1What it does

2Who it is for

Industry: Professional services
Platform: Make
Category: AI Employee
Tools used: Make AI AgentsWhatsAppHubSpotOpenAI

3How it works

Every step, in order, is in Workflow at a glance, pinned on the right as you scroll.

Workflow at a glanceMake scenario

01 Trigger
Member WhatsApp message.
02 Action
Capture member data (HubSpot).
03 Action
Internal agent enriches profiles.
04 Action
Member-facing 'operator' agent on WhatsApp answers and matches people.

Trigger: Member WhatsApp message.
AI step: Unknown.

4How it went

Per Make: one central place for member questions, 24/7 service, and personal matching across a larger international membership. The title claims 10x member experience; no metric details given.

Only what the source reports. The source did not report numbers.

5Worth knowing

Uses Make's built-in data stores instead of a separate database.

6Recipe

Adapted from source Reworded or reconstructed from what the source shows.

Pattern: data capture scenarios -> data store -> internal enrichment agent -> member agent with a 'find matches' tool.

7Source

Open the original Make case study by Make →

Source opened and read on 2026-10-11.

More AI Employee builds

All builds →

HighLevel AI Employee

Claude runs a GoHighLevel account through HighLevel's MCP server

Claude Desktop connects straight to a HighLevel sub-account with a scoped Private Integration token. From one sentence it pulls contacts, tags them, moves pipeline stages and surfaces incomplete leads.

Agency Creator's settings

HighLevel AI Employee

HighLevel AI Agent workflow action: form lead follow-up and no-show recovery templates

HighLevel's AI Agent action plans and runs multi-step CRM work inside one workflow step. You write the goal, give it up to 10 tools, and it pulls contact, pipeline and calendar context on its own. Ships with templates like Form Lead Follow Up and No Show Appointment Recovery.

Any local business Written by Builds That Work

HighLevel AI Employee

Hail alert agent that only messages contacts in NOAA hail ZIP codes

Instead of blasting every CRM contact after a storm, a HighLevel AI agent checks NOAA hail reports and county ZIP codes and filters down to the contacts actually in the hail path.

Roofing Written by Builds That Work