Ertas + OpenClaw for AI Agencies

The Challenge

OpenClaw is the most capable AI agent framework available — autonomous task execution through messaging platforms that clients already use, browser automation, file management, email triage, and cron-based monitoring. For AI agencies, it is a natural evolution from simple chatbot deployments to full-spectrum AI assistants.

But the economics do not work at scale with cloud APIs. Every client's OpenClaw instance routes every interaction through GPT-4o or Claude — generating per-token charges on every email triaged, every report generated, every message sent. A single active client can cost AU$150-300/month in API pass-through. Across 10-20 clients, the agency is looking at AU$2,000-4,000/month in variable costs that eat directly into fixed-retainer margins.

Worse, the costs are unpredictable. A client's marketing campaign goes viral and their support volume triples? The API bill triples too — but the retainer stays the same. An OpenClaw cron job scans an inbox every 30 minutes? That is steady token throughput 24/7, burning through credits even when nothing actionable happens.

The differentiation problem is equally pressing. If every agency deploys OpenClaw with the same GPT-4o backend, clients are effectively getting the same AI — just wrapped in different branding. There is no moat. A technically inclined client can replicate the setup themselves.

The Solution

Ertas transforms agency OpenClaw deployments from API pass-through to proprietary AI infrastructure. The model: fine-tune a per-client LoRA adapter on each client's domain data, deploy all adapters on a shared base model running locally via Ollama, and point each client's OpenClaw instance at the local endpoint.

The economics flip immediately. Instead of AU$150-300/month per client in API costs, inference is free after a one-time hardware investment. A Mac Studio or RTX 4090 server handles 15-20 concurrent client adapters comfortably. The hardware pays for itself in 4-6 weeks. Every client added after that is pure margin.

The quality improvement is the real selling point. A fine-tuned model for a real estate client has learned that client's listings, pricing terminology, and buyer communication style. A fine-tuned model for a dental practice knows the practice's appointment types, insurance panels, and patient communication tone. Generic GPT-4o approximates from a system prompt; a fine-tuned model internalises from training data. Agencies can demonstrate measurable accuracy improvements to clients — which justifies premium pricing and creates switching costs that generic API access never provides.

Key Features

Example Workflow

A Sydney-based AI automation agency manages OpenClaw deployments for 12 small business clients across real estate, hospitality, and professional services. Each client has an OpenClaw agent handling email triage, appointment scheduling, and customer inquiry responses through WhatsApp and email. On cloud APIs, the agency's monthly API spend is AU$2,800 — with three high-volume clients (a real estate agency, a boutique hotel, and an accounting firm) accounting for AU$1,500 of that total. The agency migrates to Ertas. For each client, they export 3-6 months of conversation history, format as JSONL training data, and fine-tune a LoRA adapter in Ertas Studio. Training takes 30-60 minutes per client. All 12 adapters are deployed on a single Mac Studio M2 Ultra (AU$5,500) running Ollama, with adapter hot-swapping based on which client's OpenClaw instance is making the request. After migration, the agency's monthly AI inference cost drops from AU$2,800 to AU$14.50 (Ertas subscription) plus electricity. The hardware breaks even in under 8 weeks. More importantly, client satisfaction improves: the real estate client's inquiry classification accuracy jumps from 76% (GPT-4o) to 93% (fine-tuned model), and the hotel's booking confirmation responses match their brand voice so closely that guests cannot distinguish them from human staff. The agency now sells the fine-tuned model as a premium differentiator — justifying a 30% higher retainer than competitors who deploy generic API-based agents.