Getting Started with Ottolax: Your First AI Agent in 5 Minutes

One of the most common concerns agencies have before trying Ottolax is that setting up AI agents must be complicated. It is a reasonable assumption — “AI-powered multi-agent automation platform” sounds like something that requires a PhD in computer science to operate.

It does not. Creating your first agent takes about five minutes, and this guide walks you through every step.

Before You Start

You will need two things:

1. An Ottolax account — sign up free at app.ottolax.com. No credit card required.
2. Access to at least one tool you want to automate — Google Analytics, Google Ads, Slack, a CRM, or any of our 70+ supported integrations. You will connect it during setup.

That is it. No software to install, no code to write, no configuration files to edit.

Step 1: Create Your First Agent

After signing in, you land on the Ottolax dashboard. This is your command center for managing agents, viewing activity, and monitoring performance.

Click “Create Agent” in the top navigation. You will see the agent creation screen with three fields:

Agent Name — Give it a descriptive name that reflects its function. Good examples: “Weekly Client Report Generator,” “Google Ads Budget Monitor,” or “Slack Daily Digest.” The name is for your reference, so make it something your team will immediately understand.

Agent Objective — This is where you describe what you want the agent to do, in plain English. Be specific about the outcome you want. For example:

• “Pull Google Analytics traffic data for each client every Monday at 8am and send a summary to the client-reports Slack channel”
• “Monitor all active Google Ads campaigns and alert me in Slack if any campaign’s cost-per-lead exceeds $50”
• “Every Friday at 3pm, compile a weekly performance summary for each client and email it to their primary contact”

The more specific you are about what data to use, what actions to take, and when to do it, the better your agent will perform. Think of it like writing instructions for a capable new team member.

Agent Type — Choose from preset templates or start from scratch. Templates give your agent a head start with pre-configured behaviors for common workflows like reporting, monitoring, or client communication. Starting from scratch gives you full control.

For your first agent, we recommend choosing a template that matches your goal. You can always customize it later.

Step 2: Connect Your Integrations

After creating your agent, the next screen asks you to connect the tools your agent needs. Click “Add Integration” and you will see the full catalog of supported services organized by category.

Select the integration you need — for this example, let us connect Google Analytics. Click the Google Analytics card and you will be prompted to authenticate with your Google account. Standard OAuth flow — you will see Google’s permission screen, approve the access, and you are connected.

A few important things about integrations:

• Minimal permissions — Ottolax requests only the specific API scopes needed for your agent’s task. A reporting agent needs read access to your analytics data but never needs write access.
• Per-agent assignment — Each integration connection is assigned to specific agents. Connecting Google Analytics for your reporting agent does not automatically give every other agent access to it.
• One-time setup — Once you connect an integration, it stays connected. You can reuse it across multiple agents without re-authenticating.

For agents that send outputs somewhere (reports, alerts, summaries), you will also want to connect a delivery channel — Slack, email, or your project management tool.

Step 3: Configure the Schedule

Most agents run on a schedule rather than continuously. The scheduling interface lets you define exactly when your agent should execute:

• Frequency — hourly, daily, weekly, monthly, or custom cron expression
• Time — when during the day to run (accounting for timezone)
• Day — for weekly/monthly schedules, which day to execute

For a weekly client report, you might set it to run every Monday at 7:00 AM so reports are ready when your team starts work. For campaign monitoring, you might run it every hour during business hours.

You can also trigger agents manually at any time from the dashboard — useful for testing or one-off executions.

Step 4: Test Your Agent

Before letting your agent run on autopilot, test it with a single execution. Click “Run Now” on your agent’s detail page. You will see the execution in real time:

1. Data collection — the agent connects to your integrated tools and pulls the relevant data
2. Processing — the agent analyzes the data according to its objective
3. Action — the agent delivers the output (sends the report, posts the alert, updates the record)
4. Completion — you see a summary of what the agent did, including any data it accessed and actions it took

Review the output carefully. Did the report include the right metrics? Did the alert trigger correctly? Did the message go to the right channel? If something needs adjustment, edit the agent’s objective or configuration and run it again.

This test-and-refine cycle usually takes one or two iterations to get right. Once you are satisfied with the output, your agent is ready to run on schedule.

Step 5: Monitor and Refine

Your agent is now live and will execute according to its schedule. But the setup is not a one-and-done process — the best results come from monitoring and refining over the first few weeks.

The Activity Log on each agent’s detail page shows every execution with full details: what data was accessed, what decisions the agent made, what actions it took, and how long it ran. Use this to:

• Verify accuracy — make sure the agent’s outputs are correct and useful
• Tune the objective — if the agent is including irrelevant information or missing important details, adjust the objective description
• Optimize timing — if reports are arriving too early or too late, adjust the schedule
• Expand scope — once the basic workflow is solid, consider adding more data sources or outputs

Building Your First Agent Team

Once you are comfortable with individual agents, the real power of Ottolax emerges through agent teams. A team is a group of agents that work together, passing data and context between each other to handle complex multi-step workflows.

For example, a reporting team might include:

• Data Collection Agent — pulls raw data from Google Analytics, Google Ads, and Meta Ads every Monday morning
• Analysis Agent — receives the raw data, normalizes it across platforms, identifies trends and anomalies
• Report Generation Agent — takes the analysis output and creates a formatted, natural-language report for each client
• Delivery Agent — sends each report to the right client via their preferred channel

Each agent specializes in one part of the workflow, and the team coordinates automatically. If the Data Collection Agent encounters an API error, it retries and notifies the team. If the Analysis Agent finds something alarming, it can trigger an immediate alert in addition to the regular report.

Setting up a team follows the same pattern as individual agents — create each agent, define its objective, connect its integrations, and then link them together in a team workflow.

Common First Agents

If you are not sure where to start, here are the most popular first agents that agencies set up:

Weekly Performance Report — connects to Google Analytics and ad platforms, generates a client-facing performance summary every Monday. Setup time: 5 minutes.

Budget Monitor — watches ad spend across platforms and alerts you when campaigns are pacing toward overspend or underspend. Setup time: 3 minutes.

Daily Slack Digest — compiles a morning summary of key metrics across all clients and posts it to your team Slack channel. Setup time: 4 minutes.

Client Onboarding Checklist — when a new client is added, creates tasks in your project management tool, sends welcome emails, and sets up initial reporting configurations. Setup time: 10 minutes.

What Happens Next

Your first agent is just the beginning. As you see the time savings from automating one workflow, you will quickly identify others that would benefit from the same treatment. The typical progression looks like this:

1. Week 1 — set up your first agent (reporting or monitoring)
2. Week 2-3 — refine the first agent, add one or two more
3. Month 2 — build your first agent team for a multi-step workflow
4. Month 3+ — expand across all clients and workflows, reclaiming 20+ hours per week

Start with the workflow that causes the most friction in your current operations. Automate that one well. Then expand from there.

Create your free account at app.ottolax.com and have your first agent running in five minutes.