1. First things first, sign up or log in to Supabase.
2. Create a new project. Just give it a name and pick your database region.
3. Once your project is up, head over to the "Database" section to check out your database schema.

1. In the "Database" section, you'll want to create new tables for your analytics data. For instance, you might set up a table called user_analytics with columns like user_id, event_name, event_data, and timestamp.
2. Make sure to define the right data types for each column. For example, user_id could be a UUID, event_name as Text, event_data as JSON, and timestamp as Timestamptz.

CREATE TABLE user_analytics (
  id uuid PRIMARY KEY DEFAULT uuid_generate_v4(),
  user_id uuid,
  event_name text,
  event_data jsonb,
  timestamp timestamptz DEFAULT now()
);

1. Navigate to the "API" section in your Supabase dashboard to find your API URL and key.
2. If needed, enable Row Level Security (RLS) and create new policies to secure your tables. For example, you might want to allow only authenticated users to insert analytics data:

CREATE POLICY "Allow logged-in users to insert data"
  ON user_analytics
  FOR INSERT
  TO authenticated;

1. In your app, install the Supabase client library. For example, if you're using JavaScript:

npm install @supabase/supabase-js

1. Initialize the Supabase client with your API details:

import { createClient } from '@supabase/supabase-js';

const supabaseUrl = 'your-supabase-url';
const supabaseKey = 'your-supabase-key';
const supabase = createClient(supabaseUrl, supabaseKey);

1. Create a function to log analytics data. For example:

async function logAnalytics(user_id, event_name, event_data) {
  const { data, error } = await supabase
    .from('user_analytics')
    .insert([{ user_id, event_name, event_data }]);

  if (error) {
    console.error('Error logging analytics:', error);
  }
}

1. Use Supabase's querying capabilities to fetch and analyze your data. For example, to fetch all events for a specific user:

async function getUserAnalytics(user_id) {
  const { data, error } = await supabase
    .from('user_analytics')
    .select('*')
    .eq('user_id', user_id);

  if (error) {
    console.error('Error fetching analytics:', error);
  }
  
  return data;
}

1. Process or visualize the fetched data as needed in your application.

1. Create database triggers or scheduled tasks to automate specific analytics tasks. For example, creating a weekly summary of user activities:

CREATE OR REPLACE FUNCTION weekly_summary()
RETURNS void LANGUAGE plpgsql AS $$
BEGIN
  -- Your summary logic here
END;
$$;

CREATE TRIGGER weekly_summary_trigger
AFTER INSERT ON user_analytics
FOR EACH STATEMENT
EXECUTE FUNCTION weekly_summary();

1. Use Supabase's built-in features like PostgreSQL functions, triggers, and cron jobs for deeper automation of analytics tasks.

2. Make sure all your triggers and automated tasks are well-tested and optimized for performance.

This step-by-step guide helps you set up custom analytics using Supabase, covering everything from initialization to advanced querying and automated analysis.