Day 7 - Data and Logic Deep Dive

The goal is to learn how to build intelligent, data-driven automations using variables, objects, loops, decisions, and Automize's Excel-like formula engine - enabling your bots to make dynamic choices and manipulate data efficiently.

1. Why Logic Matters in Automation

Logic transforms a recorded sequence of steps into a smart process that adapts to changing data and conditions.

With logic, a bot can:

• Decide what to do based on runtime values.

• Process multi-level data objects.

• Manipulate text, numbers, and dates dynamically.

• Adapt to changing environments.

2. Data Types & Variables - including JSON Objects

Common Variable Types

Important: Automize uses JSON objects (not List/Table) for structured/multi-level data. Objects can nest multiple levels and are accessed using our object access syntax.

JSON Object Access Syntax

When a variable contains a JSON object, access nested values using any of the supported syntaxes:

• {var::subvar::field}

• {var:subvar}

• {var.var}

• {var|var}

• {var||var}

• {var->var}

All of the above are accepted and equivalent for accessing nested properties. Choose the style that matches your team's convention.

Example JSON object stored in {Invoice}

{ "Invoice": { "Header": { "Number": "INV-1001", "Date": "2025-11-12", "Customer": { "Name": "Acme Corp", "ID": "C123" } }, "Lines": [ {"SKU": "A1", "Qty": 2, "UnitPrice": 50}, {"SKU": "B2", "Qty": 1, "UnitPrice": 100} ], "Total": 200 }}

Access examples

• {Invoice::Header::Number} → INV-1001

• {Invoice:Header:Customer:Name} → Acme Corp

• {Invoice.Header.Date} → 2025-11-12

• {Invoice->Lines->0->UnitPrice} → 50 (first line item unit price)

• {Invoice|Total} → 200

Note: When indexing arrays use numeric indexes: {Invoice::Lines::0::SKU}.

Casting Variables

You can force a variable to be treated as a specific type using casting syntax:

• {INT:var} - convert var to integer

• {FLOAT:var} - convert var to float/decimal

• {STR:var} - convert var to string

Examples

• {INT:{Invoice::Total}} → Converts total to integer

• =SUM({FLOAT:Price1},{FLOAT:Price2}) → Forces numeric addition even if inputs were text

Special Built-In Objects

Automize includes domain-specific objects you can work with directly (examples):

• Invoices - see help/working-with-invoices-1

• Quotes - see help/help/working-with-quotes-1

• Purchase Orders - see help/working-with-purchase-orders-1

These objects are structured JSON and expose fields relevant to that document type (header, lines, totals, supplier, etc.). Use the same multi-level access syntax to read/write their fields.

Excel-like Functions

For all supported calculation and string functions refer to:

• Help: help/what-calculation-functions-are-supported-1

This page documents the full list (CONCAT, IF, SUM, LEFT, RIGHT, MID, FIND, LEN, DATEDIF, etc.) and examples for using them inside Automize formulas.

3. Working with Variables and Objects - Hands-On

Exercise 1 - Create and Read an Object

1. Create a new process "Invoice Read Demo".

2. Add a Set Variable activity:

   • Variable: {Invoice}

   • Value: paste the example JSON above (or your own invoice JSON).

3. Add Log Message activities to output:

   • Log: "Invoice Number: {Invoice::Header::Number}"

   • Log: "Customer: {Invoice::Header::Customer::Name}"

   • Log: "First Line SKU: {Invoice::Lines::0::SKU}"

4. Run the process.

Goal: Logs display the correct nested values.

Exercise 2 - Casting and Calculation

1. Add variables: {Price1} = "100.50", {Price2} = "49.50".

2. Add a Formula activity:

   • Formula: =SUM({FLOAT:Price1},{FLOAT:Price2})

   • Output: {Total}

3. Log: Log: "Total = {Total}".

Goal: Total logs as 150.0 (numeric addition after casting).

4. Formula Engine & Practical Examples

Common Formula Patterns

• Concatenate nested fields:

  =CONCAT({Invoice::Header::Customer::Name}, " - ", {Invoice::Header::Number})

• Conditional logic on object fields:

  =IF({INT:Invoice::Total} > 1000, "High", "Normal")

• Pull a date and format (example using supported date functions):

  =TEXT({Invoice::Header::Date},"yyyy-mm-dd")

See help/what-calculation-functions-are-supported-1 for full function list and syntax.

5. Decisions using Objects

Example decision: If an invoice's total is over 1000 then send for approval.

1. Add a Decision activity with condition:

   {INT:Invoice::Total} > 1000

2. True connector → Approve flow.

3. False connector → Normal flow.

Connector colours:

• Always - Blue

• Default / True - Green

• False - Amber

• Exception - Red

Use colours to visually identify connector types on the process map.

6. Loops with Objects

Process a list of invoices represented as an object array:

Example object

{ "Invoices": [ {"ID":"INV1","Total":500}, {"ID":"INV2","Total":1500} ]}

For Each loop

• Input: {Invoices}

• Item variable: {CurrentInvoice}

• Inside loop you can reference: {CurrentInvoice::ID}, {CurrentInvoice::Total}

Hands-On Exercise - Invoice Loop

1. Set variable {Invoices} to example JSON.

2. For Each {CurrentInvoice} in {Invoices}:

   • Decision: {INT:CurrentInvoice::Total} > 1000

   • True → Log: "Invoice {CurrentInvoice::ID} is High Value"

   • False → Log: "Invoice {CurrentInvoice::ID} is Normal"

Goal: Correctly logs classification per invoice.

7. Variable Syntax Variants - Quick Reference

All of the following are supported for accessing nested values:

• {var::sub::field}

• {var:sub}

• {var.var}

• {var|var}

• {var||var}

• {var->var}

Casting examples

• {INT:var} - integer

• {FLOAT:var} - floating point

• {STR:var} - string

Array indexing

• {Object::Array::0::Field} - first item in Array.

8. Special Objects - Usage Notes

• Invoices: Use for payment processes, reconciliation, and approvals. (See help/working-with-invoices-1)

• Quotes: Use for sales quoting automations. (See help/help/working-with-quotes-1)

• Purchase Orders: Use for procurement automations. (See help/working-with-purchase-orders-1)

Each special object exposes expected fields and helper functions. Use the object-specific help pages for field names and example payloads.

9. Best Practices - Objects & Logic

• Keep object schemas consistent across processes (same field names/structures).

• Prefer objects over many flat variables for complex data - they're easier to pass and extend.

• Always cast before numeric ops: use {INT:}, {FLOAT:} to avoid errors from string values.

• Log key object fields at start/end of processing for auditability.

• Avoid deep inline formulas - break into intermediary variables for clarity and debugging.

End of Day 7 Summary

By completing the updated Day 7, you should now be able to:

✅ Use JSON objects as your primary structured data type✅ Access nested fields using multiple supported syntaxes✅ Cast variables explicitly for safe calculations✅ Combine objects with formulas, loops, and decisions✅ Use built-in domain objects (Invoices, Quotes, POs) and consult their help pages✅ Recognise connector types and their colours (Always - Blue, Default/True - Green, False - Amber, Exception - Red)

Next Step - Day 8

If you want, I'll build Day 8 - Reporting, Analytics & Continuous Improvement next, including how to extract KPIs from logs, build dashboards, and run periodic health checks on your automations.