Bitrix24 showed how to add four automation robots to a business portal

Bitrix24 has released the next installment in its practical development series for its business portal, this time focusing on automation. In the new material, the team shows how to add four robots to the system that bring data to a unified format and collect key metrics on tasks and deals.

Which Robots Were Added

This is not an abstract demonstration, but a practical set of automations for CRM and workflows. The authors take the frontend application built in the previous part and expand it with a set of robots that can be embedded directly into portal scenarios. The logic is simple: the cleaner the source data and the fewer manual actions employees perform, the easier it is to build reports and launch the next stages of the funnel without constant manual verification.

• Phone number correction — the robot brings contacts to a unified format so they can be correctly used in CRM and subsequent scenarios.
• Name normalization — automation removes inconsistencies in spelling and helps make customer and contact cards cleaner.
• Task data collection — the robot extracts necessary metrics for monitoring work and internal analytics.
• Deal sum calculation — the scenario collects financial data to quickly assess the funnel and manager workload.

What the Project Is Built On

The project uses a prepared AI starter — a project template designed for development alongside AI agents. It already contains a structure, instructions, and templates that help you get up to speed faster and avoid spending time on basic setup. Essentially, Bitrix24 offers not just a set of automation ideas, but a startup environment where these ideas can be replicated, customized for your portal, and tested on a real scenario.

This approach is important for teams just starting to build internal extensions for Bitrix24. Instead of a "blank slate," the developer gets a sequential path: take the template, spin up the project, add the necessary robots, verify data processing, and then tie the automation to specific CRM entities. An additional plus — the article promises to show not only the final result, but all changes step by step, lowering the barrier to entry for those who haven't worked with such scenarios before.

Where Complexity Usually Arises

"Today we'll show how to build automation and add 4 robots: phone

number correction, name normalization, and data collection on tasks and deal amounts."

From the description itself, you can see where problems most often arise. Any automation runs into the quality of source data: phone numbers are stored in different formats, names are entered with errors, and needed metrics are scattered across multiple entities. That's why even small robots require careful validation, clear normalization rules, and verification that after processing the data remains suitable for CRM logic and reporting. This is typically where first implementation attempts break down.

The authors specifically emphasize that the material addresses potential problems during operation and their solutions. This is an important signal: the article is oriented not just at demonstrating the "happy path," but at real development, where you have to fix configuration, catch integration errors, and adapt scenarios to the structure of a specific portal. For practicing teams, the value is precisely here — you can immediately see typical failures and understand how not to get stuck during implementation.

Another practical bonus — a project repository is available alongside the walkthrough. This means the material can be used not just as reading material, but as the foundation for your own prototype: compare file structures, replicate changes, and transfer individual logic blocks to your corporate scenario without lengthy preparation. For teams, this is a convenient way to quickly move from theory to a real working pilot within your portal.

What This Means

Bitrix24 is gradually turning AI topics from a general conversation into a set of specific business scenarios. For companies living inside a CRM, this is a useful signal: automation doesn't start with "intelligent agents in general," but with small robots that clean data, calculate metrics, and remove manual routine from standard operations.