Case Studies Umumiy_uz Case Study: Odoo Implementation for a Distribution Company in Uzbekistan and +7–9% Profit Growth
This case study shows how we implemented Odoo for Umumiy_uz and unified: product cost & margin visibility, customer hist...
The client’s team communicates in Telegram groups and tracks work requests through Jira. The assistant treated some casual messages that mentioned people as work requests and created incorrect tasks on the Jira board. Team members then had to find, review, and delete those entries.
We’re adjusting the assistant so it can distinguish casual conversation from genuine assignments more accurately. Group members can mention one another and chat freely, while only work requests that require action are added to the Jira board.
Mentioning someone in a group can mean different things. Someone may be waiting for a reply, sharing information, or directing the conversation to that person. The previous setup treated some of these messages as tasks too.
One incorrect task may seem like a minor issue. But someone still has to find it on the Jira board, confirm that it is not actually an assignment, and delete it. As these errors add up, trust in the board begins to decline.
We’re changing the assistant’s message-selection rules. A message marked as not being a task stays in the conversation. Genuine requests phrased politely, as well as assignments that include an image or file, should still become Jira tasks.
The solution should not simply reduce the number of tasks. It needs to move genuine work requests to the Jira board while leaving casual communication where it belongs.

The project also includes a separate document for the manager. It will bring together all relevant tasks, their full descriptions, and links. Instead of reviewing the Jira board page by page, the manager can read the necessary information in a single document.
The document will initially be provided privately to one user. The rest of the team will continue using the existing buttons on the Jira board. The project is still in progress, with approximately 24 hours of engineering work allocated to it.

If you need to convert Telegram messages into Jira tasks more accurately or adapt your task board to your team, contact Celion. We’ll first identify where messages are being misinterpreted, then plan a solution that fits your workflow.
Callback
Leave your details and we will contact you shortly.
Become a Partner
Leave your contact details and we will get back to you shortly.
Order Service
Leave your phone number and we will contact you shortly.