Marketing / customer management, tickets & business knowledge
350 clients. One central workspace.
Following a takeover, a marketing company’s client base grew from 150 to 350. AITJE brought together client data, tickets, projects, invoicing information and call transcripts in a central hub. Staff can search for and chat with their business experts, with access based on their role and AI integrated into an existing Mac mini.
The background
The phone rings.
What is the situation with this client?
The marketing company worked with 150 clients. Following a takeover, 200 clients were added to this. The client portfolio grew to 350, but the information was still scattered across various applications and teams.
When the phone rang, the employee didn’t immediately know what was going on. Was there an open ticket? Who was working on the project? Had anyone called back yet? And were there any outstanding invoices? Finding the answer always required searching through multiple systems.
AITJE It has built a central management hub where staff can quickly search by customer name, company name or website. The relevant data is brought together via API connections. In addition, they can ask questions via a RAG chat that utilises synchronised business knowledge.
The same need for context in every conversation.
A single point of access to the information in the connected systems.
The associated information
Standalone applications.
One client context.
The hub brings together data from the applications the company already uses. Staff can carry out targeted searches and ask questions that require information from various sources.
Clients & tickets
View customer data, tasks and tickets together with the member of staff working on them.
monday.com
Include project information when asking about progress and projects that require attention.
Rinkel
Include transcripts of telephone conversations so that previous conversations form part of the client context.
Invoicing details
View outstanding invoices alongside customer contact details, for staff authorised to access this information.
Questions from the working day
Ask for what
you want to know.
“Client X has phoned” is no longer just a casual note. Staff can ask further questions about tickets, responsibilities, conversations and projects, using the data their role is authorised to access.
Client X has phoned about Y and Z. Who is handling that ticket?
The hub brings together the client file, the relevant tickets and the assigned staff members. Conversation transcripts provide context for what has been discussed previously.
A schematic explanation of the applications; no shared customer data or actual chat logs.
A comparison of working methods
Fewer steps.
Greater coherence.
| During working hours | For the hub | With the hub |
|---|---|---|
| A customer on the phone | First, identify which company, project and ticket they relate to. | Search by client, company or website and view the available context. |
| A query regarding multiple systems | Compare customer data, tickets, invoices and conversations side by side. | Ask a question which the assistant will process using the connected sources. |
| Access to company information | Consult information via the individual applications. | A centralised workspace with access to data based on the employee’s role. |
| AI-gebruik | No dedicated central AI environment exists for this information. | Qwen 3.8 on the existing Mac mini, without external charges per AI token. |
A comparison of the approach taken for this project. No time savings or increase in turnover have been included as measured results.
The technology behind the overview
Search within knowledge.
Retrieve data via tools.
AITJE set up the Assistent software as a foundation and built the management hub around it. RAG makes the synchronised knowledge searchable; API and MCP links provide access to the connected applications. These are two routes that complement each other.
- 01
Connect
Code connects the hub to external applications via API and MCP integrations. Toolcalling provides the language model with defined ways to query the connected platforms.
- 02
Synchronise
Three times a day, the selected company data is retrieved and converted into vector embeddings. The local vector database makes this knowledge available to the RAG chat.
- 03
Search & retrieve
A name, company or website links to the available customer context. For a chat enquiry, RAG retrieves relevant passages; tool calls can also consult data from linked applications where these are accessible.
- 04
Responses with access
Qwen formulates the response using the available context. Access is restricted within the hub and when retrieving data based on the employee’s role, ensuring that the chat operates solely with authorised information.
A central hub does not mean access to everything.
Employees use their own roles. The application and data links restrict access: financial information is only made available to authorised employees. The language model does not determine these rights itself.
Making the most of existing hardware
Mac minis.
Expertise within our own network.
The company already had an Mac mini. AITJE used this as a local computing environment running Debian Linux, the AITJE Assistentsoftware and Qwen 3.8. As a result, virtually no new hardware was required for this project.
Employees access the hub via the local network. The language model, vector embeddings and RAG processing run within the organisation’s own environment. Synchronisations and tool integrations are configured via code.
The hub operates locally and can access external applications. Three synchronisations per day update the knowledge base as long as the sources are accessible.
The stored knowledge reflects the last successful synchronisation; without an internet connection, it is not automatically updated.
The costs
Existing hardware.
No external AI meter.
Mac minis were already in place. The company does not pay any external fees per AI token for local model calls, vector embeddings or the agent. The synchronisation code also runs under the company’s own management.
This makes the AI costs different from those of an external model API that charges for every processing operation. Power, management and the construction of the hub remain part of the solution. Existing external platforms, such as monday.com and Rinkel, retain their own subscriptions and any usage fees.
From customer requirements to product direction
The inspiration for
AITJE Manager.
This project demonstrated the value that a central management hub can offer: a single location for customer context, tasks, conversations and knowledge from existing applications. The project inspired AITJE to develop AITJE Manager.
For this company, the solution was built on AITJE Assistent, enhanced with bespoke features and the company’s own integrations. In this way, the growth to 350 customers was translated into a workspace that helps employees regain an overview.
Other case studies
Development agency / Assistant, Coder & model management
Ten models. One team. Control and independence.
AITJE set up a shared AI environment for a development agency, featuring Assistent, Coder and ten models to choose from. A shared token pool of €350 per month replaces nine subscriptions to €200. Model usage and token costs per task can be tracked via Asana.
View the approach
Marketing agency / level design & in-house coding agency
25 New game levels every week. With no external token costs.
A marketing agency took over an existing mobile puzzle game but ran into difficulties when creating new levels. AITJE analysed the game, built a reusable level design system and linked AITJE Coder to Qwen on its own server. Now, the agency 25 creates new levels every week from a single assignment in OpenCode.
View the approach
Power leads / 3D catalogue & workflow optimisation
From technical drawing to 3D product model.
AITJE built a workflow for a partner specialising in contact cables that converts product data and technical drawings from WordPress into 3D models. AITJE Coder controls Blender; Three.js makes the GLB files usable on the website. Following optimisation, API costs fell from approximately €9 to €0.11–€0.18 per product.
View the approachMOVING FORWARD TOGETHER WITH AI
Your next step begins with a question.
Got an idea, facing a challenge, or simply curious? Tell us what you want to achieve. AITJE will work with you to find a solution.
