Dashboard and Custom Business Intelligence
Most SMEs have tons of data but no simple way to read it. I develop decision-making dashboards that make these data usable — tailored to your profession, not generic factories.
Do you manage your business without a dashboard?
Most of the SMEs I have worked with share the same observation: The data is there, but it's not being used to steer. Decisions are made based on gut feeling, or on reports that arrive too late.
- Important numbers are scattered across dozens of Excel files, and no one consolidates them — and when someone tries, it takes half a day
- Performance indicators do not exist, or they are calculated manually once a month
- No real-time visibility on revenue, margins, inventory, or commercial pipeline
- Monthly reports arrive too late for reaction
- The ERP or CRM contains useful data, but no one knows how to export them manually
What I Build
Each solution is designed for your professional context. No pre-made models, no endless configuration — custom development with technologies adapted to your scale.
Real-time Dashboard
KPIs, revenue, margins, inventory, cash - all accessible from a browser. Refresh via AJAX every few minutes, or in real-time via WebSocket with Django Channels if necessary.
No need to install anything. Just a link, a password, that's all.
Automated Reporting
Weekly or monthly reports generated and sent via email without anyone clicking a button. Automatic Excel Reporting, PDF, or directly within the dashboard.
ETL Pipelines
Data aggregation from multiple sources: ERP, CRM, accounting, CSV files, third-party APIs. Cleaning, transformation, and centralization through Python scripts (Pandas, SQLAlchemy) or Apache Airflow for heavier cases.
Interactive Visualizations
Charts, maps, drill-down by period, region, product. Data Visualization designed for decision-makers, not for data scientists.
Alerts and Thresholds
Email or SMS when a performance indicator exceeds a critical threshold. Low stock, falling margin, goal achieved.
Integration with your tools
Connect to your ERP, CRM, accounting software, or any other system via API REST, database, or import file. I adapt to what you already have in place – no need to change everything.
Examples of Realizations
Recent projects, anonymized. Each solution was custom-built to address a specific professional need.
Multi-warehouses dashboard
Initial issue: Each warehouse had its own Excel tracking file, updated manually. No one had a consolidated view. I connected the existing WMS via its REST API, and built a dashboard with real-time stock levels, tracking of shipments, and return rates for the 3 sites. Result: logistics managers stopped spending 2 hours a day compiling numbers.
Dashboard commercial pipeline
The commercial director was making his predictions in a spreadsheet, with numbers dating from the previous week. I connected the CRM (refresh every 15 minutes) to display the sales pipeline progress, sales predictions per quarter, and conversion rate by step. Finally, he now has a reliable 90-day outlook.
Automated HR Reporting
Automated monthly reporting: absenteeism rate, department turnover, salary mass evolution. Generation of PDF via ReportLab, automatic sending to the management committee every first Monday of the month. Previously, it took 2 days for one person.
Production and Maintenance Dashboard
Factory equipped with 12 machines equipped with temperature/vibration sensors. The sensors publish in MQTT on a Mosquitto broker, a Python consumer writes the measurements into PostgreSQL (TimescaleDB for historization). The Grafana dashboard displays the real-time activity monitoring: synthetic performance rate (TRS), downtime, preventive maintenance alerts when vibrations exceed a threshold. Display on 24-inch screens in the workshop.
Not a generic tool, a tailor-made solution
The self-service BI tools (Power BI, Tableau, Qlik) are good products. However, for an SME, they often pose the same problems: recurring licenses that pile up, complex configuration, and ultimately a tool that does 80% of what's needed but not the 20% that matter for your business.
Generic Tool
- Monthly per-user license (this figure quickly adds up with 10+ people)
- Adapt to the tool, not the other way around
- Features frozen by the editor
- Dependency on a closed ecosystem
- Training necessary for each user
- Connectors to your internal tools rarely exist
Custom Solution
- No recurring license - you pay the dev, you own the code
- Built around your profession and existing processes
- Evolve when needed
- Open technologies (Python, PostgreSQL, Docker), open-source code
- Streamlined Interface, No Training Required
In practice, I start by looking at your existing data — bases, exports, files. Together we identify the 3 or 4 key performance indicators (KPIs) that you really lack. I deliver a first usable version within 2-3 weeks, and we adjust from there.
Technologies
I choose the appropriate technical stack for your project. No dogmatism: if Power BI or Metabase is the best fit for your case, I will let you know.
Treatment and Analysis
Visualization
Backend and Infrastructure
Let's talk about your data
Tell me what data you have and what is missing for you to manage. I will respond to you within 48 hours with a first concrete proposal.
Frequently Asked Questions
How much does a custom dashboard cost?
A simple **enterprise dashboard** (3-5 KPIs, one data source) starts around 3,000 EUR. A complete project with **multi-source ETL pipeline**, alerts, and **automated reporting** is between 8,000 and 20,000 EUR depending on complexity. There is no recurring license fee: you pay for the dev, the code is yours.
What are the delivery deadlines?
It depends on the scope. A decision dashboard with a few KPIs, 2 to 4 weeks. A complete project with multi-source ETL and reporting, rather 6 to 12 weeks. In practice, you see the first results as early as the second week - I deliver in iterations.
Can I keep my Excel files?
Yes. The dashboard automatically ingests your Excel files. The idea is not to replace Excel, but to replace the manual Excel reporting where it slows you down.
Do I need a data warehouse?
Honestly, for most SMEs, no. With 2-3 data sources, a well-structured PostgreSQL database does the job. A data warehouse becomes relevant when you have 5+ sources, large volumes, or historical needs over several years. It depends on your case — I'm not going to sell you a data warehouse if you don't need one.
Real-time or batch: what's the difference?
Batch = Your data is updated at intervals (every hour, every night). Real-time = continuous update via websockets. In practice, a refresh every 15 minutes is sufficient for 90% of cases. Real-time is for production tracking or logistics — not for a sales dashboard.
Who maintains the dashboard after delivery?
You are independent. The code is well-documented, version-controlled with Git, and deployed with Docker. For evolutions (new KPIs, new data sources), I propose a maintenance contract or ad-hoc interventions. Your IT team can also take over — the source code belongs to you.
What is the cost for this type of service?
From 2,000 EUR excluding tax for a simple dashboard. A complete BI project (ETL + dashboards + training) ranges from 5,000 to 10,000 EUR excluding tax.