Let your data talk
Business intelligence.
This is a major issue for businesses.
Let's explain why.
What are we talking about?
Business intelligence.
The name sounds serious, but it's very simple.
It involves using programmes to help you make decisions.
Based on the data generated by your company's activities.
This is history
Before keyboards, there were punched cards.
Before computers, there were human calculators.
Before integrated circuits, there were vacuum lamps.
Before processors, there were electro-mechanical relays.
Computers were invented to calculate quickly and efficiently.
During the Second World War, to guide shells and decipher messages.
From the 1950s onwards, to facilitate the work of office workers.
Exploiting data is nothing new!
What data?
Typically: customers, sales, finance, management.
Often also data from monitoring IT infrastructures or SaaS services. Developers are often equipped with their own tools.
For operational staff, this is less often the case.
Yet the needs are immense.
When you have to make a decision, you might as well make the right one.
The right decisions
There are two schools of thought.
First of all, there are the geniuses.
They have so much business in their blood that they know how to make decisions with their guts. Or by smelling. Or by meditating.
The lucky ones ...
They're rare.
We belong to the other school.
The second school
We like things square and systematic.
We prefer to use the information available.
It's a good way of avoiding blind spots, biases, beliefs, denials and other psychological mechanisms.
The advantage is that we have a basis (of data, #humour) for discussion.
And it's more relaxing.
Information
Useful information is often hidden away.
You have to look for it in the raw data.
This means filtering, selecting and, of course, cleaning.
Calculating averages, trends or smoothing.
The aim is to calculate the right indicators.
First step
Everyone has used MS Excel at some point.
Everyone has complained about its incomprehensible formulas.
Everyone has tried to make ugly graphs.
MS Excel isn't enough, and neither is Google Sheet.
MS Power BI is used a lot.
In the world of data, we use something else.
Python
This is the right tool for data processing.
Its graphics library is immense.
Its capabilities are infinite.
Python is practical and can be learned (fairly) quickly.
You'll always find a young engineer who knows.
Dashboards
A dashboard is a webapp.
The technology is available.
Python is the right language for data processing.
Streamlit, Dash or Panel as a framework for display.
You deploy the whole thing on a virtual machine.
And here you are, your SaaS dashboard, in just a few days' work.
Frankly, it's fast.
The layout options are endless.
And it costs less to license than Tableau or Looker.
By the way
We're working to democratise data science.
First of all, that means visualising your data.
Then using it to make the right decisions.
Data-driven management is the first step.
That's our conviction.