Get in touch

AnswerMiner > Data science  > Data analysis – it’s a kind of magic

Data analysis – it’s a kind of magic

In this post we are focusing on data analysis and visualization which are not coming from the devil, as many leaders of organization thinking. We believe the data analysis and visualization can be understandable, discoverable, manageable for an average person.


First question

Why people do not trust in computer analysis and visualization?

The question is not that simple, because people overrule, overthink, misunderstood the computed datas of the computer, while they commit their lives to a medical device without question (for example – the values of the EKG)

Second question

What stands behind this?

In our opinion, the old, outdated (no longer updated), ugly and useless softwares have a big part in it. These softwares, user interfaces made for professionals (mathematician), which ones seem useless for an average person.This causes many of thoughts like: he is not enough skilled for it, this is not his job.  On the contrary, the analysis process has to be done by the owner of the data, who also know and understands it.

Because if someone else do the analysis instead of the data owner (for example the analytical department of the organazition) the results can be different or irrelevant.

This trend is changing nowadays, the number of newer, more modern visualization tools are increasing on the market, but people hardly change, and they tend to rely on their bad experiences.

In this section we describe some typical problem and we suggest solutions for them


Many people if get a chart or datapoint which prove a certain thing, they think the problem/notion is already solved.

For instance:

There is a company, which has suppliers. In the supplier/delay chart can be seen that  the delivery of the  biggest suppliers are regularly delayed. Here we tend to believe, this is the reason of the fact we cannot sell more product. But if we approach it from a  different viewpoint, for example the product/trend chart, we can get the result our product has less demand than previously.


It worth explore our data in different views or globally. If we investigate it in just one way, not sure we obtain the correct result, beacuse we just see it one perspective, and the relations in the background are also matters.


Many people are not able to abstract from their thoughts. If someone else says different (in this case the computer), they believe it is definitely wrong.

For instance

Last couple of months demand of our product shows a growth tend. In the recent month, it decreased, but the market has not changed at all. Here we can declare that is impossible, we measure poorly, something is wrong. Although it is possible the growth tend caused by some kind of external reason, it was not the result of our work.


We should try to verify our datas with facts, relations, we must not rely just only our intuition, let us be open minded


If we scan our datas in many aspects, we might find cases are not closely connected to our tasks/problems.  

For instance

We  have a product, and we want to see our income in segment basis. If we have to many segmentations  sooner or later there will be a significant outlier which caused by a random fact.


We should focus on our tasks/problems to be solved, investigate the circumstances, what causes and effects can be possible between them, and filter out the outliers.

Bad habit

If we tried some data analysis and visualization software before and we found that useless for us, we can not be sure about there is no right software in the market for us.

For instance

Once we tried a data analysis software and we could not load our data because the UI was so difficult.


We shoud try more data analysis software, because they have very different features and we might  find a right one for us.


Why should everyone have to use data analytics and data visualization during their work?

Nowadays, everything is turned into data, everything much more related then the last century.  Almost everything connected to each other. (like the internet)

Data mining, data digitalization are a lot easier achiveable nowadays. We can definitely say that our organization has to be data driven.

Because behind our datasets there are a lot of ways to optimize our success or increase income. Be data driven!

No Comments

Sorry, the comment form is closed at this time.

Get special deals and up to date content.