As the most basic part of matter we know atoms, quarks some nitpickers would say . These can organize into molecules, via energy. These molecules can organize further, by other energy, to become minerals. These minerals can become rocks. However, to get metallurgical value from that last material, it will need to be separated chemically, using other energy.
As the most basic part of knowledge, digital, we know bits. These, on hardware media, materialized energy, become data. With a pinch of energy and some software, or code, they can become information. For example by aggregating them by month or zip code. Or by making them descriptive statistics such as mean, variance, graphs, etc. This information can become knowledge if you add some more electrical energy and some animal energy, either by energy condensed into a discipline, statistical inference, or by the mental energy that the statistical consultant will employ. This additional input allows signal and noise to be separated from that information. By adding some more energy, that signal can drive decisions.
You have just seen a description of the pyramid of knowledge. The broad base, the trapezoid, obviously contains a lot of data, which must be sifted into signal and noise. The geometric base also results in the base of the dai data, which can take more or less organized forms: .csv files, Excel, Google spreadsheets, MySQL servers, Postgre servers, data warehouses (data warehouse), data lakes (data lake), etc. The last two involve medium and large companies more than anything else.
Data without treatment is worthless, as is oil, and remains a cost and/or opportunity cost .
Data examples:
- potential customer arriving from region X
- customer arriving from region Y
- supplier invoice W
- customer White purchasing alpha product
Sample information:
- 5% of customers are from region X, while 35% are from region Y
- seasonality of supply
- discrepancy between place of residence and origin of customers
- customers who purchased the alpha product, in 5 years never reported failures, malfunctions
Examples of knowledge and decisions:
- most customers come from Y because that is where most of the companies that need my product are concentrated -> I decide to create an “essential kit” for the target type of company, I create sponsored in region Y
- seasonality of supply can lead to agreements with the supplier that provides small discounts because of the discount rate
- the discrepancy sheds light on the customer’s purchase path: those customers emigrated decades ago and want to buy products that remind them of their homeland, their affections -> stronger communication angle in the case of sponsorships, discovery of a niche in the market
- the alpha product does not suffer from clockwork defects, a consequence of its quality -> it can be specified in the sponsorships as a differentiating element, or propose an out-of-warranty insurance at a symbolic price, since the costs of the premium, company-side, will be low
With these examples you saw the difference between data and information, and likely situations of digital transformations that led to data-supported decisions. More precisely processed data, since, precisely, without energy no transformations take place. Who can do these transformations? Unfortunately, since 2015 I have heard too many “synonyms” to refer to this in some cases mythological figure. I give you a sadly non-exhaustive list: computational mathematician, computational statistician, data miner, data scientist, business scientist, BI consultant, data analyst, research scientist, quantitative methods expert, mathematical engineer, numerical analysis professional
I have identified Enterprise Statistics as a statistical consultancy, with all its limitations. If you are interested in seeing this kind of transformation in your company, we can get to know each other in a free call.