Information and Data, two sides of the same coin?
Semantics?
Recently, CFO argued with my team about the difference between information and data. His exact words were “they are the same thing” to which I politely said, "No, they are two very different things", after which he said “…semantics”. I have really never understood that answer “semantics” to imply that two things are the same. Semantics as a word collects the vast array of different values and meanings words can have, how can it be used to describe the opposite?
The Basics
Let’s imagine that you and I are asked to think about the same thing, a car or even better, a mobile phone. It is almost guaranteed that what you and I think of when prompted by the term “mobile phone” are going to be remarkably similar. This is simple to understand, both you and I have been exposed for a long time to mobile phones, we both understand what they are today and with a bit of luck you are old enough to agree with me on where they came from since the early nineties.
Exposure to something makes it familiar and familiarity turns into intuitive knowledge where we don’t really need to think much to imagine the thing represented by the words “mobile phone”. The way this works is we associate the words (an abstraction) with the thing (an implementation).The issue with this is that there are many things, concepts that we may not both be so familiar with, sometimes everyday concepts that are not really implemented physically. These are abstract and are harder to agree on as a single definition.
Information and Language
When we think about our language, the one we understand and use everyday, we simply “know” it, we know that certain sentences with nouns, verbs and adjectives will be understood immediately by the person we speak with, provided they understand the same language. We take this for granted. This is because in our language (and any language) we use words that can be nouns (things), verbs (events) and adjectives (attributes or characteristics of the things object or subject to those events).
Because we all know the rules and the meaning of the words, we can communicate specific ideas with the assurance that the other person will understand the information contained in the communication. The words represent something we both agree on.
All words are made of letters that combine to form the familiar words. As such, letters and words are building blocks for sentences, and these words and sentences encapsulate semantic value: meaning. Now imagine that we recombine those letters into words in a different language, say Spanish. Even if the same building blocks are used, we end up with sentences that, unless we know Spanish, will not really mean anything to us. Yes, we recognise the letters but the words and the entire sentence does not mean anything to us.
In this foreign language, the letters are building blocks, the words are building blocks but the complete picture is simply not there. We can say that the data is there but there is no information for us, even if the sentence in Spanish is meant to say the same thing had it been written in English instead.
So a sentence with words is data, in any language we like but information is only there if we can give meaning to the sentence. This is the difference. So far this is not rocket science, but why is it so hard to wrap our heads around it when we are using the same language that we are all meant to understand?
Metadata
Numbers can also be combined to represent stuff. Take 12544604501 for example. Simple enough, just numbers. Do you know what 12544604501 represents? You can take a guess and say it is a US telephone number (which it is) but what if we are discussing the serial numbers of defective equipment? What tells you or me what that number represents.
12544604501 is data. GlobalStar Telephone number: +1 (254) 460-4501 is information made of data and context, metadata, without which the number is useless for any practical purpose.
Glossary
What is then needed to establish a language and the possibilities of information it can carry as opposed to data (I.e. written in a foreign language?). Languages have rules, vocabulary, grammar that we learn as children, and it takes time, but the rules are all around us, we are immersed in them from birth. What happens when we are expected to know a different language without having learnt it first?
This “new” language is not a regional language but one specific to a single business, a language that has evolved gradually to mean something but that in reality, in many organisations is undocumented.
Imagine you take a new job at a company and you come from one of their competitors. You’ll think “sure, I used to work for Vodafone, I know this”. And it will be true to a degree. However, picture this: you get sent an excel document that is commonly used in your new company, one that has two columns, one with a number as above and another one with a company name. Now you don’t really know what the document really says. So you ask for the document with column headings and you receive the document with two columns, one titled CRM Phone Number and another one called Customer. The question is: what is a customer? You may think you know (we all think we do) but is a customer:
- A company that placed an order for our product/service?
- A company that we sold something to?
- A company that we actually ship something to?
- An entry in the CRM system?
Quite often, depending on who you ask, you will get a slightly different version of what a customer is. Why is this, shouldn’t a customer just be a customer? The answer is no, each organisation will have their own internal glossary if they have one, else the concept will be a different thing to different colleagues at your new role. And this is a simple example, wait until you need a definition of service and its components from a margin accounting perspective.
The reason for this is that abstract concepts don’t have implementations, and most business terms are in fact purely abstract. Without a typified definition, it is impossible to paint an accurate picture of reality. And increasingly reality is the digital reflection of the world, and not the world itself.
Glossaries go a long way to achieve this disambiguation, which is why terms like “single source of truth” or more to the example above, “Single Customer View” are permanent obsessions at every organisation. The issue with this simplistic view (everywhere you find the word “single”) is that it only works at mid to top management level. The problem is that what something means has to be combined with what each group of stakeholders needs to know about the thing in question. This is the Achilles heel of anything “single…”.
For this reason, glossaries must be extended to contain multiple “trusted” views of the concept, in such a way that every perspective can be self-contained. For the “customer” organisations should aim at producing several “trusted customer views”. After all, what the customer themselves (as one of the groups of stakeholders) and what engineering need to see are two different perspectives of the same reality. This can be referred to as Customer-360, another popular term, but this is applicable not only to customers of course, but to everything that is an information concept for your business
Business Information Model-BIM--with AVANZ.IO
So before your company rushes to buy an information system to solve a problem, some groundwork needs to be done. The idea is to create a graphical information model (not a data model) that the business, everyone, can find themselves in. This model reflects “what business are we in” and should be developed following the famous Toyota Way, by asking “why” (and what, how, where, when, who) as the model is developed into deeper more granular layers all the way to the bottom.
A BIM is the single most valuable artefact a company can create for several reasons. Our previous post in this blog dealt with the chain formed by:
People > Process > Information > Systems > Data
and information is the central link to this chain. So a BIM backed with a glossary with multiple trusted perspectives provides clarity to the business (people and process) but crucially allows you to ask IT to build or configure any system based on your information rules.
People in your organisation will know their subsets of processes, perhaps even cross-team processes with hand-off points if process mapping has been done before, but the piece that is often missing is the BIM that translates the reality of the processes with the language used to capture the details and subsequently translate them to IT specifications based on unambiguous meaning and rules.
This is what we do,
AVANZ.IO


