Have you ever made a decision based on intuition without relying on objective information? Likely, you have and even do it every day. Have you ever thought that if you hadn’t rushed a decision or if you’d taken into account certain information, you would have done it differently? Again, odds are high your answer is yes.
This also happens with the decisions that companies make. Many times, and at all levels of the organization, decisions are made based on assumptions, conjecture, or intuition rather than on real facts. Worse still, they might be made based on the opinion of the person with the highest hierarchical rank or pay.
Especially in an environment of constant change, uncertainty, and complexity (like our current pandemic), having data available to facilitate quick decisions is critical for continuous adaptation and for survival of an organization.
Organizations need to develop their ability to obtain and use relevant data that provides information-generation, knowledge, and, ultimately, learning for better decision-making.
In this article, rather than getting into types of metrics, indicators, or specific techniques, I want to focus on how organizations can develop this capability.
What kind of data are we talking about?
Data can help us analyze many different areas: our own performance and that of our team, the evolution of the market, our clients, the achievement level of strategic objectives, etc. We need both qualitative data (based on perceptions, observations, or opinions), and quantitative data (numerical and objective) to generate good conversations based on the data.
This is key: people need to be able to generate conversations from and about data. These conversations include different perspectives or ways of reading the data to guide us in effective decision-making. For all this to happen, I recommend focusing our efforts on several areas.
The data (information) we work with should start from the decisions we want to make. This can minimize the temptation to measure only what reinforces the decisions we have already made, or that shows how well we have done (so-called vanity metrics).
Organizations typically act based on the three factors included in the acronym IRACIS: Increase Revenue, Avoid Costs, and Improve Service.
Based on this model, the fundamental decisions to be made are:
- How can we increase income?
- How can we reduce or eliminate costs?
- How can we improve service?
From there, you can use a decision tree to decompose the information out to the level we need. Ask yourselves: what information would we need to make a decision? And, consequently, what do we need to know about how that information breaks down further?
A good data management strategy includes defining the processes for data definition, collection, analysis, and usage, including data quality assurance (and privacy), and the levels of accountability and collaboration throughout the process.
I also want to relate it to the traditional pyramid of knowledge management, DIKW (data, information, knowledge, and wisdom). Although there are some discrepancies on the model’s limits and specific definitions, I wanted to give you my vision from a practical and usable point of view.
The term “data” refers to “raw information”: small elements that give us an objective view of the world in which we live. Data is a collection of “what happened”. Examples of data in an organizational context can be found everywhere. Just think about all that is associated within processes and teams (units of work delivered per unit of time, total process time, waiting times, inventory, etc.) and in regard to the impact on our business (number of customers, sales volume, annual turnover, etc.).
The collection of data has four factors consider:
- Timing. When do we measure?
- Sample Size. Are we going to measure all or only some?
- Automation. Are we looking to automate?
- Quality of the Data. To what extent is the data reliable and credible for the entire organization?
Strive for “one version of the truth”. Data is our source of information and therefore can be interpreted, but it should not be debated. There will be multiple ways to visualize and use the data, but the intent is to do it without corrupting the validity of the data obtained. Today, the widespread use of software tools in digital companies provides us with massive amounts of data. It’s there, whether we use it or not.
Data becomes “information” when we can provide context and connect pieces of data to each other. It is the result of data analysis. The information answers the question of whether what is happening is “good” or not.
Examples of information could be the level of progress on a project with respect to its planning, trends associated with sales or customer satisfaction in relation to strategic objectives, increase or decrease in software errors in relation to improvements made in our development process, etc.
“Knowledge” happens when we can summarize information from our specific situation and apply it to the conclusions of other situations. Knowledge arises when we stop thinking about the past and present and look to the future. For example: specific improvements for a project as a result of progress information, or commercial actions to be carried out for a client based on analysis of client information.
“Wisdom” is when we integrate that knowledge into our actions and turn it into something useful (subject to continuous improvement and refinement). Wisdom is when we use the knowledge gained to make future decisions.
For an organization, having that continuous flow of data, correctly analyzing the data to obtain information, knowing how to use the information to generate knowledge, and using that knowledge to become increasingly wise, is key in a context of adapting to constant change and ensuring organizational growth. We should always ensure consistency and alignment (clear line of sight) from our strategic decisions to our data.
Finally, as we move up the pyramid of data, “machines” have less significance. We go from the most concrete to the most abstract, towards the human capacity to provide meaning and generate wisdom. This is the differential factor. Technology is a must, but it is not nearly enough.
We understand data governance as the set of people, responsibilities, rules, and processes that govern the management of data. Data governance deals with identification, collection, usage, and associated decision making.
Good data governance is especially important to ensure consistency, reliability, privacy, and the ability to compare and group data (between different business units, products, or customers, for example), and involve different perspectives in decision-making (the perspectives of technological, business, strategic, sales, etc.).
In conclusion, developing the ability to adapt to change through data requires working on these three key characteristics:
- What decisions do we need to make?
- How do we manage data, information, knowledge, and learning to make good decisions?
- How do we ensure good data governance?