Data & Information in the Time of Knowledge-Based Economy
Information and knowledge are essential assets for sustainable business strategy and operations. Organizations create and continuously evolve ecosystem understanding, enabled by actual data and interpreted to the level of insights. They maintain knowledge as a sustainable commodity in the accelerating world.
Essentially, this article is divided in two parts:
Part 1 — exploration of human understanding and reasoning
Part 2 — consideration of the approach for building ideas enabled by data
Part 1: Exploration
Socrates said, ‘The unexamined life is not worth living.’
But the examined one, is no bargain.
Leonard, Café Society
Continuum of Understanding
People need time to build a subject understanding before making any reasoned decision. According to Harlan Cleveland’s concept we go through several exploration stages to form a point of view.
- Data is an unstructured or structured set of raw facts
- Information is an interpretation of data that has been processed, organised and structured to give it meaning or purpose
- Knowledge is a set of our beliefs that can guide actions
- Wisdom is considered as an ultimate level of understanding
Knowledge Pyramid
Today DIKW hierarchy is also known as a Knowledge Pyramid where each level brings us towards higher understanding and rises its own questions.
- Data: source material of raw facts that needs a context
- Information: Who? What? When? Where?
- Knowledge: How?
- Wisdom: Why?
Example: Financial Statement Interpretation
Let's illustrate how understanding works using example of financial statement interpretation for a public company like Apple Inc.
Data (Raw facts)
According to Consolidated Balance Sheet 10/30/2020 company had:
— Current assets: $143,713 m
— Non-current assets: $180,175 m
— Total assets: $323,888 m
— Data for assets groups
Information (Who? What? When? Where?)
Let's identify and calculate assets with high liquidity (easy to convert into cash with minimal conversion loss):
— Cash and cash equivalents: $38,016 m
— Marketable securities: $52,927 m
— Accounts receivable, net: $16,120 m
— Marketable securities: $100,887 m
So, Apple Inc. high liquidity assets ($ 207,950 m) in 2020 represent 64.2% from total assets amount ($323,888 m).
Knowledge (How?) and Wisdom (Why?)
Using data from Annual Reports on Form 10-K 2018–2021 we may calculate percentage of high liquidity assets within total assets during this period.
Insights with more advanced level of understanding can be generated after additional data enrichment with sources like revenue, manufacturing costs, acquisitions, earnings distribution and taxation approaches.
As example financial experts Marcello Pinto and Rene Ritchie explains such big cash amounts caused by:
- Significant cash revenue from sales and low manufacturing costs
- Low spending at acquisitions due to mostly organic company growth
- Overseas earnings to be hold overseas to avoid US 2nd time taxations
Part 2: Ideas Enabled by Data
… do you happen to know where they go, the ducks, when it gets all frozen over? Do you happen to know, by any chance?
Holden, The Catcher in the Rye
Organizaion Challenges & Customer Needs
PSR Method (Problems Solutions Results) is a framing exercise aimed to help clearly define problems and expected results success criteria for further solutions ideation.
- Define problem as clear as possible
- Define expected result success criteria
- Come up with ideas to bridge the gap
Note: Before solution ideation step we need understand experience journey and identify moments that matters in client/service interactions.
Data Enablement
Data is the fuel for our ideas. When we have an early idea concept we need to validate idea feasibility by answering data related questions:
- What data we have? — quality, accessibility, completeness
- What data we need? — missed data points
- What data we want? — hard access, currently don’t exist
Based on IBM AI Essentials there are three main data types:
- Public data — data that is available to anyone
- Private data — proprietary data owned by the business
- User data — anything owned or generated by users
Example: Personalized Investment Recommendations
Let’s illustrate items above with approach to personalized investment products recommendations. Key moments of retail banking experience journey uncover advisory opportunities in client/service interactions.
Customer expectations are usually guided by the trust level to the service and bank's digital service proposition would be consumed accordingly:
- High Trust: DIFM (Do It For Me) complex tasks like insights
- Medium Trust: DIFM (Do It For Me) low level tasks
- Low Trust: DIY (Do It Yourself), at least let me know on time
Our solution is aimed at investment complexity reduction for the bank client through balancing individual's risk preferences with expected return on investment, convenient portfolio creation, and financial instruments (stocks, bonds, ETF, cryptocurrencies) updates.
Such actionable investment recommendations are able to be generated at the intersect of several data sources:
- Public data — financial indexes, market news
- Private data — investment behaviour patterns, risk level
- User data — income, savings, portfolio, behaviour, revenue goal
Summary
Nowadays world faces huge demand in knowledge and insights both on organizational and individual levels caused by growing complexity of subjects and information overloads.
This demand has laid out a foundation for broad solutions proposition in business and operational intelligence, predictive and prescriptive analytics with actionable insights powered by AI/ML solutions.
Before we try to teach machine of doing anything we need obtain precise understanding of how this task would be performed by a human and only afterwards automate what possible. We should make sure that machine conclusions can be explained to a human for building continuous trust with a digital service.
Literture
- People +AI Guidebook by Google PAIR
- IBM AI Essentials Framework by IBM
- “Information as Resource” by H. Cleveland, The Futurist
- “Hierarchy of Trust: The 5 Experiential Levels of Commitment” by Katie Sherwin, NN/g
- “Scaling pyramid of trust” by Hari Subramanian, Wipro
- “Digital Banking Benchmark” by Deloitte Digital
- “Real-life demo with SoftServe: Turn big data into real action” by SoftServe at Google Cloud OnAir