forget big data
Big Data refers to the collection of large amounts of data or information from traditional and digital sources as a tool for ongoing discovery and analysis. But the term "Big Data” has become ubiquitous as companies, businesses, governments, and marketplaces of all varieties now have the power to capture, collect, and amass their own data sets.
“The ‘Big Data’ revolution is occurring mainly because technology enables firms to gather extremely detailed information from, and propagate knowledge to: their consumers, suppliers, alliance partners, and competitors.”
(Erik Brynjolfsson, MIT Sloan Professor)
A relentless focus on capturing Big Data can lead to a narrow and misleading view in the long run. Most companies endlessly exert themselves collecting every input or data point without considering a strategic approach, without a rhyme or reason, or without a method to their madness.
Companies shouldn't focus on capturing big data; they should focus on collecting the right data.
The key for innovators and data scientists is to understand that the amount of data isn’t the critical factor — having the right types of data available is. The right approach to data should be focused around capturing information that will enhance and empower strategic initiatives. Sometimes the right data can be large-scale, sometimes the right data is limited, but precise. Collecting all the data in the world does no good unless:
- You have the right data being collected.
- You know what you're looking for in the data.
- You understand how you're going to use the data.
- You execute on the data to create insight and action.
The types of data you should be thinking about
Below are twelve unique concepts that help frame data in different ways — or how to think differently about data.
Strategic Data is made up of critical inputs that ultimately create usable information. Strategic Data is the opposite of Big Data, in the sense that Big Data attempts to collect any and all data available, while Strategic Data is designed to collect only the data that will provide meaningful value. Strategic Data is foundational: it should be the starting point for data requirements and data collection. Strategic Data ensures you have the proper data to inform decisions, fulfill goals, and deliver on objectives.
Strategic Data is the foundational starting point to data intelligence.
Clean Data is paramount to data management. If not managed properly, data may be riddled with errors, rendering it useless. Clean Data is a prerequisite to establishing trust in the information that enables you to make decisions from it; Clean, consistent data allows you to be confident in your data. Keeping a vigilant eye on ensuring Clean Data helps mitigate long-term, ill effects. If and when data errors happen, it's important to adjust, augment, or annotate in order to clean the data as quickly as possible.
Clean, consistent, error-free data is confident data.
Diagnostic Data is information that provides insight; data from which you can make decisions. Diagnostic Data (otherwise referred to as "Smart Data") is the data that makes sense in real-world scenarios. It's the difference between “data that gives you an observation” and “data that helps provide insights.” Collecting large amounts of statistics and numbers is of little benefit if the information fails to provide any added intelligence or actionable outcomes.
Diagnostic Data is information that has been analyzed, interpreted, ready to be acted upon.
Fast Data is information that can be quickly collected, analyzed, and translated into strategy or action. Fast data is critical for enabling real-time decision-making. The faster you can use your data to obtain insight, the greater the potential competitive advantage you'll have. Fast Data is crucial to increasing performance analysis of any data-connected ecosystem; businesses are now catching on to the need for fast data capabilities that increase their decision-making confidence.
Fast Data helps decrease decision-making time and increases confidence in decision-making.
Connected Data is data that combines all relevant sources, processes, and people. Connected Data means the data attributes are in a united ecosystem, allowing them to be cross-referenced and interpreted to fit any and all needs. Connected data brings together information that can be leveraged by different stakeholders, for different needs, from different sources. Connected Data empowers different teams to utilize the data in ways that best suit their needs.
Connected Data is the data that bridges divisions, inputs, and outputs.
Descriptive Data (or historical data) provides a look back at what has happened. Descriptive Data's purpose is to provide updates on performance and to tell you what has happened recently. Descriptive Data is typically used for comparison purposes (comparing to previous time frames, segments, or campaign performance).
Descriptive Data tells you what has happened in the past.
Prescriptive Data uses simulation algorithms to help forecast what might happen based on influencing specific variables, and the goal of Prescriptive data is to provide insight and advice on possible outcomes. In other words, prescriptive data attempts to quantify the effect of future decisions based on potential changes in elements. Think of prescriptive data as a tool to help answer, "If we change X, what is the impact to Y?"
Prescriptive Data helps you make informed decisions by forecasting changes and using dynamic elements.
Versatile Data is dynamic data; it can be amplified with other inputs to help create a larger, more cohesive picture. Versatile Data is adjustable, agile, and helps fill in the details of a larger picture or greater story. Versatile Data complements data with other attributes from Connected Data, such as events, variables, and dimensions to enhance the power of the information. The more versatile and enhanced data is, the more powerful it will be.
Versatile Data allows you to make more decisions, in more areas, more often.
Predictive Data analyzes data patterns to project future probability; using existing information sets, predictive data determines patterns to predict future outcomes and trends. Predictive data won't tell you what will happen in the future, but it can help forecast what might happen in the future with an acceptable level of reliability. Predictive Data uses many techniques, including data mining, statistical modeling, regression testing, and machine learning to make forecasts.
Predictive Data enables you to forecast future probability in a closed ecosystem.
Reactive Data is data that can be quickly adjusted or augmented in reaction to unforeseen events. Reactive Data is important because Predictive Data has a fatal flaw: The underlying dependency in Predictive Data is that projections for the future are based on patterns from the past. It’s important to know what underlying dependencies predictive models have, how they operate, how to determine if they are still valid, and — most importantly — how to react if something changes. When behaviors, patterns, or information change radically, the models that were used for predictions may no longer be valid, which is where having Reactive Data is of utmost importance.
Reactive Data allows you to quickly adapt to unforeseen circumstances.
Personalized Data doesn't just focus on digital experiences, marketing, or ecommerce activities. Data intelligence needs to be personalized, tailored for differing uses — whether targeting specific audiences via marketing platforms or enhancing business decisions in a quarterly review. Personalized data helps arm decision makers with the information they need to make data-driven decisions. Different data will mean different things to different people in different positions, and different people will have different data needs — this is where Personalized Data is imperative. The power of data personalization cannot be ignored.
Personalized Data is the difference between buying a suit off the rack and having one custom-tailored for you.
Actionable Data enhances strategic decisions that lead to specific outcomes. Capturing data for the sake of capturing data is useless and fruitless: Unless data intelligence is constantly used for pioneering strategic insights that drive opportunity and action, you might as well not be collecting the data in the first place.
Actionable Data empowers and enables you to make strategic decisions that ultimately drive to action.
thinking differently about data:
Turning Data into action
Data intelligence is not about the amount of data being collected, it’s about having the right data — about how that data is analyzed to create insight, how that insight is utilized to create strategy, and ultimately, how that strategy is applied to execution. Having the right data is important because data is useless if you can't do anything with it; focusing on action is important because data is useless if you don't do anything with it.
collecting Data is useless if you can't do anything with it, and collecting Data is useless if you don't do anything with it.
Forget big data – Focus on the right data: Data depth, quality, value, detail, consistency, accuracy, and comprehensiveness. By thinking about data differently, you'll be able to leverage data in a multitude ways:
- Strategic Data: Will the data help accomplish goals and enhance data-driven decisions?
- Clean Data: How will the data be consistently clean and free of errors?
- Diagnostic Data: How can this data tell us about what happened in the past?
- Fast Data: Can the data be used on-the-fly to make decisions from it?
- Connected Data: Can the data connect with different sources to complement our data sets?
- Descriptive Data: Does the data being collected provide opportunities for strategic insight?
- Prescriptive Data: If we change elements in the ecosystem around, what will be the impact?
- Versatile Data: Can this data be leveraged across multiple areas and business sections?
- Predictive Data: Can the data be used for forecasting with predictive analytics?
- Reactive Data: Can the data be augmented in reaction to changes in the ecosystem?
- Personalized Data: Is the data being personalized for end users or targeted objectives?
- Actionable Data: Does the data help direct towards meaningful, strategic action?
Capitalizing on the different types of data takes long-term planning, commitment, dedication, and strategic thinking — but the overall goal is simple:
Turn data into insight,
insight into strategy,
and strategy into action.
“It’s difficult to imagine the power that you’re going to have when so many different sorts of data are available.”
- Tim Berners-Lee