The Importance of Good Data
Business practices, theories, and applicable tools required to succeed in the “real world” are academically learned with the expectation that any student should be able to practically apply this knowledge.
This assumption has a key flaw: the requirement of good data and the impact of using bad data. In this post we’ll be exploring why good data matters to business and the detrimental effects of harnessing bad or inaccurate data.
Why Good Data Matters
This assumption may seem obvious, blatant even, but it remains at the heart of a lot of difficult issues surrounding businesses and individuals.
Car-Buyer Cognition Example
Most are uneasy buying a second hand car as their data is usually incomplete. You may think you’re getting a “great deal” but it could have a worn bottom or busted brakes. Only the person selling the car knows every detail.
The buyer will rarely have enough accurate data to make a confident used-car buying decision. A licenced dealer will have this data, making the purchase far easier.
Essentially, having access to the full scope of data available and having the ability to process it makes us far more assured with our decision-making. In business, this data is liberating.
According to IBM, more data has been created in the past ten years than in the preceding 5,000. At this rate processing and managing good data is already essential. Before long, there’ll be a miasma of information to decipher and not nearly enough time to do it.
Effect on Business
It’s essential that businesses use both money and time as effectively and efficiently as possible and this can only be done with good, reliable data.
Incomplete, skewed or faulty data can lead to wasted funds used on poor marketing campaigns, mistargeting, low quality leads, and questionable decision making.
Even more concerning are when important decisions are based on information that is old, useless, or incomplete. According to Harvard Business Review & IBM; bad data costs the United States of America $3.1 Trillion per year; punctuating the need for data control.
Data’s importance is unquestionable and as unpopular as it may seem, it’s better to use logic or a gut feeling than rely on incomplete data which will lead decision-makers astray.
Maintaining Good Data
Accepting the importance of data and knowing how to acquire and maintain it are different things. If obtaining good data was easy, the industry wouldn’t be worth $138bn per year.
Having systems and procedures in place to input, retain, maintain, and run manual data quality checks is key. Clearly defining data function roles is one example of a proper system.
Ensuring every team member knows what their job in the data management process creates a strong precedent.
Establishing ownership of data throughout all phases and departments is another assurance of good data.
Make sure that team members are held accountable for data changes or that a board or group must approve all major changes to the data or data management.
Additionally, ensuring that agreed upon metrics are followed will help ensure data quality is kept up to par with what is required for the business.
If staff know what good data looks like or what it should do they have a specific goal to meet which can be set as a bar for performance.
Finally, having team members trained in the use of data, and having the proper security systems to protect your data are imperative steps to acquiring and keeping your company’s data needs ahead of the curve.
As important as data is and despite its many potential uses, data does have limits to what it can do.
Data is a tool essential to 21st Century Business but it is not to be confused with either understanding nor wisdom.
Data Analysis Tip
A dataset can be analyzed dozens of times over but it won’t reveal any new truths if the analyst continues to look at it with the same mind set.
Having good data can be an imperative part of understanding a situation or knowing what the wise decision is. Even the best information, supported by solid data can lead you astray.
Similar to how the quality of the hammer does not make the construction worker, the quality of the data does not necessarily make an analyst or a particularly smart decision.