5 Factors of High Quality Data & How They Affect Business Decisions

Data is collected in most businesses and is often thought of as record-keeping. When an inspection is completed, an employee’s performance is reviewed, maintenance is recorded, or even when a safety meeting is conducted, data is being collected. The record is then usually kept for future reference in order to achieve a greater objective, such as making better business decisions.

iStock_000032631836Medium-1There are five factors that determine the quality of data, and when each of these components is properly satisfied, the result is high quality data. When data is of high quality, it is more effective at driving greater organizational success because of the reliance on fact-based decisions, instead of habit and human intuition.

Five Factors of High Quality Data

  1. Completeness
  2. Consistency
  3. Accuracy
  4. Validity
  5. Timeliness

We will discuss each factor of data quality, and provide examples about how each can be improved to have the biggest positive impact on your business.

1. Data Completeness

Data completeness refers to whether there are any gaps in the data from what was expected to be collected, and what was actually collected.

  • Example: An inspection is done on a vehicle and the inspector accidentally does not indicate the current hour meter reading on the vehicle, which is a required field for that inspection. This has rendered the inspection incomplete and less valuable because important information is left out.

How this can be solved:
The problem of incomplete data can be resolved by ensuring that the data cannot be submitted, unless all expected data is present. This is exceptionally difficult with paper because it is prone to human error.

Paperless data collection, on the other hand, uses smartphone and tablets and allows companies to gather the same information that their paper forms do, but data is recorded on a supporting device instead of on paper. With smartphone and tablet-based data collection, data completeness can be easily achieved with a function called mandatory fields.

When a necessary field such as “Hour Meter Reading” is mandatory, the person completing the inspection will be not be able to finish and submit the data without the mandatory field being filled. This ensures that data is complete when it is gathered and results in less time wasted fixing mistakes revolving around incomplete data.

2. Data Consistency

Data consistency refers to whether the types of data align with the expected versions of the data that should be coming in. This may seem similar to data completeness, and while they can be similar, they are also quite different.

  • Example: Referring back to the same example as above with the vehicle inspection, if the vehicle operator misspells the type of vehicle, the data is not consistent with what is expected to be written and can create problems further down the road. For example, this limits searching functionality once the data is entered into a computer, because a search function that pulls up a vehicle’s history will now exclude the latest inspection because of the spelling mistake.

How this can be solved:
In order to ensure that all data is consistently collected in the expected format, drop down menus can be utilized in a mobile data collection app, so that rather than writing in free form, the operator only has a predetermined number of options from which to choose. This ensures that all data is consistent and allows for proper accurate event and asset history and complete search results.

3. Data Accuracy

Data accuracy refers to whether the collected data is correct, and accurately represents what it should.

  • Example: Sticking with the vehicle inspection theme, if the operator performs the inspection and puts in a value for every field, and spells the type of vehicle correctly and records the correct units of measurements, he or she has complete and consistent data. However, on the same inspection, if the operator records the mileage at 40,000 miles instead of 60,000 miles, this is inaccurate data, resulting in misinformation and related issues.

How this can be solved:
Data accuracy can be slightly more challenging to fix than completeness and consistency, and accurate data is often the result of proper training and competent employees. However, in order to minimize human error as much as possible, sometimes it is necessary to implement extra measures, even with great employees.

In order to ensure that all data is accurate, additional, yet quick steps can be added to minimize or eliminate inaccuracies. By adding elements to the data collection process, such as picture capture and GPS location and time stamp to recorded events, the likelihood of inaccuracy is reduced. For example, if the operator was required to record the mileage during an inspection, but also take a picture of the odometer, this inaccuracy would be identified and rectified

4. Data Validity

Data validity can be a bit trickier than the previous examples, and fixing invalid data often means that there is an issue with a process rather than a result. Validity of data is determined by whether the data measure that which it is intended to measure.

  • Example: One challenge with paper forms is the rigidity of them, because they are a hassle to change. Often new information needs to be included in order for the data to remain valid, in that it accurately measures what it is supposed to. When new information is needed but forms don’t get changed, the data is no longer valid because it does not properly measure what it is supposed to.

How this can be solved:
As we mentioned, issues of validity are often derived from processes, instead of the final result of the processes. Paper-based processes make issues of invalid data more difficult to change because changing forms can be costly, wasteful, and the more widespread a company is, the harder the change. With paperless data collection, there are no physical pieces of paper since all data is collected on a smartphone or tablet, so changes to forms take seconds and are immediately implemented company-wide. There are no old sheets to throw out or extra printing costs, and certainly no one left using an old version of the form.

5. Data Timeliness

Data timeliness refers to the expectation of when data should be received in order for the information to be used effectively. The expectation and reality often do not align, leading to ineffective use of the data and a lack of data-driven decisions.

  • Example: When data is collected on paper in the field, there is a significant lag between when the data is collected to when it is used to drive informed decision making. If a vehicle is inspected and has been determined to need maintenance or repairs, and this information is recorded on paper, often it can be days before the information is submitted to the right person, inputted into a computer and ultimately received by the workers in the shop.

How this can be solved:
Anything slower than real-time data is becoming an increasingly inadequate source of information in most industries. By utilizing real-time data and timely analytics, companies can make more effective decisions. Real-time data allows for personnel to collect the same data that they would on paper but rather than being collected on forms, it is recorded on a smartphone or tablet and instantly submitted into a cloud database as soon as the task is completed. This completely eliminates the time lag between when a form is completed in the field and when it is received by the relevant person, leading to timely, informed decisions.

Conclusion

High quality data is determined by optimizing the completeness, consistency, accuracy, validity, and timeliness of the data collected. By following the best practices of ensuring high quality data, companies can improve their operational processes and organizational visibility through informed, data-driven decisions.