What does data entry mean, and what does it have to do with scanning and digitization? If you’re digitizing your records, data entry is critical to ensure document accessibility and usefulness: it’s the method of capturing data that is later used to index your digitized files.

Although it’s an important part of your conversion project, data entry is often misunderstood because there’s little information about the topic. Here, we’ll define data entry, describe its importance to your digital conversion project, illustrate methods of classification, and provide a glimpse of our data entry process. 

What Is Data Entry?

Data entry is the process of capturing information for indexing and file identification purposes. It’s the process of extracting, or capturing, important fields (letters, numbers, and/or symbols) through a process known as “keying.” The person who does the keying will systematically input the data needed into a computer system or software so it can be processed and utilized in various ways. 

A keyer is sometimes referred to as a “data entry clerk” or “data entry specialist.” They typically handle a large amount of information that can sometimes be sensitive or confidential data. Some data entry clerks focus only on one type of information while others focus on a variety of records such as student records, medical records, financial data, and more. 

When it comes to data entry, three things matter most: capture, accuracy, and security. It’s essential to have efficient, accurate, and secure data capture since data is the lifeline of many organizations. 

How Data Entry Relates To Your Digital Conversion Project

Data entry is important, especially when it comes to your digital conversion project, because it allows you to index your newly digitized files and easily identify and search for the content you need. For example, if you want to find the sports section of a specific newspaper from January 1987, you can search using the specific fields that were keyed and locate the information within a matter of seconds. 

If your organization has a small number of records, indexing may not be critical. However, if your organization has a lot of data, it might be nearly impossible to find specific records efficiently if you don’t have some form of indexing. If your files are organized and indexed, you can easily search and find any piece of information you need. 

Data Entry Classifications

Now that you have a basic definition of data entry and an overview of how it relates to your digital conversion project, you might be wanting more examples to get a better understanding of the process. We’ll help you solidify your understanding of data entry by showing you the various ways it’s classified. 

1. Simple vs. Complex

The first way to classify data entry is if it’s simple or complex. You can figure out if it’s simple or complex by looking to see if a “key” is presented. A key is a set of instructions or rules stating which data is suitable to be keyed. 

If there’s a key present, then the data entry is simple. This means that there are directions on which fields to key. 

For example, the rules may say to key all the medical record numbers (MRN) so the keyer will only input the MRN on the files. If the key says to input the MRN and insurance claim number, then the keyer will input those two fields of information. Simple data entry is straightforward with clear rules on what to key. 

Complex data entry is when there is no key present. In this case, there are usually a set of rules to follow that aren’t as straightforward as in the “simple” method. Usually, a keyer can input a field of data into the system only if it checks out all the criteria. 

Using the same example from above, let’s assume there’s no key to tell the keyer to capture the MRN or insurance claim number. Instead, there will be a set of conditions for a keyer to follow before they can key a field. For instance, the keyer can only input an MRN if an individual’s birth date is before a certain year, if the individual had a recent visit with the doctor in the last 2 months, and if they live in a specific zip code.

Adding multiple conditions to data entry can make your final data more precise and usable, but you can see how it can become complex very quickly. 

2. Manual vs. Automated

Another method to classify data entry is to determine whether it’s manual or automated. Historically, data entry is done manually by humans, but with technological advances, it’s now sometimes completed using machines and software. Most data entry projects are either manual or automated but some forms are a hybrid of both.

Manual data entry is the traditional method of data entry. When data entry is manual, it’s largely dependent on a person to input the data into a system. Manual data entry heavily relies on the perception of an individual or a group of individuals and their interpretation of what they see and capture.

When data entry is automated, this means that it’s done mostly by a machine and/or software program and doesn’t need a human involved other than to input the capture settings. There may be some form of machine learning involved and the keying is done through algorithms.

Robotic hand doing data entry

3. Capture and Verification Methods

The last way to classify the type of data entry is by the capture and verification methods. There are three main methods: single-pass keying, double-pass keying, and key-key-compare.

Single-pass keying is when one person does the keying for a data set. They generally key one value and only do the keying once. The accuracy for this method is 95%, which means that 1 out of 20 characters is likely to be incorrect. 

To ensure greater accuracy, there is double-pass keying. With double-pass keying, one individual will do the keying twice and then verify it. 

Double-pass keying is the most common method of data entry, which has a 99.95% accuracy rate. However, there’s still room for errors since the keying is executed by one person. Even if the person does the keying twice, they can still misread a field of data two times in a row, which means the entry they’ve created is incorrect. 

As an example, we’ll say the keyer is capturing a student record number (handwritten from a document) that is JS1230038. The keyer may see the number on the first pass and capture “JS1230088” (remember, this is handwritten!). Then they’ll be presented with the number a second time and again, they read and capture it as “JS1230088.” When the values are verified, it’ll seem correct because both of the captured values are the same, even though the original value is different.   

Key-key compare typically involves three separate parties: two parties do the keying and another verifies if the data keyed is accurate. Two individuals will key the same data set separately. After they’re both done, a third person will review and compare the differences. When they see any discrepancies, they’ll review the data and choose the correct entry. 

One reason you may want to choose this type of data entry is if you need the highest level of accuracy. This method ensures the greatest degree of accuracy because it has multiple people working with the same data set – they’re more likely to catch mistakes than using the single-pass and double-pass methods. 

Similar to double-pass keying, key-key-compare has a 99.95% accuracy rate. We get that this may sound odd since we’re saying it’s the highest level of accuracy. The reason double-pass and key-key compare methods are the same accuracy is because they’re technically the same methodology (key data once, key it again, check it) but key-key-compare splits the data between three people, making catching a mis-keyed value more likely. 

Using the student record example again, let’s do it with key-key-compare. The actual student record number is JS1230038. The first keyer types “JS1230088.” The second keyer (a second set of eyes!) captures it as “JS1230038.” The discrepancy is flagged and the verifier checks it out during the compare step. They look at the two values, then go to the data that was presented to the keyers. The verifier decides which value is correct. In most instances, the verifier will choose the correct value. 

The words "personal data" on a computer screen

How Do You Protect My Sensitive Records During Data Entry?

Depending on which industry you’re in, you may have records that contain sensitive data or personally identifiable information (PII). Some examples of PII are social security numbers (SSN), student records, and dates of birth. If there’s protected information, we may use “fractured keying” to randomize sensitive data before the capture phase takes place. Its main purpose is to secure the data so our keyer will have no way of piecing the information together with other “fractures” to know who the data belongs to. After the data is fractured, the keyer will do either single-pass, double-pass, or key-key compare. Once all the keying is completed, the captured fields are reassembled to present a complete record.

BMI Data Entry Process

For your digital conversion project, the data entry component follows the scanning phase, post-image processing steps, and QA checks. It’s done right before the indexing stage because data entry allows information to be pulled for indexing. 

We conduct your data entry based on the scope of work (SOW) that you’ve created with your sales rep. Before getting to the actual keying, our operations team will meet to determine what kind of keying processes your specific project needs. 

First, we have to determine if there’s heavy keying. This means that there are many fields to key because there’s a lot of indexing that needs to be done. Then we’ll determine the level of accuracy your project needs so we can figure out if we need to employ single-pass, double-pass, or key-key-compare.   

Next, we’ll determine if your records have any sensitive information or PII to see if fractured keying is needed. This step also allows us to know which of our keyers have the right security clearance to work with your records. Security is paramount to us: we’re HIPAA self-certified, SOC 2 Type II certified, a CJIS-listed vendor, and are NIST SP(800)-53 compliant. By reviewing what kind of data you have early on in the project process, we can determine the security level of your project and put the appropriate protections in place. 

After reviewing the type of keying needed for your project and the security level, we pass it onto the appropriate data entry clerks to do the keying. Our keyers have 10-20 internal tools at their disposal to help them with the keying process. Once the data entry process is finished, it will be moved into the indexing phase. 

Throughout our data entry process, our goal is to have efficient, accurate, and secure capture. Every step in our process serves to accomplish this goal. 

Data entry is an important process in your digital conversion project that makes indexing possible. It lays the foundation to make your data and records more easily searchable and in turn, more accessible. It’s a crucial step to take to have effective information management. 

Next Steps

Reach out to us today! Click the “Get Your Quote” button below, fill out the form, and we’ll quickly reply to you to discuss your project.

Further Reading

Take a look at some other articles and pages to learn more about digital conversion and protecting your records:

“Quality Assurance & Digital Conversion” covers quality assurance in your digital conversion project. Quality assurance is the step after scanning and before data entry in your conversion project. 

“What Is OCR?” defines optical character recognition, which is another option to consider when you are indexing. 

“The BMI Difference: Security” goes over security measures our company takes to ensure that your data is secure during a digital conversion project. Security is an especially important component of data entry.