Data Collection

Digital Marketing - Study Notes:

What is data collection?

Essentially, data collection is the process of gathering and measuring information on targeted variables in an established systematic fashion, which then enables you to answer relevant questions and evaluate outcomes. By following best practices for data collection, you can ensure the trustworthiness and better use of the collected data.

Qualitative data

Qualitative data is effectively a depth of insights that are captured through methodologies such as interviews, for example. You should apply qualitative approaches when looking for deeper insights on already established trends or statistics, or to fill gaps in data collection that can't be filled by quantitative methods.

Common qualitative data collection techniques include:

• Interviews
• Focus groups
• Observation
• Self-study
• Ethnography
• Action research

Quantitative data

Quantitative analysis is looking at pure metric data to unearth insights. This can be anything from thinking about your management information that you have with the core of your organization. Sales figures or the number of complaints are good examples of quantitative measurement.

You can also look at data in terms of the customer journey from a quantitative point of view. Think about possible bottlenecks that exist within your customer journey and examine the metrics that show how many people ‘drop off’ at different stages to try to improve the journey. This is a very useful and unbiased way to use quantitative metrics to drive business forward.

Common quantitative data collection techniques include:

• Questionnaires
• Scales
• Physiological measurement

Recent trends in data collection

We will now take a look at some of the recent or current trends around data collection. You can collect data in a number of ways, but some of the more conventional ways can be somewhat inefficient and non-responsive from a customer perspective.

Some of the more recent trends in data collection include:

• Web analytics
• Mobile market research methods
• Biometric market research techniques
• Prediction markets
• Virtual shopping
• Social listening
• The Internet of Things (IOT)

It’s no longer enough to simply deliver a survey through someone’s mailbox and then expect that person to complete and return it to you. You need to think carefully about what type of data you are looking for and at what point in time you capture it.

Data-collection considerations

When collecting data, you need to consider:

• Sample size: You need to ensure that you have the appropriate sample size. If the sample size is wrong, the resulting data isn’t reliable. For example, if the sample is too small, then some of the key insights may be missed as we have made decisions on data that is not representative of the size of the issue we are trying to analyze.
• Operational issues: Occasionally there is a difference between putting a plan down on page and implementing that plan in reality. It is important to develop ways to overcome these operational issues. You need to be clear at the beginning of the process exactly who you’re looking to target.
• Accuracy: The accuracy of the information is very important. If data is inaccurate it can lead to difficulties for the analysts analyzing the data and they can derive insights that are inaccurate since the starting point data wasn’t accurate to begin with.
• Bias: Always remember that all data collection has bias. Therefore, try to be aware of potential biases in your collection and overcome them as best as possible. Simply being aware of bias in data can help you to understand the data better, particularly when it does not align with predicted results.

Good data

In general, good data is:

• Accurate
• Complete
• Consistent
• Unique
• Timely

Accuracy and completeness are essential. If there are big gaps in the data, you won’t be able to draw big insights. You also need a certain consistency within your dataset. To draw insights, you want the data to be unique. In other words, you want it to be something that nobody else knows about. So you need to think about how you can capture things that you think only you have access to. And finally, as previously mentioned, good data needs to be relevant and time bound.

Bad data

In general, bad data is:

• Inaccurate
• Conflicting
• Irrelevant
• Misleading
• Outdated
• Incomplete

If the data is inaccurate, the insights based on that data will also be inaccurate. If the data is conflicting, your dataset will tell you different things, leading to confusion. If the data is irrelevant, you are simply wasting your time analyzing it. Remember to capture only what you need! Only ask for things that will be important to the questions you are seeking to answer. Outdated data can offer limited and often irrelevant details, again leading to confusion and inaccuracies. Finally, if the dataset is incomplete, it will be difficult to discern correct or reliable insights from it.

Sources of bad data

Some common sources of bad data quality include:

• Input data
• Processing
• Data identification
• Information integration
• Usage
• Aging
• Organizational quality

Let’s take a look at two common sources of bad data. How is the data being input in the first place? What are the ways in which that data could start to turn bad? At what point do you need to start to refresh that data, to ask the question again? Be careful, however. If you ask too often, you will simply annoy your customers. Conversely, if you ask too infrequently, your data may become irrelevant or too dated. Make sure that you understand the importance of getting the data right the first time.

Another consideration is process quality. You need to have a robust way of being able to process data and identify when the data is going wrong. Also consider where you store the data. You may have captured a wealth of data in one system but if it doesn’t translate onto another you will encounter problems. Therefore, the usefulness of that data is compromised.