Interesting
One of the topics discussed that caught my attention was predictive analytics. Businesses more often use this knowing how effective it can be when predicting sales. It can be anything from text analysis to even forecasting. The weather is a major predictor for sales on a day to day basis and marketers know this using predictive analysis to their advantage. I found this interesting knowing that certain purchases are made more frequently based on whether or not it is a sunny, cloudy or rainy day. Although, not all data is good data, you have to know which data to study which can be difficult to understand.
While reading the chapters, it talked about how while using a bigger sample size, it was less likely for you to get random results. Random results done skew your data as much with a sample size that is much larger. If there is a random result in a smaller size then you have to consider it may not be too random. If there is one random result out of 100 then that is much different then if there was a random result out of a sample size of 5 people.
Chapter 11 more so talked about A/B testing and how it wasn't always the most reliable source of getting good data. The readings talk about how the data can be hectored due to some errors which will lead to poor data results. Many businesses heavily rely on their data testing which can be concerning due to the large area of error that it can lead to. It surprises me that so many companies will use it well knowing that it may not always be the most accurate data. This chapter also discussed correlation and how import that can be when it comes down to actually taking action. Businesses have to be confident with their decisions based off of their data results.
1. How can one know exactly what is good vs what is bad data?
2. What do marketers look at when data contradicts itself?
3. Is it worth it to take risks when not 100% sure if your data is correct?
While reading the chapters, it talked about how while using a bigger sample size, it was less likely for you to get random results. Random results done skew your data as much with a sample size that is much larger. If there is a random result in a smaller size then you have to consider it may not be too random. If there is one random result out of 100 then that is much different then if there was a random result out of a sample size of 5 people.
Chapter 11 more so talked about A/B testing and how it wasn't always the most reliable source of getting good data. The readings talk about how the data can be hectored due to some errors which will lead to poor data results. Many businesses heavily rely on their data testing which can be concerning due to the large area of error that it can lead to. It surprises me that so many companies will use it well knowing that it may not always be the most accurate data. This chapter also discussed correlation and how import that can be when it comes down to actually taking action. Businesses have to be confident with their decisions based off of their data results.
1. How can one know exactly what is good vs what is bad data?
2. What do marketers look at when data contradicts itself?
3. Is it worth it to take risks when not 100% sure if your data is correct?
Comments
Post a Comment