A statistical analysis of the effectiveness of feedback requesting emails from third party service providers on Amazon.
There are currently a lot of third party service providers on Amazon featuring automated feedback-request email software. While products might slightly differ between each company, their main idea is similar, which is to take on a proactive approach to help build up positive Amazon feedback base for sellers.
How this works…
A lot people are curious and have been asking us just how effective these feedback request emails are. To answer that, we ran regression analyses using our own customer database, hoping to give us a better understanding of its effectiveness.
Our data is based on our Feedback Central users from October of 2014 to June of 2015. We were most anxious to know the effectiveness of the feedback request emails. For the analysis to be more precise, we targeted only the accounts that are using the feedback request feature. Some of the users were only sending product review requests, and some were using Feedback Central to manage feedback as opposed to requesting for more. We excluded those sellers from this analysis as well. We ended up with a sample size of 897.
We decided to first test the relationship between feedback counts and emails sent. The results show very little correlation (0.577), which means increasing feedback and emails sent have little direct relationship.This is intriguing, because what is the point of sending emails if not to get more feedback? We suspect there should be a second external factor influencing feedback count increases.
The obvious culprit was orders. Feedback is left by buyers, so with more buyers, there should be naturally more feedback. We then tested the relationship between feedback counts and orders. We predicted that there should be a positive correlation, as sellers with more orders will just simply have more buyers to leave feedback. As expected, the correlation was much higher (0.892). While this confirms our suspicions, this does not help sellers answer the question: does sending email requests increase amazon feedback counts?
We then decided to do a multiple regression using feedback count, orders, and emails sent. This regression should take into account the major players we want to analyze:
Our regression model is as follows:
Feedback counts = 9.970756 + 0.010634*Emails Sent + 0.047112*Orders
So what does this tell us? This equation, also known as the line of best fit, gives us an estimate of the amount of feedback that a seller is expected to get on average, given the numbers of his orders and emails sent. Let’s quickly look at an example first:
Suppose one of our clients this month has 500 orders and he sent out 250 feedback request emails. Plug these numbers back to the equation and we would expect the seller to have: 500*0.047112+250*0.010634 = 26.2145 feedback count.
To evaluate the effectiveness, the numbers that we should be looking at are the coefficients, which are the numbers in front of the variables, 0.047112 and 0.010634, and what they tell us:
- If we were to hold emails sent constant, then for every order we get, we are expected to get 0.04711 feedback counts in return.
- If we were to hold orders constant, then for every email sent, we are expected to get 0.010634 feedback counts in return.
To put it simply:
According to the regression…
Putting it this way, some of you might think 94 is a relatively small figure. In other words, sellers who use feedback request emails can generate approximately 22.6% more feedback than without, if they were to send out on average, one email for every order.
While our regression was based on a fairly large sample size, there are still other external factors that could be influencing the results. For example, perhaps buyers of a certain kind of product are just simply more willing to leave feedback. What our findings can do is help sellers decide if sending out emails requesting for more feedback is a necessary action to take. Not all of our clients use Feedback Central to send email requests. Some clients simply use it so their customer support department can manage seller feedback. From speaking with our clients, what we can conclude with confidence is that more feedback is great, but quality positive feedback is the most important to increasing sales.
Try BQool Feedback Software. 14-Days Free Trial Now!
Learn more about:
Published by BQool Marketing on Oct 2, 2015