I am pretty sure that if you are reading this, you’ve done your fair share of online research.
It could have been to find the best selling products, to read reviews of one specific product or to find out how a certain product works.
I have done it numerous times. In fact, I can’t recall a single time I bought something without having done my online research first.
Maybe not with everyday things like food and clothes, although there would have been instances like that.
In this article I will discuss:
The essence of online research
At the end of the day, when you think about it, online research of a product basically means reading up on what heaps of other people are saying about that product.
Product reviews, comments in forums, comparison lists etc, they are all things that somebody somewhere says of a product.
And MOST of the time, people say those things about a certain product because they have had some kind of experience with it.
Maybe they bought it themselves, maybe they have friends and family that have used it.
You will get people voicing their opinions without having ANY personal experience with it, but you quickly see through those.
What I am getting at, is that we inevitably base our decision to buy or to not buy something on what someone else said of it.
And that is PERFECTLY fine.
Who better to trust than someone with personal experience.
Hold this thought for a moment, though, I will circle back to it.
Online product ratings
Many sites that sell products, like Amazon, have product ratings listed with the item. Most of these platforms will have these ratings in a form where two variables are given.
The rating
The first is the rating of the product. That could be a percentage, a number of stars out of 5 etc. So it tells anyone how good a product is on a scale.
Sometimes these ratings are an average of several sub category ratings, combined and averaged as a total rating.
These sub ratings can include parameters like affordability, ease of use, power consumption and so on.
Usually the more authoritative the ecommerce platform, the more extensive these ratings and the more sub rating categories they have.
It’s a matter of being thorough and covering all the bases so the end user, you and I, can make a decision as informed as possible without actually ever touching the product.
The number of ratings
The second variable these ratings include, most of the time, is the number of people that have rated that product. The final rating you see is then an average of all the ratings of all these people that have rated that specific product.
The statistics
Remember that thought I asked you to park, about basing our decisions on what others say and trusting the people that have had personal experience with a product?
It boils down to the following: The more people that give a product a rating as discussed above, the more accurate the overall rating is.
It’s statistics.
The outlier ratings are drowned out by the sheer number of other more honest and objective ones.
Let me explain.
The collection of these ratings can be represented by a classical bell curve.
For you who don’t know, a bell curve, also called the normal distribution, is the statistical range into which all data points from a normal distribution dataset falls. For example, an adult’s height.
Take a look at the graph below.
If we plot the number of people with certain heights on a graph, we will get the most people in the center of the bell curve, because that is the average.
As you go either way out from the center of the graph towards the edges, you will get less and less people per height group.
On the fringes you will get the lowest number of people. They will be the outliers, either people that are well above average tall or well below average short.
The authority of a rating
As pointed out before, all of the above just means that the trustworthiness of a rating is defined by how many people’s ratings are incorporated into that final rating.
In other words, the more ratings a product has, the more accurate and representational that rating is.
Let’s illustrate the principle with an example .
You can have two products in the same category, like milk frothers, both with an average rating of 4 out of 5 stars. Milk frother A, though, has been rated by 3 people, and milk frother B has been rated by 300. What product, then, has the better rating? Milk frother B.
It’s just statistics.
So if you want to find ratings that carry more authority, you want to find large numbers of ratings. And to find that, you should look at ecommerce sites with authority. Sites like Amazon. Huge numbers of buyers, products and last but not least, ratings.
I have to add, a huge number of ratings for one product is not necessarily a huge number for the other. A $20 handheld milk frother will most likely have more ratings than a $5000 induction cooktop. Simply because more people would have bought the milk frother.
The other factor to take into account is that new products will obviously have less ratings.
The best seller lists
This brings us to the next point: what is my aim with these bestseller lists.
Well, taking all the above into account, it makes sense to use these ratings, and these numbers of ratings, to determine what people think of certain products.
Other people have done it, researched all the products and read all the reviews and checked out the ratings. And then they bought what seemed to be the better product. They then probably had the opportunity to rate that product as well.
And so the number of ratings of good products went up and for bad products stayed the same because the better products get bought more. And the average rating for the good products went up and the ratings for the bad products went down.
So you have all these data sitting out there, all over the internet, which are all collectively used by people like you and I, to make informed purchasing decisions.
The KSR rating
To incorporate all these data into one single rating for a product, I have come up with the Kitchen Stuff Reviewed (KSR) rating.
I will calculate this value by taking into account the rating of a product, the number of ratings for a product and how they rank on certain ecommerce sites.
Also, the price of a product will be taken into account to make provision for the principle that more expensive items will be sold less, so less ratings.
Newer products will logically also have less ratings in general but we have made provision for that as well.
The scale of the KSR rating will go from zero upwards towards 5.
It is important to note that I have allowed for exceptional products to go above 5, so there is no maximum KSR value.
The higher the KSR value, the harder it is for a product to achieve it.
Conclusion
My lists will have 7 products, all viewed as best sellers with KSR values to compare them.
I will update those lists continuously to provide for new ratings and product rankings over time.
As for the calculation of the KSR rating, it is still dynamic and as time goes on I will update the formulas to keep up with current trends and improve on its accuracy.
I will elaborate more on how it is calculated in a follow up article.
If you have any suggestions on other factors to consider when calculating the KSR value, or how to improve its accuracy, please let us know here!