PolyRated

PolyRated

PolyRated means that a product has been given a score out of 1000 to indicate the level of market sentiment for it - the higher the score the more people like it.

Nonaligned Ratings

You are probably familiar with the 5 star ratings you see on many retail websites that indicate how happy shoppers are with a particular product. When a product has been PolyRated the score takes into consideration the ratings from many places and sources to provide a more accurate rating than any individual store can offer due to the limited pool they draw from.

Because PolyRated scores are based on larger sample sizes than individual stores have access to, they can be significantly more reliable. For example, a store might sell a product at a discount to attract new customers but in doing so they may generate higher 5 star ratings than they would have at the original price because they have provided better value for money. On the opposite side of the coin, a store might have had a temporary problem with delayed deliveries and in doing so receive lower ratings than they otherwise would have for the same product.

The Polyphonical Algorithm 'averages out' these and many other biases to produce a rating that can be relied upon for a product regardless of which business you buy it from.

How The Ratings Are Calculated

Trained researchers scour the internet for reviews and ratings about a product and feed that information into the Polyphonical Database. These sources include everything from 5 star ratings on retail and review websites, forum threads where product pros and cons are discussed, video and audio reviews, blog posts and just about anywhere product opinions can be found online.

Part of the researcher's job involves reading information that doesn't have a numerical rating associated with it, like most forum discussions, and assign numerical values according to a check-list.

Once all this information has been collated, the Polyphonical Algorithm processes the data and produces the PolyRated score.

How The Algorithm Works

Some opinions have more predictive value than others and this can be measured by comparing a particular source's previous ratings for a particular product to the ratings provided by every other source for that same product.

By performing this and other similar kinds of analysis, it is possible to assign weightings to individual sources that determine how much influence their new ratings should be given.

The net result is a composite rating that minimizes the impact of incentivised and other types of aberrant reviews.

How Consumers Can Benefit

By comparing the PolyRated scores of similar products you can see which one is held in the highest regard by others and this can help you decide which one to buy the same way you currently do with 5 star ratings at online stores.

How Retailers Can Benefit

By displaying PolyRated scores on product listings, retailers can provide reliable third party product ratings in place of, or in combination with, their own customer ratings. This can be particularly useful for retailers that don't yet have many customer ratings of their own, either site-wide or just for select products. Note that a 5 star equivalent rating can be produced for snippets and other purposes simply by dividing the PolyRated score by 200.

How Review And Comparison Websites Can Benefit

By displaying PolyRated data in comparison articles readers will see that there is rational basis upon which products have been recommended which leads to more trust and higher sales. Just as importantly, using these ratings to select products to recommend is a more cost effective approach than purchasing large numbers of products and reviewing them independently. For websites that prefer to produce hands on reviews, significant savings can be had by using PolyRated scores to narrow down the short-list of products purchased and tested.

Further Information

If you are interested in exploring the possibility of using PolyRated data for your business or website, please send a message using the Contact Form.