Be careful! This kind of five-star review will actually deduct the average score

However, the content of the new review policy does not end here.

Yesterday, Alice was in our seller communication group

I also mentioned this issue to my friends.

That is the change of review weight and algorithm

Based on our sellers’ past experience in evaluation (shuadan)

Five-star and one-star reviews are reflected in the overall rating

It is a simple addition, subtraction, multiplication and division algorithm.

The more five-star reviews there are, the higher the average score of the product will be.

But sellers with good operating foundation know

Even the same five-star reviews

The weight and quality are also different

A five-star review with pictures, detailed and true content, and Helpful

The weight in the algorithm is higher than other general five-star reviews.

▲Although they are all five-star VPs, their weights will vary.

But this weight is implicit.

Sellers cannot directly see the pros and cons of their own reviews

Now after the policy update

The algorithm for overall review star rating has changed

From simple algorithm updates to intelligent machine learning models

A not so good system

Five-star reviews suspected of fake orders

It may even lower the overall rating of the listing.

This is when the article begins

The friends in the group responded with five-star reviews

The reason why the overall score was deducted by 2 points

As for what this new version of the intelligent algorithm is

What factors are used to determine the weight of the review?

Amazon gives us a little reference on its updated purchase page

The algorithm’s criteria include: Review’s age, whether it is a VP review

And the credibility of the buyer's account

From the quality of the review itself

Alice believes that there may be the following potential factors

1. The content of the review itself: reviews with rich content and pictures

Whether in terms of weight or influencing customer conversion

Or in the system to determine whether the order is fake and other aspects

There are considerable advantages, so whether you are asking for reviews or brushing reviews,

The content must be as substantial as possible

2. HelpFul is difficult to use to manipulate Review sorting rules

However, the bonus to Review weight is still valid.

I suggest you don't order too much at once.

Just like the previous comment, click in batches irregularly

Let’s summarize today’s new policy content

In other words, the Review weight that was invisible and intangible in the past

Now it has been brought to the fore by intelligent algorithms

And also added intelligent judgment

Reviews suspected of fake orders will damage the overall rating.

Just like what Alice said to her friends in the chat group