QUICK SUMMARY:
Amazon just rolled out a seemingly small, but massively significant change to its ratings. They are now displaying a numeric average star rating in search results. This change appears to still be in the beta phase as the numeric average star ratings are not displayed for all users on all searches across Amazon.com
Take-aways:
- Shopper behaviors will change, favoring selection of products with higher displayed numeric average star ratings.
- Some products will see conversions increase while others will see drops as a function of changed shopping behaviors.
- Advertisers will see changes in conversion rates based on a product’s star rating and may need to allocate spending to focus on higher rated products.
THE DEEPER LOOK:
Historically, both in the search view and on a product listing, you’d find two rating metrics:
- The iconic 5-star image, filled by half-stars
- The total number of global ratings
BEFORE the change:
The search view now displays a numeric average star rating as a third rating metric.
AFTER the change (desktop):
AFTER the change (mobile):
Why does this matter? It seems like a small change, but it is not. This is a massive addition and will significantly alter the landscape of winners and losers on the world’s largest ecommerce marketplace- Amazon.
Follow along with me for a moment and I will explain.
Star ratings directly impact sales, for better or worse. In my book, The Review Cycle, I share my researched principle of Review Elasticity. Like the economic principle of Price Elasticity, Review Elasticity measures change in demand. Instead of measuring that change as a function of price, we evaluate demand as a function of average star rating.
Having nearly a decade of experience managing reviews for thousands of product listings, I have observed the relationship between an average star rating and sales. Here are the trends I have identified.
Principles of Review Elasticity
- Whatever the demand is at around 4.5 to 4.7-stars is considered maximum demand for that product.
- From there, every half-star rating drop results in demand cutting in half:
- So, around the 4-star mark demand is half what it is at 4.5-stars. 3.5-stars is half the demand of 4-stars and so on.
- 1-star average rating has almost zero demand because it is the lowest rating possible to give on the standard 5-star metric.
- From 4.8-stars to 5.0-stars we see a dramatic drop in demand to as low as the 2.5-star demand threshold. This is because consumers will generally believe the ratings to be too good to be true, faked, manipulated or otherwise not trustworthy.
- All of these principles are relative to the ratings of similar products within a category.
Visually, Review Elasticity of demand looks like this:
Figure 2.14, The Review Cycle: Image provided courtesy of Matt R. Vance©, 2022
Figure 2.14 is a “Smooth Star Rating Elasticity Curve”. A smooth star rating elasticity curve is expected to exist on product listings and review profiles where the 5-star image reflects each and every 0.1-star change. Smaller incremental changes in that displayed image result in a smoother elasticity curve. Marketplaces like BestBuy.com and Target.com follow this model. As you can see in the images below, the graph of 5-stars changes with each 0.1-star change in an average rating.
EXAMPLE: Best Buy search results display 0.1-star rating changes in the 5-star ave. rating image.
Not all ecommerce platforms have smooth star rating elasticity curves. Platforms that round displayed star ratings by half stars have a much different distribution of demand. Demand is artificially concentrated at each half-star level and looks more like this:
Figure 2.16, The Review Cycle: Image provided courtesy of Matt R. Vance©, 2022
Rounding average star ratings by half stars changes the principles of Review Elasticity as follows.
Principles of Review Elasticity on a Stepped Curve
- From 4.3 to 4.7-stars is the peak pocket of demand. This is where a 4 ½-star image is displayed.
- As is true for the smooth standard star rating elasticity graph in figure 2.14, demand cuts in half with each half-star drop in a product’s average rating.
- The difference is, there is much less of a curve. The half-life demand change is concentrated at the 0.1-star change where the star rating rounding occurs.
- In the middle of a rounded rating, only marginal demand change occurs with each 0.1-star rating change. For example, changing from 4.4 stars to 4.5 stars will only lift demand a little because no change to the displayed star rating is made with that shift. Either way, the site displays an image of 4 ½ stars.
- It is also true here that a 1-star average rating has almost zero demand.
- A rating below 1 star is still not possible as the lowest rating possible to give on the standard 5-star metric is 1 star.
- Unlike the smooth standard star rating elasticity model, the half-star rating model displays ratings for 4.8, 4.9, and 5 stars with an image of 5 filled stars. What this does is effectively drop sales as low as that 2.5-star demand at the 4.8-star mark and higher.
- All of these principles are relative to the ratings of similar products within a category.
Amazon displaying a numeric average star rating will smooth out the previously rigid demand curve. That means:
- A 3.8-star product previously benefiting from a displayed 4-star image will see organic sales decrease.
- A 4.2-star product also displaying a 4-star image will see organic sales increase.
- Ad spending ROI for sponsored placements will shift to a smoother curve, favoring products with higher average ratings.
Will Amazon keep displayed numeric average star ratings permanently? Only time will tell. If the net impact for Amazon is more sales, then I would guess this new feature will stay.
Take-aways:
- Shopper behaviors will change, favoring selection of products with higher displayed numeric average star ratings.
- Products with an average rating on the high end of a previously half-star rounded scale will see organic conversions increase as a function of changed shopping behaviors.
- Products with an average rating on the low end of a previously half-star rounded scale will see organic conversions decrease as a function of changed shopping behaviors.
- Some products will see organic conversions increase while others will see drops as a function of changed shopping behaviors.
Recommended actions:
- Measure your current conversion rate and see how your products are now performing.
- Measure your ad-spend efficiency compared to the past and see how it compares.
- Reallocate ad spending on products in your catalog with higher star ratings and stronger conversation rates.
A final thought. Not all is lost if your product is at 3.8-stars and is now negatively affected by this change. Amazon is still displaying the iconic filled star images by half-stars in search results alongside the newly added numeric average rating. Within product listings, Amazon is only displaying the iconic filled star image. This means the addition of the numeric average star rating will not completely smooth out the pre-existing stepped star rating curve. It is likely to be a little bit of a wave, almost a hybrid between the smooth curve and the stepped curve.
REVIEWS.ai is here to help you stay ahead of all curves shaping the #CPGindustry. Schedule a free audit of your product reviews or contact us for a free demo account and start turning your reviews into valuable insights.
Thanks for reading! If you found this article helpful, check out my book, The Review Cycle, for sale on Amazon.
Director of Data & Analytics @ REVIEWS.ai
Author of The Review Cycle
