How should you Optimize your Amazon product titles?

Machine learning is widely used at Amazon. Product search rank is one of its best known usage. To get the best match between users search terms and sellers products, Machine Learning algorithms are crunching tons of data points and gathering as much relevant information as possible.

Some of the inputs of these algorithms are :

Product titles
Product descriptions
Backend keywords
Product sales volume
Number of pictures
Number of reviews
Quality of reviews
Age of reviews
External traffic
Add to cart events

Improving your Amazon products search rank can be achieved by working on one or more of the topics listed above. In this post, we will focus on product titles, as our research clearly demonstrates the impact which good titles can have on organic ranking.

One recurring question most of our customers are asking us is:

‍

How should I build my Amazon products titles ? Which keywords should I include ?

To answer this, let’s look at the performance of our customers’ product titles. The data clearly speaks for itself.

We selected a representative sample of accounts which are running their advertising on Amazon with our technology. We split the accounts included in this experiment into four categories, according to their trailing 12 months sales on Amazon:

CAT1 : Under €300k sales,
CAT2 : Between €300k and €1m sales,
CAT3 : Between €1m and €10m sales,
CAT4 : More than €10m sales.

For each category, we computed the performance of the keywords included in their product titles and compared it to the performance of the keyword which were not included in their product titles.

The results are displayed in the table below.

‍

Note that the amount of sales by keyword is averaged per product and per month to allow relevant comparisons between the 4 categories.

We clearly see that the keywords which are used in the product titles generate a significantly higher amount of sales: at least 50x more sales than the keywords which are not included in the title. And the ACOS is lower, which means better profitability.

One would argue that there is a selection bias in this study. Keywords which are driving more sales are used before hand in product titles. It is not because they are included in the titles that the these keywords are getting more sales. True, and it is very hard to measure exactly the impact of one over the other. What matters most is how to use this correlation to improve your products ranking.

We also run the same analysis over all the accounts, mixing big and small ones. The results are pretty much the same: