Listing Quality Analysis
Introduction
The Listing Quality Analysis (LQA) datasets are designed to help you optimize your Amazon and Walmart listings. The datasets provide an in-depth analysis of key criteria that impact the quality and performance of your listings, including title, description, bullet points, and media criteria. With these datasets, you can identify areas for improvement and make data-driven decisions to enhance the visibility and sales potential of your products.
Considerations
This article details new LQA datasets. The following tables are independent material:
PRODUCT_LISTING_QUALITY_SCORE
PRODUCT_LISTING_QUALITY_SCORE_BREAKDOWN
Glossary
Criteria
Criteria are specific characteristics that can be observed in a listing, such as the number of images or the length of the title. Every day, we analyze your tracked products and extract the data corresponding to each of these criteria. This extracted data is the observed content for a specific criterion.
Rules
Every day, we compare your products against a set of Amazon guidelines and recommendations that we call Rules. They are sourced from Amazon’s official documentation. Walmart will be supported in the future.
Listing Quality Score (LQS)
The LQS is a default score that we compute based on the LQA rules, using weights assigned to each rule according to Amazon guidelines. The more important a rule is for the quality of your listing, the greater its weight.
This score is based on an in-house weighting system; we provide it as a starting point. You can build your score using all the available data.
Section
A section is a basic content element of your listing. We analyze four sections: Title, Description, Bullet Points, and Media.
Business Needs
This section details how to use the LQA datasets to:
- Diagnose the quality of your listing content
- Rank higher in search results
- Convert more users to purchase
- Diagnose the quality of your competition’s listing content
Content
With these LQA datasets, you will be able to improve the quality of your listings’ content.
- Monitor the evolution of your listing quality over time to refine your listings with the historicized and daily updated datasets. Narrow down the listings that require improvements, implement the recommended changes, and track your progress.
- Finetune your listings to be compliant and optimized on Amazon and Walmart with the LQA rules dataset. Leverage the LQA Criterion dataset to understand why each recommendation is provided by looking into the detailed analysis of your content.
- Create your listing quality score by combining the available analyzed criteria and implementing the most relevant formula to your business.
Market
With these LQA datasets, you will be able to get Market insights.
- Compare your products’ listing quality score with your closest competitors by leveraging the Listing Quality Score and the Competitors datasets.
- Study how competitors with high listing quality build their listing.
Benefits
- Improved visibility and sales: by identifying quickly which changes to implement on your listings, you help your products stand out on Amazon & Walmart, increase your visibility, enhance your customer experience, and increase your sales.
- Efficient use of your time: by leveraging the LQA datasets, you prioritize your efforts and focus on the listings that are likely to have the most significant impact on your business.
- Identify which of your products have an edge on the competition and which ones are falling behind in terms of listing quality.
Tip
With destinations, you can enrich your analysis with all the information about your products, such as their sales or estimated monthly sales, price, rating... It enables to analyze specific subsets of products, such as those with the highest sales, the most significant volume, or newly launched ones. Additionally, you can leverage the tags system to analyze a specific selection of products.
Data Model
Using destinations, your accompanying DataHawk-powered Snowflake or BigQuery database contains multiple tables about Listing Quality Analysis. These tables have a very exhaustive list of data points as columns and are organized into schemas. All Listing Quality Analysis data is available under the PRODUCT schema.
| Schema | Table | About | Refresh Rate | Primary Key |
|---|---|---|---|---|
| PRODUCT | PRODUCT_LQA_CRITERION | Displays the observed content of multiple criteria for each tracked product | Daily | product_key, criterion_key, observation_date |
| PRODUCT | PRODUCT_LQA_RULE | Provides data on the compliance of tracked products with a set of Amazon or Walmart rules and recommendations | Daily | product_key, rule_key, observation_date |
| PRODUCT | PRODUCT_LQA_SCORE | Displays the global Listing Quality Score as well as Title, Description, Bullet Points, and Media Listing Quality Scores of all tracked products | Daily | product_key, observation_date |
| PRODUCT | PRODUCT_LQA_RULE_CRITERION_ASSOCIATION | Matching table between rules and criteria tables | Daily | rule_key, criterion_key |
Technical Specifications
Available criteria
| CRITERION | SECTION | OBSERVED CONTENT EXAMPLE | EXPLANATION |
|---|---|---|---|
| Character count in title | TITLE | 256 | The title is 256 characters long. |
| Stop words in title | TITLE | for | The titlte contains one stop word, for. |
| Capitalized first letter in title's words | TITLE | true,true,true,true,false | The first four words of the title are capitalized, the fifth one is not. |
| Capitalized conjunctions in title | TITLE | For,With | The title contains two capitalized conjunctions, For and With. |
| All caps words in title | TITLE | AMAZING,APPLE | The title contains two words in all caps, BEAUTIFUL and APPLE. |
| Brand name in title | TITLE | Thisismybrand | The title contains the brand name Thisismybrand. |
| Promotional words in title | TITLE | discount | The title contains the promotional word discount. |
| Decorative characters in title | TITLE | <,> | The title contains two decorative characters, < _and >_ |
| NON_ASCII characters in title | TITLE | ©,§ | The title contains two NON_ASCII characters, © and §. |
| Restricted keywords in title | TITLE | Rated #1 | The title contains 1 restricted keyword, Rated #1. |
| Emojis in title | TITLE | 🙂 | The title contains 1 emoji, 🙂. |
| Price currencies in title | TITLE | $ | The title contains 1 price currency, $. |
| Alphabetical numbers in title | TITLE | Five, seven | The title contains two alphabetical numbers in the title, Five and seven. |
| Bullet points count | BULLET POINTS | 5 | The listing contains 5 bullet points. |
| Character count in each bullet point | BULLET POINTS | 400,326,161,249 | The first bullet point is 400 characters long, the second one 326, the third one 161, and the fourth one 249. |
| Total character count of all bullet points | BULLET POINTS | 600 | The total count of characters of all bullet points is 600. |
| Ending punctuation in each bullet point | BULLET POINTS | false,false,true,true,true | The first two bullet points do not have ending punctuation, the last three bullet points have ending punctuation. |
| Capitalized first letter in each bullet point | BULLET POINTS | true,true,true,true,false | The first 4 bullet points have a capitalized first letter, but not the 5th one. |
| Character count in description | DESCRIPTION | 858 | The description is 858 characters long. |
| A+ content included | DESCRIPTION | TRUE | The listing contains A+ content. |
| Image count | MEDIA | 4 | The listing contains 4 images. |
| Video count | MEDIA | 2 | The listing contains 2 videos. |
The observed content may be null or empty.
A null observed content means that we were unable to extract the corresponding data, so we do not know what the observed content is.
An empty observed content means that we did not observe the criterion in question. For example, an empty observed content for the criterion Emojis in title indicates that we did not find any emojis in the title.
Available Rules and Listing Quality Score (LQS) weights
Our LQS score is based on a subjective weighting system, we provide it as a starting point for your analyses. You can also build your own score using all the available data. The more important a rule is for the quality of your listing, the greater its weight.
| CHANNEL | RULES | SECTION | DATAHAWK LQS WEIGHT | LQS WEIGHT EXPLANATION |
|---|---|---|---|---|
| Amazon | No restricted keywords in title | TITLE | 5 | Amazon prohibits a list of keywords |
| Amazon | No characters decoration in title | TITLE | 2 | Moderately important to maintain readability |
| Amazon | Total bullet points length below 1000 characters | BULLET POINTS | 4 | Amazon recommends keeping the total character count for all bullet points to 1000 or less |
| Amazon | At least 3 bullet points | BULLET POINTS | 5 | Amazon requires at least 3 bullets points that are concise and easy to read |
| Amazon | Description length at least 1000 characters | DESCRIPTION | 5 | Very important for a comprehensive explanation of the product and its features |
| Amazon | At least 1 image | MEDIA | 5 | Very important as it provides a visual representation of the product |
| Amazon | At least 1 video | MEDIA | 5 | Very important as it provides a visual representation of the product |
| Amazon | Brand name included in title | TITLE | 5 | Important for searchability |
| Amazon | Between 5 and 6 bullet points | BULLET POINTS | 3 | Moderately important to maintain readability |
| Amazon | No capitalized conjunctions in title | TITLE | 2 | Moderately important to maintain readability |
| Amazon | Between 5 and 7 images | MEDIA | 3 | Considered optimal to showcase the product |
| Amazon | No Non-ASCII characters in title | TITLE | 3 | Important to maintain compatibility and readability |
| Amazon | First letter of bullet points capitalized | BULLET POINTS | 2 | Not a strict requirement but helps improve readability |
| Amazon | No promotional words in title | TITLE | 5 | Use of promotional language is prohibited by Amazon |
| Amazon | No ending punctuation in bullet points | BULLET POINTS | 3 | Not a strict requirement but helps improve readability |
| Amazon | No price currencies in title | TITLE | 3 | Moderately important to maintain readability and avoid confusion |
| Amazon | Words' first letters capitalized in title | TITLE | 3 | Important to maintain readability |
| Amazon | No words in all caps in title | TITLE | 3 | Title may look aggressive and spammy with all caps words |
| Amazon | Title length below 150 characters | TITLE | 5 | Amazon requirement. Product title should be displayed properly on all devices. |
| Amazon | Bullet point length below 255 characters | BULLET POINTS | 5 | Amazon limits the number of characters per bullet point to 255, including spaces |
| Amazon | No emojis in title | TITLE | 4 | The use of emojis is prohibited by Amazon |
| Amazon | A+ content included | DESCRIPTION | 5 | Very important to improve the customer experience |
| Amazon | Description length between 1800 and 2000 characters | DESCRIPTION | 3 | Moderately important to avoid overwhelming customers with a long description |
| Amazon | Title length between 60 and 100 characters | TITLE | 4 | Important for readability |
How we compute the Listing Quality Score
The formula used to compute each product’s LQS is based on the weight of its corresponding section and the weight of the rule.
Rules weights are described in the table above.
Sections weights are as follows:
| SECTION | LQS WEIGHT | LQS WEIGHT EXPLANATION |
|---|---|---|
| TITLE | 5 | Highly important for searchability |
| MEDIA | 4 | Very important for understanding and visibility |
| BULLET POINTS | 3 | Moderately important for readability and understanding |
| DESCRIPTION | 2 | Slightly important for understanding |
In the 'DESCRIPTION' section, we consider both the traditional product description and the 'a+ content'. If all data for this section is missing, the score isn't computed. However, if only some data is missing, we adjust the score accordingly. For all other sections, a missing data point means the score isn't computed.
Each section listing quality score (title, bullet points, media, and description listing quality scores) is computed using the following formula that takes into account the rules relative to the corresponding section:
Once all sub-scores are computed, we compute the global LQS following this formula:
Why are some LQS not filled out?
If one section score is missing because of missing data, then we do not compute the global LQS.
Future Development
Rules and Listing Quality Score are only supported for Amazon for now. Walmart will be supported soon.
Updated 8 days ago