3 Lessons to Learn from Product Data Scoring

Complete and accurate product data is one of the most important success factors in turning shoppers into buyers.

According to retail analytics provider RetailNext, the web will account for, or influence, 59% of U.S. retail purchases by 2018, and omnichannel customers shop more frequently and spend 3.5 times more than other shoppers. What’s more—84% of customers believe that retailers should be doing more to better integrate their offline and online channels.

That’s why it’s important for merchants to use “data scoring,” a process that makes sure critical product data is populated across all channels and that it’s optimized for the customer’s experience. How so? Once information about a product has been entered into the system – including name, price, description, etc. – data scoring technology analyzes it and automatically assigns a value to the effectiveness of that information. The algorithms gauge a variety of factors, and the feedback helps the merchant build a comprehensive, robust content profile for the product.

To understand a “data score,” think back to your school days. A grade of 75 or higher on a test is generally “passing,” and anything below that mark usually means there’s still work to do. The same is true with a product data score. Like the age-old grading scale, if you get a data score of 100, your products are most likely to succeed.

Whether or not data scoring technology is already implemented in your product information management system, the strategy behind the algorithms are helpful for any retailer or distributor seeking to maximize the success of its products across all sales channels. In fact, data scoring itself teaches three valuable lessons.

Lesson 1: Don’t publish substandard data

It happens all the time. Product data is populated to an e-commerce platform or an in-store product database that is out of date, incorrect or lacking so much information the shopper hardly knows anything about the product she might wish to buy.

The beauty of product data scoring is that it provides an objective benchmark that data must meet before it can be published. Anything below a 75 is below the minimum standard for e-commerce, and therefore, should not be populated until a higher score is reached.

It’s good practice to keep this standard with all product data. If your content doesn’t at least contain the bare essential elements, it likely wouldn’t meet a data scoring threshold, and therefore should be amended before it is ever published. At the very minimum, make sure your data includes:

  • A unique identifier (UID) code given to a single product.
  • A short product description, which should not exceed 35 characters.
  • Main product image at least 1,000 pixels on one side. The image should be on a white background and should be a solo image of the product outside of the packaging—no lifestyle shots for the main product image.
  • Brand logo, at least 400 pixels on one side.
  • Marketing copy that does not exceed 2,000 characters. This should be a persuasive product description directed to the consumers. It’s best to include what the product does, its distinguishing features and why the consumer should buy it—without overloading the consumer with too much information to wade through.
  • At least four feature-rich bullets with a maximum of 300 characters to highlight, clearly and concisely, the benefits and features of the product.

Before you publish, these elements must not only be complete, but also correct. Be sure to double check:

  • Spelling
  • Grammar
  • Formatting for dates, numbers, text, etc.
  • Character counts

Lesson 2: Overachievers in product data see better results

Adequate data isn’t the same as rich data. Data scoring is an automatic guide for going from average product content to high-quality product content that is more likely to translate into sales. So once you’ve checked the boxes for the basics above, look at improving your data a few steps further.

Adding more data attributes, such as package height, weight and depth aren’t always required, but they can help customers, retailers and anyone within the supply chain better account for the product. Include more feature/benefit bullets to provide more clarity about the product to customers.

Product data scores hit the top of the chart when additional digital assets are added. That could include multiple detailed product images that represent different views of a product, product installation guides, user manuals, videos and planograms, among other assets.

The more data that is thoughtfully placed within product content, the more information people on both sides of the transaction have to maximize sales (and reduce the number of returns processed).

Lesson 3: Product data is never one-size-fits-all

Content shouldn’t be reused for multiple products, but tailored to each specific product. Data scoring technology will automatically trigger new fields as products are classified into their respective categories, making certain that pertinent product-specific details are included in the content.

For example, customers purchasing a drill would be interested in battery compatibility and torque, versus those who are looking for a light bulb, who would need to know wattage, bulb color and shape.

Even if product data scoring technology isn’t integrated into product content management, the lessons and best practices it teaches are valuable for the retailers and suppliers looking to make their products stand out online and on the shelves—and ultimately into the shopping carts of consumers.

Brian Roloff is the campaign and media director and Regan DeHaven is a project manager at Edgenet

Leave a Reply