Webinar Recap: Building Product Descriptions with AI - Indix



Webinar Recap: Building Product Descriptions with AI

This article was originally published on the Narrative Science website and authored by Gwen Smith. Click here to read it.

Complete product attributes and enticing descriptions can be the differentiating factors between a customer buying products from your website or shopping somewhere else. To attract more visitors, marketers are constantly working to unlock the secret behind search engine optimization (SEO) when describing their solution within the digital world.

Will producing more high-quality content improve SEO? Or should they ensure the content is dynamic and up-to-date? Perhaps it is the element of uniqueness—describing features in a differentiating, enticing way? The answer to SEO success is all of the above.

However, creating descriptions at scale could mean hiring an army of product writers to find missing product details and write unique content, which is expensive, time-consuming, and tough to manage.

In a webinar on May 17, 2017, Shankar Sundaram of Indix and Gretta Hermes of Narrative Science discuss how to solve this challenge through the Indix and Narrative Science partnership. You can watch the recording here, and read a few highlights below.

What Challenges Does Indix Help Overcome?

Have you ever shopped online and read different core-product details for the same product on different websites? Often, e-retailers don’t have enough data points on their products to generate comprehensive product details and descriptions, resulting in the consumer receiving a multitude of inconsistent information from various resources.

E-retailers struggle to deliver the information customers and search engines alike are looking for. Using the Indix Cloud Catalog of more than 700 million products, e-retailers gain access to the world’s largest collection of structured product information in the cloud.

What Challenges Does Narrative Science Help Overcome?

When working with incredibly high volumes of product pages, refreshing the content for millions of products can seem like a daunting task. With Quill, the Narrative Science advanced natural language generation (Advanced NLG) software, e-retailers can automatically scale their content coverage in seconds, without adding additional resources.

Quill allows brands to leverage their style guides and SEO strategy in a scalable way that can write thousands of descriptions, indistinguishable from what a human would write, in seconds.

How Does the Indix and Narrative Science Partnership Address These Challenges?

The Indix Cloud Catalog contains product information from thousands of websites and feeds; each product has a comprehensive list of attributes. For each product, Quill, the Advanced Natural Language Generation (Advanced NLG) platform from Narrative Science, uses these attributes to automatically write an accurate and detailed description alongside the product specs, which reads the same as if human would write it.

What Are the Long-Term Benefits?

E-retailers can scale the way they engage with their audience and appeal to search engines through dynamic, automated descriptions. With these valuable tools, clients have seen an increase in site traffic, click-throughs and conversions, and an enhanced quality of web content with informative, engaging descriptions.

Plus, content teams are now able to focus on measuring what is working and how to tweak their keyword strategy, rather than getting bogged down in constantly writing and updating product descriptions.

If Two Customers Have the Same Product With the Same Attributes, Will the Content Be the Same on Their Different Websites?

This is where Quill writes custom product descriptions. We take your brand stylebook and configure Quill to talk about products as you would want them to be described. Each configuration is unique to the client or brand and can be as specific as possible.

Also published on Medium.

  Download the Pervasive Commerce White Paper
Register for webinar

Leave a Reply

Your email address will not be published. Required fields are marked *