Technology Provider: Automated Insights is a provider of a natural language generation platform, Wordsmith, that converts big data into narratives or summaries understandable by humans.
User: Vivint Smart Home, an American smart home technology provider based in Provo, Utah.
Industry: Smart Home Services
Function: Localized Landing Page Copy
Vivint, a smart home technology provider was looking to add personalized landing pages to support its products and services for applications like home automation, control and security.
The company’s strategy was focused on providing a more personalized customer experience than its competitors in order to increase organic search traffic, and the conversion of online visitors into sales of the company’s products.
In order to stay ahead of the curve in a rapidly evolving market space, Vivint needed a cost-effective and speedy solution to achieve their digital marketing objectives. According to Jacob Parry, Digital Marketing Manager at Vivint Smart Home:
“We needed to create highly localized copy for 12,000 unique web pages, which we calculated would take a team of human writers over two years to complete at a cost outside of our budget.”
Vivint used Automated Insights’ Wordsmith Natural Language Generation (NLG) platform to generate website copy on a highly personalized level. Jacob added that Vivint used Wordsmith to successfully produce personalized content for almost 30,000 pages (more than twice the original expectation) over the course of a few months.
For example, NLG was used to create unique content on the Vivint website for each town in the state of North Carolina. An example of the type of content generated by Wordsmith for Vivint is showcased below:
According to Laura Pressman, marketing manager for Automated Insights, the Wordsmith platform configuration for Vivint’s requirements took about two months, during which the Automated Insights team worked alongside a team from Vivint to continually tweak the output to fit Vivint’s desired narratives.
Vivint claims that adding personalized landing page content has had a positive effect on the company’s SEO efforts (i.e. attracting more organic search traffic). According to Cole Riebenm, SEO Manager with Vivint Smart Home, the company saw increases to the top one or two search results positions (for search terms like smart home solutions, home security, home automation etc.) in thousands of local areas where the company operates. Cole added that replication of the improved SEO results would have been very difficult without the personalized content.
Vivint also claims that within three months of deploying the personalized content, local search-based monthly sales increased by almost five times over the same period. The figure below shows the number of ‘smart home systems’ sold (in unit shipments for the year 2017) by Vivint from organic local search traffic, post their contract with Automated Insights.
Laura adds that a typical smart home system from Vivint would include one of more products and services including home security, energy management, home automation, local cloud storage, and high-speed internet solutions.
Although at the time of writing we are unsure if any other SEO or digital marketing initiatives were undertaken by Vivint over the same period of time and do not claim that the results are purely derivative from their NLG implementation.
Personalization and Localization are big themes when it comes to Natural Language Generation Applications in general. When applied to website content, NLG can increase increase the likelihood of a paying customer and can help improve search engine optimization.
NLG platforms can help improve SEO and sales efforts by creating localized content that engages audiences with an aim of converting site visitors into paying customers. The added benefit of improving SEO traffic is achieved by ensuring that search engines do not downgrade search position for certain keywords by detecting ‘duplicate content’, as AI is capable of creating unique sentences and copy for a variety of similar pages.
NLG is also a good fit for creating customized content from relatively ‘templatized text’ by utilizing data that add details specific to a geographic region or for a particular product or service, at scale. For example, in an ecommerce application, NLG can help generate differentiated versions of a given product page or description, depending on where users are accessing it from or which particular product is being accessed.