3 Ways to Make Your Natural Language Generation Solution a Success

January 18, 2016 Katy De Leon

Natural Language Generation solutions

We work with all different types of organizations, from the U.S. Intelligence Community, to some of the biggest Financial Services companies in the world, to e-commerce platforms with less than 5 employees. Regardless of the use case, we've identified best practices in ensuring your natural language generation solution deployment runs smoothly and fast.

1. Assign Ownership

Every project needs an owner. In the case of implementing a natural language generation solution, it’s best to have two: a subject matter expert (SME) who understand the business and analytic requirements, and a developer or development team who is familiar with the data.

The SME will be critical to providing feedback and ensuring the system meets the goals of the business.  The developer will be on the hook to automate the extraction and delivery of data, likely to an API or FTP, and automate the retrieval and distribution of the reports that the NLG system produces.

2. Develop an Integration Plan

An NLG project will either be integrated into an existing workflow, or it will be the catalyst to build a new workflow and distribution process. It’s critical upfront to understand and articulate how this is going to work.

For example, if you plan to call an API, how will you do this and to where will you distribute the content? An internal content management system? A dashboard? What is your channel for distribution?

3. Establish Metrics

One of the surest ways to fail is a lack of metrics. It’s critical to know why you are implementing a natural language generation solution and identify how you are going to measure success. Will you be looking at the impact on internal man hours, or page views on your website, or revenue tied to your new information product?  

Putting the appropriate measures in place and then articulating improvements from your NLG solution will ensure that your decision to implement it receives the appropriate recognition - and you’ll be a hero!

In order to get the most value out of your natural language generation solution, we recommend that you follow these 3 steps.  By doing so, you may even find more use cases for NLG as there are many places it can add value to your organization.

Automated Analyst White Paper

Previous Article
Employee Spotlight: Jordin Huffman
Employee Spotlight: Jordin Huffman

Read an interview, as part of our Employee Spotlight series, with one of our Business Development Associate...

Next Article
Being a Robo-advisor Client for a Year: The Great, the Good, the Ugly
Being a Robo-advisor Client for a Year: The Great, the Good, the Ugly

After being a robo-advisor client for a year, here are a few areas where I think robos have succeeded and f...


Get Narrative Science blog posts in your Inbox

Keep an eye out for your confirm email!
Error - something went wrong!