Do You Need Natural Language Generation?

Answer the following questions:

  1. Do you have more data than you can efficiently analyze?
  2. Do your dashboards take more than two minutes to understand?
  3. Do you have people writing lots of data-driven reports? (financial summaries, profit / loss reports, asset reports)
  4. Does your organization perform any daily, weekly, monthly, or otherwise repetitive, data-based writing tasks? (sports recaps, product descriptions, financial news)
  5. Does your organization outsource numerous but simple writing tasks?
  6. Does your organization send mass communications or notifications to users, subscribers, or potential customers?

If you said yes to any of these questions, then natural language generation (NLG) tools and solutions can help you. If you said yes to more than one, I suggest reading my latest report (The Ultimate Guide to Natural Language Generation) and browsing NLG vendors’ websites (Automated Insights, Arria, AX Semantics, Narrative Science, Narrativa, and Yseop). And if you said yes to all six, I don’t know what you’re waiting for…

NLG is a successful offshoot of artificial intelligence with a growing market of six primary vendors and hundreds of customers. NLG offerings ingest data and use a combination of templates, rules, and artificial intelligence technology to produce reports, summaries, product descriptions, analysis, and more in plain language. There are three overarching use cases for NLG: automation, BI and reporting, and personalization.

Automation (Qs 3, 4, 5)

If your organization repeats writing tasks, particularly data-driven ones, NLG can help. NLG offerings automate writing tasks from e-commerce product descriptions and sports recaps to quarterly financial reports and daily manufacturing equipment status reports. NLG can free people from labor-intensive writing, reduce outsourced work, improve accuracy, and ensure timeliness and consistency.

BI and Reporting (Qs 1, 2)

Business are complex, and dashboards built to relay key information and trends in business data can be complex too. If you look at a dashboard for more than two minutes and it’s still unclear what the data says or what your next steps should be, it’s no longer serving its purpose. NLG offerings integrate with BI tools to summarize key points in dashboards and data sets and even prescribe next steps, accelerating time to insight and action.

Personalization (Q 6)

Many organizations collect personal data on their customers and track user history. NLG enables organizations to automatically send personalized messages to thousands of customers or users based on their user history, rather than send the same generic email or notification. Personalization increases communication effectiveness, increasing click-rates, Website traffic, etc.

In the past, NLG vendors built solutions for their customers, and most still do. But NLG gained steam the last couple years because vendors released their technology as self-service tools and existing tools reached a certain level of maturity. These tools enable developers and non-developers to build NLG solutions at reasonable prices. Now, Fortune 500 companies use NLG daily and chances are you’ve already read something written by NLG software without even knowing it.

If you have any inquiries about natural language generation please leave comments below!

Henry H. Eckerson

Henry Eckerson covers business intelligence and analytics at Eckerson Group and has a keen interest in artificial intelligence, deep learning, predictive analytics, and cloud data warehousing. When not researching and...

More About Henry H. Eckerson