<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=2006193252832260&amp;ev=PageView&amp;noscript=1">

2 Min Read

How to Think About Natural Language Processing in Marketing

Featured Image

Wondering how to get started with AI? Take our on-demand Piloting AI for Marketers Series.

Learn More

Natural language processing (NLP) can give marketers superpowers—if they approach the technology in a smart way.

That’s the key insight from Christopher S. Penn, cofounder and Chief Data Scientist at Trust Insights, in his talk at the Marketing AI Conference (MAICON) 2021.

In fact, Penn has three core pieces of advice for marketers that he shared at MAICON on how to think about NLP in marketing much, much smarter.

PS — Have you heard about the world’s leading marketing AI conference? Click here to see the incredible programming planned for MAICON 2022.

1. Understand how NLP works

In general, NLP can’t comprehend anything, says Penn. It analyzes language, but can’t understand it the way a human can.

Instead, NLP classifies language, then makes predictions of what comes next to generate new (and hopefully accurate) language.

NLP classifies language in two ways:

  • Basic counting. NLP counts letters, words, etc.
  • Tokenization. NLP takes something big and splits it into different groups (words, phrases, etc.), then starts to assign values to this text at the token, sentence, paragraph, document, and corpus levels.

NLP isn’t magic. It’s math. Understanding this can broaden your horizons about what’s possible with the technology in marketing.

2. Know your NLP use cases

So, what can you actually use NLP for?

There are several compelling use cases for NLP in marketing:

  • Word counting. By counting words, you can create word clouds, which are an easy example of NLP at work. They’re not great for analysis, but make good visualizations.
  • Word and phrase analysis. NLP can analyze language for keywords, which can help you track trends in media or across social media.
  • Topic modeling. NLP can identify how words relate to each other and group with each other, which enables all sorts of capabilities like generating language and identifying recurring themes in content.
  • Language processing. NLP can process many different types of language, which becomes useful across a lot of domains. One example: you could analyze recruiting ads and compare that data with call center language to make the ads more relevant.
  • Content optimization. NLP can even identify which keywords and phrases are being used in high-performing content, so you can mimic what works in top pieces of content.

3. Decide if you should buy or build NLP

Should you buy or build NLP?

The answer comes down to what you’re trying to do and the resources you have available:

  • Do you have time?
  • Do you have money?
  • Is there a use case for NLP?

If NLP is going to be an add-on to make your marketing more efficient, buy what you need from a vendor.

If NLP is going to be an integral part of your company’s DNA, you’ll probably want to build.

To get started with building, there are two paths to follow:

  • At the lower scale, Watson Studio is very powerful and easier to get started with.
  • At the higher scale, you can use R or Python to write your own code, and understand the nuts and bolts of NLP.

PS — Have you heard about the world’s leading marketing AI conference? Click here to see the incredible programming planned for MAICON 2022.


Related Posts

How to Use AI to Automatically Create Content from Keywords and Images

Paul Roetzer | June 6, 2022

GoCharlie is an AI tool that generates content using natural language processing (NLP). We spoke with the company to find out how it works.

5 Ways Natural Language Processing (NLP) Is Changing Digital Marketing

Sam O'Brien | February 22, 2022

One popular AI innovation for marketing in recent years is natural language processing (NLP). Here are the top ways to use it in digital marketing.

7 Key Differences Between NLP and Machine Learning (and Why You Should Learn Both)

Sam O'Brien | June 7, 2022

The term AI is often used interchangeably with complex terms such as machine learning, NLP, and deep learning, all of which are complicatedly intertwined.