Analysis Featured

Tracking Consumers On Their Journey From Google Search to Purchase

Marketers and other professionals have to figure out how to create advertising that interests the consumer and put those ads in front of the people who would be interested. How do you organize information from human interaction with its variables of individual language use and action? Organization of that data would help marketers figure out where consumers’ interests are and target individuals with specific ads. This means consumers would see the ads they want to see and marketing dollars aren’t wasted.

DeepIntent uses artificial intelligence to understand customer interests. CEO Chris Paquette and co-founder Mohamed Nouh enjoyed working together on tech problems at Binghamton University, New York when they came upon the idea.

“I was getting my degree in machine learning and bio-engineering,” Paquette said. “Mo majored in Computer Science.”

The Deep Structure of DeepIntent

They reconnected three years ago while Nouh worked on high frequency trading platforms for an investment bank and Paquette worked on natural language programming for a hospital.

“I was combining structured inputs with unstructured data and unlocking patterns by blending two types of datasets together,” Paquette said. “It’s like a research project that transformed into a business once we realized we had people who would pay for an application of data that fits in the media world.”

The technology behind DeepIntent is its ability to extract meaning in a structured way from unstructured documents.

“Can we annotate metadata,” asked Paquette, “and blend it with structured data?”

He was concerned with information like how many users visit a particular website. The core technology is natural language processing and handling data in data-agnostic ways so they can work together.

“DeepIntent is an ad tech and marketing technology stack,” said Paquette. “At the core, we have NLP annotation giving a tremendous wealth of web data on where ads can appear. If you are a shoe company, you want to show your ad where there is positive talk of your brand or negative talk of competitors. As we annotate data, the second part of the stack is the data management part of the platform. During the course of serving ads, we accumulatE a load of user data, user profiles, and understanding the affinity of users toward a wide array of concepts. The breadth and depth of coverage is pretty immense. We can go beyond users interested in buying a car to the specific type of car they are looking for.”

Looking for behavioral patterns and how users surf the web provides the data. Acting on it is how money is made. Finding the common interest of users reveals the audience that performs best to the brand so ads can be targeted. DeepIntent also works with data enrichment, syncing data and enriching the current understanding of users and audience on their platform. In addition, they are working on developing a white label platform.

“The core technology has been under development for three years,” Paquette said. “It’s a mix of deep learning neural networks and unsupervised data discovery. When we see an URL or page come through a request from a supply source, we automatically index that page. It consists of real-time scraping, real-time analytics for topics and sentiments, and we store it for ad serving. If an article is on Tesla, we determine if it’s talking about the vehicle or the inventor based on deep semantic language targeting.”

The distinguishing component between DeepIntent and other ad tech companies is the level of drill-down on the data, Paquette said.

“We go head-to-head against IBM Watson and successfully outcompete on a number of data points,” he said. “It’s a constant R&D effort, improvement of algorithms, and segmentation. Ad tech is a highly competitive market. We compete with DataXu, Facebook, and Google, to some degree. But we differ in the way we approach the problem.”

DeepIntent tracks the customer from when they begin their research on a type of product to when they visited the brand website. They deliver stories and insights into customers and each customer’s journey across the Web.

“We can see things about user location, sites visited, and patterns of behavior,” Paquette said.

What’s the Future of AdTech?

Paquette thinks there is a massive consolidation on the horizon where MarTech eventually consumes AdTech. Companies that can’t differentiate will start to fade away.

DeepIntent strives to be a “data first” company providing power in their tools and the product.

“Our marketing platform is customer-centric,” said Paquette. “You can go into it with a couple of clicks and form marketing decisions based on insights so your brand is messaged more effectively.”

The company has grown 33% every month for the last six months and is looking for talent. They plan to hire data scientists, salespeople to tell their story, and people who can craft the perfect product, test it, and get it into hands that can use it.

Author:

Written by Nicki Jacoby.

About the author

Allen Taylor

An award-winning journalist and former newspaper editor, I currently work as a freelance writer/editor through Taylored Content. In addition to editing VisionAR, I edit the daily news digest at Lending-Times. I also serve the FinTech and AR/AI industries with authoritative content in the form of white papers, case studies, blog posts, and other content designed to position innovators as experts in their niches. Also, a published poet and fiction writer.

Add Comment

Click here to post a comment

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