Tag! Your Website Is It—with AI-Generated Analytics

Corey Spencer demonstrates #LaunchIt during Sneaks at Adobe Summit 2018.

Data is crucial for creating good digital experiences. You have to understand both your website and your customers—who’s visiting your site, the content they engage with, what they buy, why they leave, and more—to support the customer journey.

Of course, getting reliable and actionable analytics takes more than a mountain of data. It requires strategic planning. You need to ensure you’re measuring the most relevant actions and customer behavior, and that means every element of the site needs to be tagged thoughtfully, efficiently and consistently. Doing it correctly takes an understanding of the site’s technical structure, along with your business goals. It’s a complex, challenging and unglamorous task.

“Marketers and IT teams have to work together to tag a site,” explains Corey Spencer, director of product management for the Adobe Experience platform. “If they’re not using the right tools, it can take months. Using a tool like Adobe Launch, which is an API-powered tag management system, a good technical marketer might be able to reduce that time to days, or even hours if they have a good understanding of the site structure and what they need to measure. We wanted to see if we could use the power of AI to make it better.”

The result is a sneak called Launch It, recently demonstrated at Adobe Summit. Launch It uses natural language processing and image recognition capabilities to analyze a website, understand its content, and automatically tag it with the appropriate JavaScript.

“Launch It reads not just the site, but all the data associated with it as well, and then makes sense of it: ‘This looks like an e-commerce site. This looks like a product page, this is a product image. This is a product name. This is a price.’ It can learn your terms and tag site elements appropriately as it learns about the data structure of the site,” Corey says.

Once the site is analyzed, and the recommended tag structure is generated, Launch It uses the open APIs of Adobe Launch to automatically deploy the new tags throughout the site. What used to take days or hours, now takes just minutes or seconds. Corey adds, “In the use case we demonstrated, you get all the traditional e-commerce metrics—paths, devices, referrals, product performance, sales, campaign tracking, and more.”

Although the AI capabilities of Launch It get the most attention, it’s the automation and integration with Launch’s APIs that are potentially most useful to the marketer. “With Launch APIs, we’re able to do things like use natural language processing to note that a check-out page would benefit from an industry best practice such as product recommendations, and then tie-in Adobe Target to help deploy personalized product recommendations, ultimately increasing revenue,” Corey notes.

Having scanned an e-commerce site, Launch It notes the shopping cart could benefit from a “recommendations” area, powered by Adobe Target.

While Launch It is currently a concept technology only, Launch itself is already free to users in the Adobe Experience Cloud. “A customer can use our APIs to automatically deploy code through Launch right now,” Corey says. “It’s really exciting to me, because I started my career at Omniture as a curriculum manger in 2004, helping customers use the technology. Since then I’ve worked in start-ups, been an analyst, a VP of marketing and more—and I keep bumping into these same issues and challenges. As a practitioner, it’s really rewarding to me to finally be able to automate the replicable parts of creating a tagging strategy and free up time for marketers to focus on higher-value tasks.”

As for what’s next, Corey hopes to continue to build out the open extensions and ecosystem of Launch and grow the ecosystem to include as many partners as possible. He’s also interested in exploring concepts like Launch It that leverage new capabilities like AI and automation. “I’ve never seen more excitement and fervor around deploying contextual AI at Adobe,” he says. “We take a lot of inspiration from the other smart people at Adobe using machine learning to increase value for our customers.”

This story is part of a series that will give you a closer look at the people and technology that were showcased as part of Adobe Sneaks. Read other Peek Behind the Sneaks stories here.