Reimagining content creation with our Azure AI-powered Inside Track story bot

|

We created the Inside Track content bot to help us create content faster at lower cost.

Engage with our experts!

Customers or Microsoft account team representatives from Fortune 500 companies are welcome to request a virtual engagement on this topic with experts from our Microsoft Digital team. 

In Microsoft Digital, the company’s IT organization, innovation is the fuel that drives us.

To stay ahead, we are continuously exploring ways to use technology to drive efficiency and value. One way we’re innovating is by using generative AI to accelerate content development for our Inside Track blog.

While there will always be humans in the loop, using generative AI, we have embarked on a journey to revolutionize how we engage with our customers through compelling stories and case studies. In fact, this very article was created with the help of Artificial Intelligence.

The vision and challenge

A composite photo of Jones, Pydimarri, Sengar, Boyd, and Velush.
The Microsoft Digital internal Azure AI bot team works together to constantly improve the AI-powered Inside Track story bot. The team includes Dwight Jones (left to right), Revanth Chandra Pydimarri, Urvi Sengar, Keith Boyd, and Lukas Velush.

The vision for this project was clear: use AI to accelerate content creation, reduce turnaround times, and enable any member of our Microsoft Digital team to quickly share their story with customers. However, the path to realizing this vision was not without its challenges.

“We were venturing into uncharted territory, using AI in its infancy with a limited budget and few resources,” says Dwight D. Jones Sr., a principal product manager on the Frictionless Device team who led the initiative.

To overcome these challenges, the team created a pitch deck based on their vision, built a detailed technical specification, and worked with stakeholders to build support for the project. They formed a virtual team across The Microsoft Digital Inside Track team, Frictionless Devices team, and Employee Productivity Engineering team. With the approval of Microsoft Digital leadership, they launched a small proof of concept that proved successful—the Inside Track content bot. In the pilot, the team was able to use an AI-powered interview bot to produce excellent first drafts of blog posts, saving significant time per story. 

Accelerating publication and reducing costs 

Senior Director Keith Boyd emphasized the dual challenge of telling more stories at a lower overall cost and accelerating publication.

“The key technology that’s powering the bot is OpenAI on Azure,” Boyd says. “It’s already helping us by making it easier for subject matter experts in Microsoft Digital to share their expertise on their own schedule.” 

By using the bot, the team hopes to increase the number of stories in the pipeline and decrease the time it takes for publication.

“Our primary metric is time to publication,” Boyd says. “Can the bot help us move from a six-week cycle for story production to end-to-end story authoring and publication in half that time?” 

Watch this demo of the Inside Track bot. Customers or Microsoft account team representatives from Fortune 500 companies are welcome to request a virtual engagement on this topic with experts from our Microsoft Digital team. 

Streamlining the process 

Revanth Chandra Pydimarri, a senior product manager on the Inside Track content bot V-team, highlighted the need to streamline the drafting and publishing process.

“We were wasting resources through inefficient operational processes, and it took a lot of time for us to reach out to relevant subject matter experts, get the content from them, write the story, and eventually publish it,” Pydimarri says. 

The team turned to Microsoft Azure AI, building a bot that uses generative AI capabilities—including Retrieval-Augmented Generation (RAG) and Azure OpenAI—on your data to interview subject matter experts (SMEs) at their convenience.

“We have trained the bot in the style of Inside Track,” Pydimarri says. The results were impressive, with a time savings of at least five hours per story, while maintaining publication standards.

Capturing input and maintaining style 

Urvi Sengar focused training the bot how to write stories with the same look and feel of human-written stories, and on how to use the bot to capture input from SMEs and turn it into content. The bot was trained using a comprehensive dataset of previously published stories and with writing guidance provided by the editorial team, enabling it to emulate the Inside Track team’s storytelling style.

 “The bot’s sophisticated content summarization and curation techniques allow it to draft high-quality stories quickly, maintaining the nuanced details and tone provided by the SMEs,” says Sengar, a senior software engineer on the team in Microsoft Digital. 

Despite the challenges of transferring subject matter expertise and iteratively improving the process based on user feedback, the project has been well received by stakeholders.

“The bot is now able to generate drafts that are more than 70% complete, significantly reducing the time required to produce stories,” Sengar says.

The team is adding the ability for the bot to verbally interview SMEs by integrating VoiceRAG using Azure AI Search for audio to text capabilities. This will enable the bot to transcribe spoken interviews accurately and efficiently, streamlining the content collection phase and creating richer, multi-dimensional stories through a more natural interviewing process.

Combining multiple perspectives 

Lukas Velush, the managing editor of Inside Track, tackled the challenge of using an AI-powered bot to interview multiple SMEs and combine their interviews into one coherent story.

“Our challenge was to continue telling the stories of our subject matter experts across Microsoft Digital, despite having a smaller team and a shrinking budget,” Velush says.

This downward pressure on resources was the inspiration for the bot, which could help create stories at a lower cost and at a faster pace. 

The bot is designed to emulate human writers, interviewing SMEs about their IT work, their challenges, and how they overcame them.

“One of the biggest challenges that we’re still working on is getting the bot to be able to interview multiple people for a story then weave what each person has to say into a coherent narrative,” Velush says.

The current solution is having the bot interview each person separately, then combine the resulting stories into one comprehensive narrative. 

The initial results are promising, with bot-written stories costing 50% less than human-written stories and being completed at least 30% faster.

“This has been an amazing journey—we’ve learned a ton about AI and about humans and what it takes to report and write stories like this,” Velush says. “I see the bot as a great helper that can work right alongside our human writers.” 

The future of AI-driven content creation 

As Microsoft Digital continues to innovate and push boundaries, the Inside Track bot stands as a testament to the power of AI to transform the way we tell our stories. Using AI, the team has shown that it can accelerate the content creation process, enabling delivery of high-quality, engaging content faster and more efficiently than before. 

Looking ahead, the team plans to further enhance the bot’s capabilities. Jones shared that the team plans to use AI to produce multiple types of content with a single entry, such as articles, blog posts, PowerPoint decks, and white papers. They also plan to generate content in multiple languages and to support readers with accessibility needs. 

Boyd emphasized the potential for AI to augment human capabilities.

“One goal is to reassure other content teams that AI-powered bots are not here to replace them, but to augment their capabilities and make them more productive,” he says.

This is just the start.

“This is V1,” Pydimarri says. “We have plans to add more features and also expand the functionality to accommodate other team’s requirements, too.” 

Future enhancements will focus on expanding the bot’s capabilities and functionality. “That will ensure it continues to meet the evolving needs of the Inside Track team and other users,” Sengar says.

Looking to the future

Using AI to accelerate content creation in Microsoft Digital exemplifies the team’s innovative spirit and commitment to customer engagement.

“This project is just the beginning,” Jones says. “We’re excited about the potential of AI to revolutionize how we engage with our customers and look forward to seeing where this journey takes us.” 

The Inside Track bot has demonstrated the potential of AI to streamline processes, improve efficiency, and deliver high-quality, engaging stories that resonate with the audience. It’s a great example of digital transformation in action, paving the way for future innovations. 

As we continue to push the boundaries of what’s possible with AI, we invite you to join us on this journey. Explore the capabilities of the tools and technologies available through Microsoft Azure AI in your own content creation processes and see firsthand how it can enhance productivity and storytelling quality. 

Key Takeaways

We’re using AI to accelerate content generation at Microsoft, and you can too. Here are some highlights from our journey:

  • AI-powered content creation: Microsoft Digital is using generative AI to automate content creation for the Inside Track blog, significantly reducing turnaround times.
  • Successful pilot: The AI-powered chatbot proved successful in a pilot, producing excellent first drafts of blog posts and saving significant time per story.
  • Efficiency and cost reduction: The initiative aims to tell more stories at a lower overall cost and accelerate publication, with Inside Track content bot-written stories costing 50% less and being completed at least 30% faster.
  • Streamlined process: The AI bot interviews subject matter experts (SMEs) at their convenience, eliminating the need for coordinating schedules and saving at least five hours per story.
  • Maintaining quality: The bot is trained to emulate the Inside Track team’s storytelling style, capturing nuanced details and tone provided by SMEs.
  • Adding voice: We’re integrating Azure AI Speech Studio for audio-to-text capabilities, which will make getting interviewed by the Inside Track content bot a more natural experience for our SMEs.
  • Future enhancements: We’re expanding the bot’s functionality to help other Microsoft Digital teams.

Recent