Business

Understanding Key Success Factors in Generative Conversational AI Applications

A qualitative analysis of the strategies employed by companies that have implemented Generative Conversational AI (GCAI) into their products.

► Jump straight to the full case study ◀

If you have been following this space over the last couple of years, you've likely noticed the rapid pace of change in the Conversational AI sector. Hardly a week goes by without significant news emerging: a new large language model boasting unprecedented performance, a major company reporting extreme levels of customer service automation thanks to Conversational AI, or a novel voice-first consumer device attempting to render smartphones obsolete.

With all these innovations emerging, it's sometimes beneficial to take a step back.

Drawing conclusions in a sector that is still “in the making”, where projects are often in experimental phases, can be challenging. However, patterns are beginning to emerge, and even when definitive conclusions cannot be drawn, there's always room for discussion.

That's why we've decided to examine various real-world cases of initiatives by leading companies that combine Generative AI and Conversational AI.

Speaking of successful initiatives, the genesis of this analysis came from the “Matching Market" of the Platform Strategy Research Symposium at Boston University's Questrom School of Business. The Matching Market is an initiative aimed at bridging academic knowledge and research with practical applications.

This work was conducted in collaboration with Sarah von Bargen, researcher at Harvard Business School, to whom we extend our sincere thanks for her contributions to this case study, along with Paolo Cervini, who facilitated this collaboration.

What is this about?

The result is an in-depth examination of recent instances where leading companies integrated Conversational Generative AI into their products and the possible managerial implications. 

We intentionally selected cases that are significantly different from one another, not with the aim of comparing the companies, but rather to showcase the diverse applications of GCAI.

Here's what you'll find:

  •  Introduction:
    • Background on the Rise of Generative Conversational AI (GCAI)
    • Business models
  • Case studies:
    • Duolingo: Using GCAI to build ground-breaking products
    • Shopify: Helping two sides of a market with GCAI
    • BloombergGPT by Bloomberg
    • OpenAI’s GPT store
  • Conclusion and key managerial takeaways

Read the abstract

Conversational AI (CAI) has been making headlines since the mid-2010s. In the last two years, as CAI and Generative AI (GenAI) have begun to combine through products such as OpenAI’s GPT series and Meta’s LLaMA, this hybrid has exploded in popularity. Companies and individuals across industries are rapidly adopting it, finding case-specific applications for the technology. The sudden explosion of both established companies and start-ups developing products in this space leads to the question, What makes a successful generative conversational AI product?

The purpose of this case study is to provide a qualitative analysis on how companies have recently and successfully implemented generative conversational AI (GCAI). For this case study, we have chosen 4 companies to analyze: Duolingo (education), Shopify (e-commerce), Bloomberg (finance), and OpenAI (tech). After our analysis, we summarize each company’s GCAI application with a Key Success Factor (KSF), which can be applied more broadly. We chose these companies for two primary reasons: not only are these digital companies with platform elements, but also they are at the top of their industries from a brand perspective, and thus provide excellent examples of digital companies which have successfully implemented GCAI. By studying companies across industries, we focus on universal KSFs, rather than industry-specific ones. We also chose companies with their primary regional loci in the United States for simplicity.

This case study is meant to give a qualitative approach to understanding some of the KSFs necessary to (1) build a CAI product, company, or platforms, or (2) integrate CAI efficiently and effectively in a company’s processes. For a more in-depth, quantitative analysis, we refer you to our paper, How does device type and product reliability affect user engagement and retention in Conversational AI? An empirical analysis using Alexa Skills data. The purpose of developing both a qualitative and quantitative approach to understanding KSFs in this space is to capture effects which may be difficult to identify in one study type alone, thereby creating a thorough and encompassing approach to our analyses of this cutting-edge technology. 

For managers, we provide a succinct summary at the end.

[...]

Ready for more? Get the full case study

More case studies & conversational AI guides

Leggi di più

Iscriviti alla nostra newsletter

Ricevi aggiornamenti importanti dal nostro team esperto

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Iscrivendoti accetti la nostra Privacy Policy

Voice technology and Conversational AI: not so easy to digest.

Get our free Voice Bites 1/month

Waaaaaay easier to digest.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
By signing up you agree to our Privacy Policy

(Wanna check Voice Bites first?)