We stand with Ukraine and our team members from Ukraine. Here are ways you can help

Home ›› Artificial Intelligence ›› Using Conversational AI Platforms to Find the Real ROI

Using Conversational AI Platforms to Find the Real ROI

by UX Magazine Staff
5 min read
Share this post on


As conversational AI reshapes business processes, major platforms like OneReach.ai and Cognigy are emerging as pivotal players in scalable implementations. Discover how these platforms orchestrate AI ecosystems, providing long-term value and transforming product design in ways that tools like ChatGPT alone cannot.

With Forrester releasing their new report on conversational AI for customer service, all the major analyst firms (including IDC and Gartner) have compiled similar reports on the top platforms many companies are using to create conversational AI product experiences for their employees and customers. This is happening in a moment where AI agents have come to dominate discussions about how companies can orchestrate business processes using conversational AI. 

While everyone has heard of OpenAI and Google, companies like OneReach.ai and Cognigy might not be well-known by those outside this marketplace. Still, they are among the platforms in Forrester’s report that highlight an often overlooked layer of the market that is critical to scalable implementations of generative AI. 

There is a disconnect between the short-term capabilities of generative tools and the much longer-term strategies that organizations need for actually leveraging the technologies associated with conversational AI. This wave of reports from major analyst groups seems to recognize this, as platforms for orchestrating these kinds of technology are designed to deliver the real business value that organizations are searching for.

Technology That Will Change Business Forever

The conversational AI platforms typically used by enterprises to create, orchestrate, and manage their conversational products and implementations include tools like ChatGPT as part of orchestrated solutions, not as solutions unto themselves. This is one of the reasons why major analyst firms have released reports about the top conversational AI platforms that have no mention of OpenAI or the lot. 

There are a growing number of companies training LLMs to automate meaningful tasks in a reliable fashion. On UX Magazine’s Invisible Machines podcast, Jeff McMillan and David Wu of Morgan Stanley described the massive amount of effort that went into training an earlier GPT model on more than 100,000 internal documents so that their advisors could essentially have conversations with the investment bank’s wealth of data.

The project has been a success, but it’s still just one piece of organization-wide automation. If an organization is going to use conversational AI across departments in both employee- and customer-facing use cases, their AI agents are going to need an ecosystem. The platforms that consistently appear as leaders in these reports seem to be what analysts see as the answer for product owners, technologists and product teams: platforms that provide ecosystems instead of building ecosystems from scratch.

OneReach.ai, Cognigy, and Amelia are fixtures at the top of these reports. In the Forrester Wave: Conversational AI For Customer Service, Q2 2024 report, all three companies garnered numerous perfect scores in categories that matter to product owners, technologists, and business leaders: 

OneReach.ai scored perfectly in 10 categories, including: 

  • Orchestration of AI assets (5/5) 
  • Scalability and reliability (5/5) 
  • Trust and security (5/5) 

Cognigy scored perfectly in 9 categories, including: 

  • Integrations (5/5) 
  • Answer management (5/5) 
  • Vision (5/5) 

Amelia also scored perfectly in 9 categories, including: 

  • Digital user experience (5/5) 
  • Partner ecosystem (5/5) 
  • Answer management (5/5) 

Given the strengths of the different platforms, it remains to be seen how each might contribute to the maturation of conversational AI in product design. What seems inevitable is that once more organizations start finding ways to help customers and employees harness advanced conversational technologies in automating business processes, product experience design as we know it will enter a new era.

As Principal Analyst Max Ball points out in Forrester’s Wave report, “For over 30 years, customer self-service applications have consistently disappointed consumers while 

delivering underwhelming results for the brands that worked so hard to bring them to market. Large language models (LLMs) and generative AI (genAI) promise to finally change that dynamic through inherently natural conversations and significantly reduced deployment times.”

A New Paradigm for Design and Development

With the proliferation of conversational AI platforms and the rise of powerful language models, enterprise leaders face a critical decision. Which path is best suited for their organization’s needs? Let’s take a deeper look at the key considerations product and technology leaders will have to make as they prioritize long-term success by incorporating these technologies into their product ecosystem. 

Speed and Deployment Timeframes

  • Building your own from-scratch solutions using large language models directly may require more custom development work, potentially increasing deployment times.
  • Conversational AI platforms generally enable faster deployment due to their pre-built components and low-code environments.

Security and Scalability

  • Platforms typically offer better security with enterprise-grade features like data encryption, access controls, and compliance certifications. 
  • Most platforms will also provide built-in scalability features, load balancing, and the ability to distribute workloads across servers or cloud infrastructure.

Cost and Long-Term Value

  • While platforms have licensing fees, they can be more cost-effective due to reduced development and operational costs, pay-as-you-go pricing models. 
  • Prioritizing long-term value over short-term speed and cost-savings can also help prevent expensive failures and replacements.

Talent Requirements

  • Conversational AI platforms require a balanced mix of technical AI proficiencies (ML, NLP, programming), in integrations,and folks that are experienced in conversational design.. 
  • Building from language models directly has a higher barrier in terms of the depth of AI engineering, development and design talent required.

By prioritizing these considerations, companies can ensure they’re on a path that drives sustained value, adapts to changing needs, and helps them to avoid the pitfalls of poor user experiences or costly replacements.

Evaluating Conversational AI Platforms

As Ball states, “Vendors achieve differentiation in this market through mastering the art of orchestrating multiple AI assets.” When evaluating platforms like the ones evaluated in the Forrester report, companies should take the following actions: 

  • Conduct interactive demos with your own data and use cases to assess real-world performance.
  • Evaluate key features and capabilities, such as NLU accuracy, dialog management, integrations, analytics, and multi-channel support.
  • Assess scalability, enterprise readiness, and compliance with industry regulations.
  • Consider pricing models, total cost of ownership, and the vendor’s expertise and support.
  • Prioritize thought leadership, design expertise, and the platform’s ability to adapt to changing needs,  including integrating emerging technologies like large language models (LLMs).

Three Layers of Tech

Orchestrating generative AI to deploy automations at scale is likely to become more complex the larger an organization is. There are, however, three critical layers that every company deploying generative technologies at scale will have to address:

  1. Infrastructure. This foundational layer includes databases, large language models (LLMs), AI models, and other backend services that power AI functionalities.
  2. Orchestration. This layer manages and coordinates the underlying infrastructure technologies, facilitating efficient data flow, task execution, and resource allocation across AI systems, including the orchestration of various LLMs and interactions with multiple databases, enterprise systems, and vendors.
  3. User Interface. This layer enables users to interact with AI capabilities across various channels, such as voice interfaces, graphical user interfaces, web chat, SMS, messaging platforms (e.g. Teams, WhatsApp, Slack), and more.

Choose a Viable Strategy

In the rapidly evolving conversational AI landscape, a strategic mindset focused on long-term success is crucial for enterprise leaders looking to harness the power of these game-changing technologies. Prioritizing long-term value over short-term speed and cost-savings can also help prevent expensive failures and costly replacements. By evaluating platforms holistically and taking into account factors such as security, scalability, design expertise, and thought leadership, organizations can unlock the true potential of conversational AI and drive sustained value for their employees, customers and operations.

To read the complete scoring for all platforms along with meaningful breakdowns, get The Forrester Wave™: Conversational AI For Customer Service, Q2 2024: https://www.forrester.com/report/RES180735


Related Articles

The age of productivity, minimalism, and monochromatic color schemes.

Article by Elvis Hsiao
Why is The World Losing Color?
  • The article delves into the global shift towards monochromatic color schemes in various industries, exploring its implications and underlying social theories.
Share:Why is The World Losing Color?
7 min read

MFA Interaction Design at the School of Visual Arts paves the way for more thoughtful future designers. 

Article by Brooke Viegut
The Next Generation of Design is Inclusive
  • The article explores the MFA Interaction Design program at the School of Visual Arts, focusing on its emphasis on inclusive design principles and the impact of its innovative curriculum.
Share:The Next Generation of Design is Inclusive
5 min read

Synthetic insights are not research. They are the fastest path to destroy the margins on your product.

Article by Pavel Samsonov
No, AI User Research is Not “Better Than Nothing”—It’s Much Worse
  • The article critiques the growing trend of substituting human-centered design processes with AI-generated insights, highlighting the potential pitfalls and implications for design quality and user experience.
Share:No, AI User Research is Not “Better Than Nothing”—It’s Much Worse
6 min read

Tell us about you. Enroll in the course.

    This website uses cookies to ensure you get the best experience on our website. Check our privacy policy and