"Which conversational AI platform should I choose?"
We get asked this question A LOT.
Unfortunately there's no single, nor simple, answer to this question.
It's a big decision for any organisation embarking on a conversational AI journey, whether in the form of a chatbot, digital human, or smart IVR solution, and one that is important to get right at the beginning of the project to save the headache and regret in the future.
Our team at Spark 64 has worked with several of the major platforms in the past and we want to answer the common questions that crop up and share what factors you should consider with choosing a conversational AI platform.
If you're building any digital assistant, you're going to need an engine to interpret the user's query and respond appropriately. Conversational AI platforms provide you with the platform to do just that, wrapping all of the latest advancements in natural language processing and tried and tested patterns for designing your dialog.
They're also known as chatbot platforms, but I prefer the term conversational AI because they can be used in solutions far beyond chatbots; digital humans, sophisticated IVR, and even as part of a web/mobile application for interpreting search queries!
Since 2016, there has been an explosion of conversational AI platforms available that range from simple drag drop interfaces to full blown programming framework allowing you to implement a custom solution in code. These are offered by open source communities, large cloud vendors, contact center software providers, and also independent software providers (some of which have been rolled up into a cloud offering).
Unless you're in the business of building conversational AI platforms, a lot of common patterns and requirements have already been thought out by the existing solutions. Even if you're concerned about data security, there are solutions you can self-host and never leave your premise or private cloud environment. Virtually, all conversational AI platforms can integrate with your backend so there's really no need to go full custom.
Is the bot going to be a simple FAQ bot, where it's a single answer for each question? Or do you intend to take the user down a conversational journey with various paths, depending on their responses along the way?
We've seen two main types of platforms. There’s platforms that make it very easy to build simple dialogs, but get more complicated when deviating beyond. Then, there’s those that make complex dialogs very easy to manage, but may have a higher initial learning curve and some options that might not make sense to someone less familiar with the platform.
The analyst, contact center staff, developers, or both? One of the major factors of success is the ability to iteratively improve on the conversational ability of the digital assistant over time.
Some of the solutions are designed to be used by more business facing users, whereas some are intended more for developers. Traditionally, it would be an inversely proportional to the complexity of the dialogs you can build on it, but some of the platforms these days strike a good balance between the two.
In many applications, it is critical to be able to hand the user over to a human agent for complex questions. What's worse than having your users get stuck in a dead end, and have to call in anyway with one extra complaint?
Some of the platforms have out-of-box integrations with helpdesk solutions, which will save you heaps of time in building your own (which is possible with almost any option).
There's far too much for us to go into in this blog post, so we've created a report containing a review and comparison of the platforms. This has been put together with the top business questions in mind, as presented above. You can download it here:
In this report, we will be focusing on the conversational AI platforms that are backed by the top cloud vendors: AWS, Google, Microsoft, IBM. These have been around long enough to be considered quite mature, and are amongst the most popular. We’ve covered:
Again, this is not an exhaustive list; there are plenty others out there as well (notably Rasa) - some tailored for specific use cases or integrated directly with a particular contact center suite. If you've found one you've really liked, we'd love to hear about it!
If you're embarking on a conversational AI journey, it is worthwhile taking some time to pick the platform you're about to put the time and effort building content on, and training staff to use. We've put a few comparisons and commentary on some of the top conversational AI platforms out there and we hope it is useful for you to start your journey.
Spark 64 has experience in using all of these platforms. As a tech-agnostic AI agency, we will guide you to select the best platform to suit your company’s needs and specifications. Feel free to reach out to us if you want to have a chat about your own requirements and we can give you an impartial opinion on which you should choose!
This Lunch and Learn webinar provides a detailed overview of reinforcement learning, its applications and use cases, and a selection of resources you can explore to learn more about reinforcement learning.
What is the edge? What are some of the key considerations to explore when you're planning an edge AI implementation? How can you best leverage the edge to give you the best business outcomes? Read on to learn more about how the edge brings AI to your fingertips.