Petal

Connecting chatbots and home assistants with a headless CMS

Customers expect to engage and transact on their preferred channels. As chatbots and voice assistants grow in popularity, more and more organizations are asking: do we need to our expand channel availability? And just how difficult will it be with our current CMS?


Monica RaszykPublished on May 23, 2022

What makes omnichannel experiences truly great? For many people, it’s about the level of seamlessness: the ability to engage across a range of channels, confident that a switch won’t fracture their current flow or cause them to start from scratch entirely. For others, it’s about choice – the opportunity to interact with a company on a specific channel and have that experience feel just as comfortable as anywhere else.

For companies looking to expand channel availability strategically, it’s key to really understand your unique customer journey. It’s not necessary to be everywhere, but it is crucial to meet customers where they are – and have these touchpoints be exceptional. In other words, if you’re going to do it, do it right.

More recently, chatbots, voice assistants, and smart speakers have grown in appeal and usage globally. So, should you get started with them? That depends. Here, we’ll explore some sensible use cases, then share two real-world stories of companies who expanded to these channels with the help of Kontent.ai. 

Chatbots are back – when they’re used right

Think back a few years and you’ll likely remember the less-than-stellar reputation of chatbots. The promise of having a fully automated conversation or customer care experience was a bit rich for the technology. People felt that chatbots were wasting their time, not saving it. Today, chatbots are making a deserved comeback. Supported by better content, realistic objectives, and improved intents and machine learning, many chatbots can skillfully manage a nice range of use cases. 

Consider a vertical example of an FMCG company with a large portfolio of products, from breakfast cereals and baked goods to granola bars and nut butters. A customer may engage with a chatbot to understand whether a specific product is vegan, gluten-free, or allergen-friendly. Satisfied with the breakdown of ingredients, the same customer may then want to locate the product in-store for purchase. For this conversation, a chatbot would be a useful first line of contact, saving human agents for more complex scenarios. 

Answering frequently asked questions, suggesting products, or allowing for order completion or form fills can certainly be handled by a well-configured chatbot. As long as there’s always an option for a customer to be transferred to a live agent, chatbots may have finally found their place in the world of customer care. 

List of four main chatbot use cases
Deliver realistic chatbot experiences that deliver on their promise

Not every scenario can, and should, be solved with a chatbot interaction, but with the rise in popularity of self-service, it may be an important option to have for your customer base.

Voice assistants & smart speakers make life a little easier

While chatbots have found footing in specific customer service and commerce use cases, voice-activated assistants and smart speakers such as Siri, Alexa, and Google Home can be enlisted to help out more casually, hands-free-style. Like any good assistant, they do pretty well with ad-hoc search and quick questions, and can be called on to help out during the daily grind. According to research from Capgemini, more than two-thirds of consumers say a voice assistant allows them to multi-task and complete tasks hands-free.

68% of consumers said a voice assistant allows for multi-tasking
Voice assistants find their place in everyday life

Consider the opportunity for brands like Baking Mad, who offer ideas, tips, and recipes for baking enthusiasts. Where could a voice assistant or smart speaker fit the bill? Probably not so useful during a casual browse through dessert ideas. But for bakers who need to hear the next step in a cake recipe, mid-stir? This in-the-moment support is the cherry on top. By understanding their audiences and identifying “sticky fingers” problems, Baking Mad has their answer: yes, this channel makes sense. 

With the growing preference for voice search, voice commerce, voice-activated assistance, and smart speakers, it’s definitely worth considering your unique opportunities in this area. And what it would really take to create the right experience for customers on these channels. For many companies, Kontent.ai has helped them scale to new channels without typical growing pains. 

Content-first development through the use of APIs

Using a headless CMS for chatbot and voice assistants has its advantages. Michael Kincaid, CTO of digital agency Reason One Inc., shared his perspective on why API-first solutions are key for expanding channel availability, without the burden of additional resources. 

His team created both a chatbot and a Google Home voice assistant using Kontent.ai. “Developers love the technology freedom provided by a headless CMS and having 100% ownership over the presentation layer. This has allowed us to fully optimize the frontend for performance, which, in turn, makes sites score higher with search and ensures users have a great experience,” he said. 

Kontent headless CMS allows companies to deliver omnichannel digital experiences via API.
A headless CMS allows for content delivery to any channel

The Create Once, Publish Everywhere (COPE) advantage of Kontent.ai APIs means the promise of omnichannel is much more attainable than organizations may initially think. “The focus with a chatbot is on content and conversation, which is a different world to content for web pages. Kontent.ai is a fantastic solution for developing and managing content regardless of the channel – web, chatbot, home assistant, smart TV – so we felt it was a great fit,” said Michael.

For the team, modeling content for conversational interfaces was about fulfilling different types of customer actions and intents. Let’s check out how. 

Michael shared, “The model for conversational interfaces can represent the dialog around initiating an action, like what a user might say to invoke the assistant. It can also define how to identify conversation parameters or arguments in what people are saying, such as, I want a quickhot lunch. Finally, you can model the response and how to go about fulfilling it. For our chatbot, we modeled both the dialogue content and fulfillment content in Kontent.ai.”

Modeling content for chatbot interfaces differs from web
Modeling content for chatbot interfaces

For Reason One’s Google Assistant, OtterBot, all fulfillment content was stored in Kontent.ai and the conversation design was developed using DialogflowDialogflow lets users build out natural language interactions for all sorts of channels such as chatbots, IoT devices, Smart TVs, etc. Simply configure web hooks to pass the conversation parameters to your own fulfillment application that can then integrate with Kontent.ai as the content store.

The technologies involved in modeling content for home assistants like Google Home
Designing for voice assistants with Kontent.ai

By creating a single cloud-based content respository in Kontent.ai, rather than storing content in multiple platforms, it is much easier to expand to other channels when the time is right, while also ensuring the customer experience continues to feel seamless, consistent, and on-brand. 

Watch Michael’s presentation to learn even more about connecting chatbots and Google Home Assistant with Kontent.ai

Expanding to voice made possible with a headless architecture

Let’s revisit the Baking Mad example and see how it all came together. Speaking at the Kontent Horizons conference, Tom Marshall, Head of Technology at Kyan, shared his experience designing a solution for Alexa, using the same recipe content that drives the website. 

“Baking Mad is all about encouraging the baking hobby and having fun whilst baking, so we wanted to see if we could remove this friction and improve the baking experience. With our hands out of action, voice was an obvious candidate for a solution,” he said.

Before voice was even in scope, they first migrated Baking Mad from a traditional CMS to the Kontent.ai modular content platform as part of a wider modernization initiative. This opened up a lot of future possibilities for Baking Mad. With a headless architecture and a completely new website, omnichannel experiences could also become a reality. Why, exactly?

Tom shared, “It was the headless architecture that made expanding into a channel like this feasible. It wouldn’t require content editors to maintain a second catalogue of recipes; there would be no complicated or automated data sync, just a single canonical source of recipes maintained in the headless CMS that could drive both the Next.js website and the voice experience,” he said.

A single source of recipes maintained in Kontent drives the Baking Mad website and voice experience
A single source of recipes maintained in Kontent.ai drives the website and voice experience

Expanding on the benefits of headless, Tom said, “Structured, modular content is empowering. Baking Mad have been trying to do something with voice for years, but it’s Kontent.ai as a headless CMS that’s really made that feasible and possible. ​Content creation and delivery with Kontent.ai gives brands the opportunity to meet customers’ unique channel preferences​.”

Modeling content for voice
Modeling content for voice

Digging deeper into the details, Tom breaks down three phases of designing for voice interfaces, which requires a different approach to visual. 

Defining the voice interaction model for Baking Mad and Alexa
Defining the voice interaction model for Alexa

“There’s a lot to consider, but there’s plenty of resources out there to help us with this. And the Alexa developer experience is pretty awesome. With the design to the developer console, it’s a great place to get started with voice,” he said. As Tom shared, with over a quarter of UK households now having a smart speaker, and three quarters of those speakers being Alexa, it simply was the obvious choice.

Play button iconPlay button icon
Play video

Watch Baking Cakes with Jam(stack), Kontent.ai, and Alexa to get the full story.

Why going headless will set you up for future growth 

While omnichannel is still considered a future state for many organizations, it’s important to recognize how much, and how fast, customer preferences and behaviors are changing. Sooner rather than later, omnichannel availability will be more than a trend to watch, but a non-negotiable. 

By staying close to your customers and audiences, you’ll be able to decide what channels makes sense, and when. Before that time comes, it’s crucial to assess whether your technology stack, content management, and ways of working will help you grow or hold you back. To learn more about omnichannel digital experiences and headless content management, check out the following resources: 

And if you’re interested in assessing your channel strategy and learning how Kontent.ai can help, connect with one of our content experts by scheduling a demo.

This blog was originally published in July 2017 and has been updated to expand on the original premise and include an additional use case.

Written by

Monica Raszyk

I head up the product marketing team at Kontent.ai. On the blog, I like to explore and share how organizations can achieve huge returns on one of their most strategic assets: content.

More articles from Monica

Feeling like your brand’s content is getting lost in the noise?

Listen to our new podcast for practical tips, tricks, and strategies to make your content shine. From AI’s magic touch to content management mastery and customer experience secrets, we’ll cover it all.

Listen now
Kontent Waves