Feb 21, 2018
By: Jason Smith
With the explosion of IoT connected devices and their integration into our daily lives, we’re seeing a shift in how we both create and consume content. One of the biggest advancements is the Amazon Alexa device, which is helping to shift content to a service-based approach, instead of being purely informational and consumption based.
By building their own Alexa skills companies and brands can take advantage of this innovative device and create voice-based conversational content that helps to better solve user problems.
One of our clients, Gettysburg College, recently built a set of Alexa Skill using the dotCMS platform. So, we got back in touch with the historic college to glean some advice about building effective Alexa Skills.
To limit overheads and streamline the development process, Rod Tosten, VP of xxx at Gettysburg College, recommends that brands first take the time to understand where there content is, how it’s formatted and how it can be delivered to an Amazon Echo,
“First, you really have to know how all of your content is already formatted and how someone can consume it. Second, you need to put a lot of thought into how conversations are going to happen: What will people be asking for and how will you give them that information with the content that you have?”
There are many different considerations you’ll have to take into account in regards to making sure your users are delivered the correct information through their interactions with your Alexa Skill. The process below will help you work out the information delivery process, so you can ensure your users are able to access the information they need with a streamlined experience.
Keep in mind that you can also bring in additional forms of media to the Alexa skill. For example, you could have a user input their phone number and receive a link, or additional media that furthers the Alexa skill interaction.
Don’t just build a whacky skill because you think it will entertain your audience. Build a skill that your customers or clients actually need — which is something you can discover by listening to the voice of your customer.
“Based on your content delivery system - whether it be your website, mobile app, or other - look at what information people want the most and start to figure out how you’ll turn that information into a dialogue. That’s what we did at Gettysburg with our campus dining menu and now our course requirements,” said Rod Tosten.
Before launching your Alexa Skill, consider the potential pitfalls that you may have coded or written into your skill. After all, this is likely a new channel for your brand, so there will be some unfamiliar roadblocks.
For example, Tosten advised developers to, “think about any potential language issues. For example, should 123 be pronounced as “one two three” or “one hundred twenty-three”?”
Such considerations need to be made within the context of your audience. For younger users for example, you’ll want to shorten Alexa’s responses to avoid losing the child’s attention.
Developing audio content for an Alexa Skill is an entirely different kettle of fish compared to developing content for a web browser. The user interaction with your content will be back and forth, just like a conversation. Such conversational content allows your users to build a relationship with your content, instead of pure consumption.
Because of this, your content needs to be formatted differently. This is especially true if your skill is going to require back and forth engagement, instead of a standard question and answer approach.
When examining your existing content see if you can create a structure that caters to a back and forth spoken conversation. Consider opening lines or questions, responses to the user’s responses and additional content to keep the conversation flowing.
As previously mentioned, it’s likely that your brand will run into some issues when developing at scale for the Amazon Echo, simply because it’s virgin ground for most marketers. It’s helpful to think of your Alexa skill as an iterative process. With any new piece of technology, there will be certain kinks that’ll need to get worked out. This is especially true since you probably won’t have the existing data available to predict every single user input or response from the get-go.
Instead of leaving your user frustrated when they can’t execute a command, or Alexa can’t retrieve the requested information, you should create a fail state that’s fun and enjoyable for your user. Similar to how a lot of sites today have upgraded their 404 pages to show off a little humor and personality. It makes the process of not being able to find what you’re looking for a little more lighthearted.
As your experience with audio content expands, you can expect your Alexa Skill launches to run more smoothly. But until then, it’s better to be safe than sorry.
Just like when you’re submitting an app to the Apple App store, you’ll have to go through an approval process before your skill can be accessed by the public. During this approval process, Amazon will take an in-depth look at your skill. So, make sure that there aren’t any glaring red flags or undocumented features when you submit it for review.
Sure, the first iteration of your Alexa skill won’t be perfectly polished, but nonetheless, it’s important to treat this first version as the final version, as long as you want approval by Amazon.
As you gain more real-time user data and feedback the language variances will become more true to life and the amount of information that’s able to be readily accessed will grow.
Amazon Alexa and other related devices are bringing about a huge shift in the way we think about and create content. Instead of content being static and based on a digital display, brands will need to adopt Content-as-a-Service models to aid them to deliver audio and visual content, text content and audio-based content to an increasingly wide range of devices— starting with the Amazon Echo.
Read our white paper, ‘IoT In Education: How Gettysburg College Built Student-friendly Alexa Skills”.
Brands are turning to Content-as-a-Service (or Headless CMS) as a way to future-proof their content. In CaaS, content is stored separately from the presentation layer, meaning it can be pushed to any connected device in the Internet of Things , think Alexa, Google Home, etc. Moving to CaaS means brands need to shift their thinking when it comes to creating, storing, and presenting content.Download
69 percent of Alexa Skills have zero or just one customer review, signaling extremely low usage. Here’s how to build an Alexa Skill that your audience will love.
Alexa for business is up and running, and we’d love to see developers take it a step further with these five ideas for business Alexa Skills.
What is voice shopping, how does it work and what’s coming next? We’ve taken a look.