- Artificial intelligence is one of the major business drivers today.
- Alexa is a smart assistant with built-in capabilities to help you manage office tasks efficiently.
- For Alexa to be effective, you should have in mind the skills you need it to have for the efficient progress of your business.
Alexa isn't just for the smart home. Amazon's smart assistant is on a roll, gradually making its way into the business world. FinancialForce demonstrated how it could turn an Amazon Echo into a smart office aide that can fetch expense reports and other important business documents with just a few voice commands. If you're intrigued about the idea of a command-powered office, here are some key things you could do with Alexa.
Earning cash with new applications for Alexa
Amazon compensates developers for top-performing and engaging voice apps across half a dozen categories. The program pays out for high-quality games, but now that program extends to education and reference, food and drink, health and fitness, lifestyle, music and audio, and productivity apps. The company doesn't do a great job of describing what it means by engaging, but it does say it's looking for "voice-first" apps that are unique and feature fresh content. The dollar amounts aren't astronomical, with the best skill earning $5,000, but the 300th skill would get about $100. The company is also giving away Amazon Web Services promotional credits.
How to make a skill
Making an Alexa skill is for those with developer experience, but there are several good tutorials that can walk you through the process. To get started, check out the Alexa developer site and grab the Alexa Skills Set. This has all the requisite APIs along with code tools, documentation and other resources for getting started.
Another good resource to explore is AWS Lambda. Built on the Amazon Web Services platform, it's a cloud service that lets you run code without the need to provision or manage servers. It can be a crucial cloud-based back end that has minimal costs, thanks to Amazon's dominance in this space.
Developer Brian Donohue also has an excellent walk-through for those who are new to the world of Amazon Alexa development. Experienced developers can be up and running with making a skill in short time.
Whether you publish the skill or keep it in-house, the next key step is to take it through a beta test. Developers can run one with Amazon's tools and control access to who can give the Alexa skill a try.
Here are the basic steps to make a skill in Alexa, based on guidance from Amazon.
1. Plan and design your skill.
Always have the end user in mind, creating a skill that will bring value to them. This could mean adding support screen interaction, adjusting the audio or publishing your skill to additional markets.
You should also include a voice interface. This will map out how users interact with your skills.
2. Set up the skill in a developer console.
Set a name that identifies the skill. This becomes the public name that your clients will find it under on the skills store. Start with one main language, though you may want to include support for additional languages later.
3. Use the voice design to build your interaction model.
The interaction model should include sample utterances, a dialogue model and a collection of your intents (the requests your skill can resolve). The sample utterances will map your intents to the phrases and words users will use to interact with your skill. The dialogue model will identify the information your skill requires and the prompts Alexa will use to communicate with the user.
To create these components, you can use the developer console or create the JSON for interaction model. Use these best practices when writing your sample utterances.
4. Write and test the code for your skill.
Your most essential coding task for your skill is to create a service that can accept requests from Alexa and keep feedback. The quickest way to write your skill is to first create an AWS Lambda function to host service for your skills. This will let you run your code without managing servers. Make sure you update your skill with your endpoint in the developer console, and test your skill with the Alexa-enabled device or the test page.
5. Beta-test your skill.
This stage is optional, but it's ideal to set up a beta test for your skill. This will make your skill available to a limited group of testers you have personally chosen rather than the general public.
Possible business uses for Alexa
Once you know what to do, the question is what to do with that knowledge. As the FinancialForce demo revealed, there are many potential uses for Alexa in business.
Some current skills point to this use. Quick Events, for example, lets you add items to your Google Calendar. Alexa's chatbot for Slack gives you the power to post messages to Slack by voice. Additionally, Complete Tasks works with the Complete mobile app to give you a quicker way to add tasks than searching for your phone.
None of them are dramatically life-changing just yet. Alexa may need deeper access to a company's database or other APIs before this becomes a real must-have type of workflow.
Despite it still being the early days of this type of technology, developing a skill just for your own business may also be valuable. If your organization is beginning to dabble in using Alexa for the office, a simple skill that allows you to request a form would be a great way to get started.
The key is to think not only about the use cases but if they can be accomplished well with voice. Since the Echo doesn't offer any type of visual confirmation right away (some answers are shared through the Alexa app), consider whether the workflow you're trying to achieve can be accomplished in just an auditory form. (Note: Amazon doesn't yet allow developers to charge users for voice apps, nor does it allow for in-app purchases.)
Bottom line
Artificial intelligence is already a driving force in the enterprise world. Recent innovations by Google Cloud, Microsoft Azure and other competitors in this space demonstrate that companies are seeking a smarter way to work. Alexa could fulfill that mission, but it's going to take some clever applications and more robust capabilities to get there.