Personalizing the experience for customers rewards them for their use of the skill, creating familiarity while providing personalized information for them alone. You can increase or decrease the degree of context depending on the individual skill that you are creating. To some degree, you need to collect information from the customer to move the discussion forward. It is within those moments that you need to determine what information you need to collect and store or what can be discarded once the individual session is complete.
Patterns and use cases
Recognize new and returning customers
When a customer invokes a skill without a request ("Alexa, open [skill name]"), the skill should deliver a welcome message, and then prompt the customer to respond. You should have several variations of welcome messages, including one for first-time use, a return welcome message, and personalized welcome messages. Immediately following the welcome message, the prompt asks the customer what they’d like to do. Consider including a variety of reminders about the skill's basic functions as a quick refresher.
For example, in a first encounter with a skill you might want to discuss what is possible and frame the conversation Alexa is able to have with the customer via the skill. After a certain number of subsequent invocations, shorten the greeting with customers to help speed their access to the interaction they want.
When using the skill for the first time
“Alexa, open Baking Pal.”
“Welcome to Baking Pal. I can help you find delicious recipes for your favorite pastries, cookies, and desserts. What would you like to make today?”
When using the skill for the second time
“Alexa, open Baking Pal.”
“Welcome back to Baking Pal, Steven. Last time you made cupcakes. What would you like to make today?”
Capture information through skill use
In the previous example, notice that Alexa addressed the customer by name. This is a piece of information that you can ask for as part of the flow of the conversation and store it to that customer’s profile. This personalizes the experience so that each time customers return, they feel more comfortable with Alexa.
This personalization is not limited to information you request from the customer. You can also use their interactions with the skill to guide Alexa’s behavior in subsequent interactions. For example, in a fitness skill designed to measure progress toward a customer goal, the skill could capture information about a specific exercise routine customers do over time to track their progress. When they return to the skill, Alexa can tell them if they’ve reached a new personal record.
Use adaptive prompts
As a customer uses a skill more and more, they become increasingly comfortable with using the skill and remembering what happens. Consider making subsequent prompts shorter and more direct, and even acknowledge the frequency of use.
For example, here are some variations in a music lesson skill:
Avoid data entry
It can be difficult for customers, even those without speech impediments, to tell Alexa long strings of numbers or detailed information. Telling Alexa an address, phone number, or zip code can be tedious. Offer alternative methods for inputting this kind of information, such as using the Alexa App or the methods listed in Access a customer's location in case they can’t relay the information by voice alone.
Access a customer’s location
A customer must give you permission to access their location. The customer’s location is dependent on where they registered the Alexa-enabled device they’re currently using. The Device Location API provides the customer’s address as long as the customer has allowed access for that specific skill. This API gives you the full street address or just the country and postal code. Knowing a customer’s location is great for personalizing skills such as greetings (good morning, good night) or services such as maps, ordering food, finding theater times, etc. To learn more about account linking and other ways to appropriately get customer information, see Understand Account Linking and Request Customer Contact Information for Use in Your Skill.
Resume a skill session after exit
Another way to create context with the customer is to keep track of their progress within the skill. For example, let’s say the customer exits the skill while in the middle of a task, such as listening to the ingredients in a recipe. When they come back, Alexa should ask them if they want to pick up where they left off. Alexa: “Let’s get back to your Poached Salmon with Dill recipe. Do you want me to read the ingredients?”