Improving discoverability for news and media voice apps
Smart speakers have seen tremendous growth in the last few years and are becoming more popular than ever, reaching critical mass in 2018. Yet there is a little problem for people trying to take advantage of this trend. Only 49% of users say they have tried third-party voice apps.
Most users are happy with using the smart speaker to play music, and check the weather, or ask questions. And when they do, there is only a 6% chance that user will be active in the second week. Unfortunately no matter how much time and money you spend on developing a voice experience for your website, it won’t help anyone if your audience doesn’t even bother using it. This makes discoverability for third party skills is a serious problem. We’ll go over ways to increase engagement with your voice app including crafting a good invocation phrase, designing for stickiness, and the role of email.
Invocation phrase
The user needs to invoke the action saying its correct name. This means your action needs to be at the top of the users mind when they want the assistant to do something. With hundreds of competing voice apps this becomes an even greater challenge. The good news is Alexa and Google both support name free invocations — where the assistant automatically invokes a 3rd party skill. Crafting a good invocation is the first step in creating a way for your audience to engage. Google outlines some of the top tips for implicit invocation and discovery.
In short you need to craft your invocation phrase so it reflects your action’s use cases. For example: Ok Google, talk to Personal Chef to look up chicken soup recipes. Is a good phrase to be implicitly invoked by, “Ok Google, I need a chicken soup recipe.”
Your sample invocation phrases should be broad and cover the various ways a user might try to invoke your action. And of course you want to make sure your name is easy to remember and pronounce. If you are building for Alexa, your action must support the CanFulfillIntentRequest, to be automatically invoked even if the user did not enable it.
Build better voice apps. Get more articles & interviews from voice technology experts at voicetechpodcast.com
Designing for Stickiness
Getting your users to try your skill is great, but it kind of sucks when they never use it again. A 2017 Voice Report found that when a voice application acquires a user, there is only a 6% chance that they will be active in the second week. This analysis found that utility skills see the highest repeat usage. However since a flash briefing skill is not considered utility your need to focus on increasing affinity. Your skill needs to be memorable and bring real value to your audience. For a news or media site a flash briefing is a great place to start, however there are a few things you can optimize to maximize affinity.
You can start by using a human voice whenever possible. Listing to a flash briefing is much more memorable when the audio is read out by a person. However, recording quality audio every day might be challenging for some. An alternative strategy is combining both human voice and speech synthesis. This is a low cost alternative to automatically generating a quality briefing. Essentially you can look for the parts of the briefing that are consistent day to day, and get them professionally recorded. Welcome messages, transitions between headlines, and CTAs are all prime candidates for this. You can then use Speech Synthesis Markup Language (SSML) to combine the recorded audio with speech synthesis for today’s headlines. This technique was originally pioneered by The Guardian.
Another strategy to increase affinity is to give your users the ability to interact and explore more content. For example, flash briefings can be very compelling however they don’t provide much room for the user to actually engage and control what they want to listen to. One great way to add a level of interactivity to your voice app is to add the ability for the user to ask to get more details on a story. This may include a longer in depth commentary or a curation of related content. By allowing the user to ask for related content, the voice apps starts feeling more like a personal editor, rather than a static recording. And a personal editor provides users with real value, resulting in a memorable experience they are more likely to re-engage with.
The role of email
Even when optimizing the in app experience your ability to re-engage a user would greatly benefit from a reminder. Account linking gives developers a way to get the user’s email by asking them to log in or grant permissions to the app. Imagine your voice assistant saying the following, “I’ve found some more stories that focus on [TOPIC A] and [TOPIC B], would you like me to send them to you?”. The assistant can send an email containing more relevant stories, and even explain how they are related to a topic to a story the user was just exploring. This leads to a more memorable experience and helps your voice app stand out in the ever growing crowd. If the email is scheduled to be sent later, it can help keep your voice app fresh in the user’s mind. You can even go as far as scheduling the email before the user starts their morning routine by using their local time. This way, when they are more likely to engage the next time they use their smart speaker.
Discoverability is a serious challenge for voice apps. But by optimizing your invocation phrases for implicit discovery you increase your chances of reaching new audiences. Using personalization and relevance to create affinity for you voice app, increase the chances people will engage in the future. With a measly 6% retention rate for 3rd party voice apps it is more important than ever to focus on designing for engagement.
At Vecgraph, I help brands build their Google Assistant and Alexa experiences. If you’re thinking of taking advantage of this ever growing opportunity but are not sure where to start, let’s get in touch so we can figure out how voice can work for you: marko.arezina@vecgraph.com
