Alexa Display Template Skill Deep Dive Part 3: Things That Kill Alexa Skills

My Alexa Display Template Skill Design & Coding Deep Dive series continues with a look at some things that kill Alexa skills before they’re even published. You’ll want to go down this checklist before you invest too much time or effort in any new skill, and the Visual Tarot skill this series is based on is no exception.


Nooo Squirrel


1. Did someone else already write it, or something nearly identical to it?
If there’s a skill that does more or less the same thing as yours already out there, you might want to go back to the drawing board and come up with a new idea. The situation is a no-win. If the other skill is getting lots of great reviews, what reason will people have to try yours instead? If the other skill is getting bad reviews, your skill can be unfairly lumped in with the other skill and not given a chance. If the other skill has no reviews, it’s still a crapshoot as to whose skill will ultimately catch the consumer’s eye and be enabled.

1a. Is there another live skill out there with the same name you plan to use? If there’s an existing skill with the same name as yours, you’ve got even bigger problems because Amazon does not require skill names or invocation phrases to be unique. Consumers may want to enable your skill but accidentally end up enabling someone else’s when they go to look it up in the skill store or Alexa app. You can even end up getting negative reviews that were intended for the other dev’s skill!

Alexa device owners can enable skills by voice too, and this introduces yet another problem. In cases where skills with duplicate titles exist Amazon will not say precisely how Alexa decides which one to enable, other than to say it has to do with things like longevity, user engagement and “other factors”. A skill that’s already been out there for a while will almost certainly have better engagement than your newly-published skill, and it will certainly score better in the longevity department. All of which makes it likely that even if the consumer intends to enable your skill, they’ll get the other one when enabling by voice.

There were already several Tarot skills available when I started on Visual Tarot, but none designed for use on the Show and none that employed Tarot images. Because imagery is such a crucial element in Tarot card readings, I knew my skill had something more to add. I also knew that my title would include the word “Visual”, so it would be distinct from any existing competitors.

A note about foreign markets: when researching for similar skills or duplicate skill names, be sure to check all the Amazon sites where skills are available. A similar skill or one with the same name as yours in the India skill store may not seem like a problem if your skill is released only in the US, but the developer who released in India may choose to release in the US as well, and if that happens all the reviews and usage stats already collected by the India skill will come along with it when it’s released in the US store. In that situation you can end up with a strong competitor you never saw coming.


2. Is it dependent on content that may be difficult or impossible to obtain?
The Echo Show implementation of Visual Tarot is entirely dependent on the Tarot card images. If I had not been able to source royalty-free Tarot card images, my skill would’ve been dead in the water. Luckily, the classic Rider-Waite Tarot deck and associated guide entered the public domain in 2012.

If your skill depends on inclusion of certain audio, video or graphic files, do the research up front to ensure you can legally obtain the needed content, and at a level of quality that’s acceptable. You don’t want to go through all the work of building, testing and debugging, only to find the cert team will not publish your skill due to content quality or intellectual property rights issues.


3. Is your intended implementation a fit for the target user, and target device(s)?
A game like Pictionary won’t work in a voice-only version, whereas a guess-the-sound game doesn’t need any supplemental visuals.

A workout skill that intends to teach the user how to do certain exercises will offer a much better user experience with images or video, and that means a voice-only skill of this type will struggle to compete against similar skills that do employ images or video.


4. Can the Alexa SDK support what you’re trying to do?
If yours is a Display Template skill, check out Amazon’s Display Template documentation to verify the lists and images your skill must show will be presented effectively via the available templates, and that available screen controls can do everything you need them to. It’s not safe to assume the available templates will deliver the user experience you have in mind but even if existing templates aren’t an exact fit, it may be possible to alter your intended design to make the skill work.

Similarly, if your skill will employ the AudioPlayer Interface or VideoApp Interface, it’s a good idea to read through the applicable documentation first because very often you’ll find the features and methods exposed to independent developers are much more limited than you expected. For example, as of this writing (on 1/8/18), you can’t specify ‘album art’ in the Audio Player App, and there’s no native support for playlists in the Video Player App.

If limitations like that could require a partial or complete redesign of your skill, it’s definitely better to find out before you start the work.


5. If you’re hoping to earn money from the skill, will it be published in one of the categories that earns payouts?
Many a dev has been disappointed to learn the skill they were sure would be a hit and earn a big payout from Amazon isn’t categorized under a heading that currently qualifies for payouts.

It’s true that the dev can put a skill into any category at the time of submission for cert, but a little birdie told me Amazon’s making their own, behind the scenes judgments about skill categorization for payouts. Sometimes there’s not a category available that’s a 100% accurate match for your skill; in that case I suggest looking up mobile apps that are similar to your skill in Amazon’s app store to see how they are categorized.

If you’re in this to try and make some money, finding out your hard-won skill doesn’t qualify can be heartbreaking. Click here to see which categories of skills currently qualify for payouts based on popularity and engagement.


That’s it for today. I’ll be back here next week to continue the deep dive.


Other Posts In This Series


Sourcing Alexa Skill Content

A Look At The Visual Tarot Alexa Skill Code

Mapping Alexa Skill GUI Flow

Alexa Skill Image Cheatsheet

Alexa Platform Changes

Persisting Data With Session Attributes In Alexa Skills

Using Alexa Display Template Directives