We are excited to share our plans for 2017. We have ambitious plans to extend the AppSheet platform and to address many of the feature requests that you have asked for. This post lays out our plans in broad strokes.
From the time we launched more than 2 years ago, the development of AppSheet has been driven by our customers asking (and sometimes demanding!) features. We know the broad general direction we want to go, but our customers educate us on the details, what we've missed, what should be improved, and what will really make the platform better.
It might interest you to know how we function as a product development team.
As a team, we have a rough idea of the roadmap for the year. We deploy a new version of the AppSheet platform every workday. In 2016, we must have deployed more than 250 new versions. Because we deploy every day, the incremental change between versions is small and this promotes stability, yet it also allows us to deliver new features rapidly. In the last few months, we have also started deploying new features using a "rollout" process. The rollout initially goes to a small fraction of our users and then we gradually ramp this up to 100% of our users. At any point in time, we have about 10 rollouts in flight.
We expect AppSheet to be a platform for anyone to build apps. That mission statement is intentionally broad and deep.
We know that the current version of AppSheet is limited, but 2017 is going to be all about removing some of those limitations.
When it comes to broad access, we really want everyone in the world to be able to create apps.
When it comes to deep features, the goal is to support as many apps as possible:
It is difficult to describe all the improvements we want to make without presenting a laundry-list of small features. Instead, I've tried to lay out some broad themes to capture the work. If your favorite feature request doesn't fit in one of these themes, don't panic -- these themes guide our work but are not meant to exclude everything else.
Today, AppSheet supports five sources of authentication and eight different data providers. We expect to grow this number significantly. Apps are driven by data, so we want to make sure you can access any useful data to build your apps. As a first step in January, we expect to launch support for Socrata as a data provider. Socrata is a repository of public data sets from various government agencies.
Another important integration point is via workflow rules. We expect to broaden the set of external systems that workflow rules can integrate with, not only "outbound" but also "inbound". Early in the year, we also expect to fundamentally improve the workflow system by introducing scheduled workflow rules.
Currently, the app editor has some basic infrastructure to suggest hints for next steps to take. We expect to make major investments in this area. For the advanced user, the hints should be really specific and useful. For the beginning user, the hints should steer through obvious early steps in app creation. The purpose is to significantly shorten the ramp-up time to create useful apps. As a first step in January, we expect to launch the ability for users in the mobile app to directly create new apps without ever opening the app editor or going to the web site.
Another significant opportunity is to analyze the actual usage of the app and suggest changes to the app creator. Currently, we collect this information and provide some of it via usage reports. We expect to use this data to provide much richer feedback that will guide the app creator. For example: which features of the app are being used and which are not? Can moving some UX controls lead to better usability?
On mobile devices, we will provide the ability to better integrate with native capabilities like bluetooth, payment devices, push notifications, authentication mechanisms, etc.
So far, we've supported apps on Android, iOS, and web browsers. However, the world is changing and one of the really interesting emerging opportunities is the use of apps running within messaging platforms like Facebook Messenger, Skype and Slack. We expect to support as many of these form factors as we can. As a first step in January, we expect to launch the ability for users to create a Facebook Messenger chatbot app using AppSheet. Another way to think of this is that (some) existing AppSheet apps will now be able to run hosted inside Facebook Messenger.
Most of the feature requests we receive fall in this category -- our customers want to push the boundaries of the behavior of the app, especially by having dynamic behaviors. We expect to gradually replace many static app configuration options with dynamic (formula-based) options. This will lead to a lot of expressive power that our advanced users can benefit from. A couple of features in this category coming in January are deep app links and app-wide settings.
We are in the middle of a major overhaul of the app editor web component. In the last few months, you will have seen new version of some panes like the Format Rules pane and the visualization of tables and their relationships. We expect to finish most of this work in the early part of the year.
In the mobile apps, we expect to deliver a number of user interface improvements, in particular richer presentation views and richer data capture controls. In the latter category, we recently introduced Actions as well as QuickEdit. In January, we will be launching a bulk-edit capability so that you can select multiple rows and apply an action to all of them with a single click.
As teams and organizations use AppSheet and build multiple apps, it becomes important to have good tools to manage the team of app builders and to manage the apps that they build. These tools include reporting tools to see the activity of a team of app creators, as well as security and policy tools to build common security whitelists and enforce common settings. In January we expect to launch a mechanism to better manage the deployment of new versions of an app while an existing version of the app is already deployed.
We are constantly updating and in some cases, completely rewriting some of the internal code components of AppSheet. For example, in 2016, we completely rewrote much of the website and the code that handles the presentation of data views in the apps. We are in the midst of a complete rewrite of the app editor as well. We expect to make significant infrastructural improvements in 2017 with a specific focus on app and sync performance. You should see improvements in sync time and in the ability to handle larger data sets.
As you can see, there's a lot that we plan to deliver. We've already at the beginning of January. We can't wait to see all the amazing apps that will be built!