In my previous post, I’ve introduced the concept of Digital Assistants, or in a more classic definition I’ve introduced the concept of ‘bots’. I provided a simple definition of bot as: an autonomous program on a network (especially the Internet) which can interact with systems or users, especially one designed to behave like a player in some video games. A robot.
We were also able to define a couple of scenarios where our bot may be useful and change how we interact with One Identity products by leveraging their native capabilities, such as natural language support and machine learning algorithms.
While the topic is definitely interesting from a technical standpoint but before I dig in every aspect of how to build a proper bot, I’d like to take a moment and look at digital assistants from a business perspective.
In other words, I’m wondering: are Bots a Business Opportunity?
First and foremost, we have to define the landscape. A bot is not an App. If you define a bot as a mobile app, you’ll fail. A bot is not alternative version of a desktop program. It’s an integrated user interface designed to help you to use a program in a more organic way regardless of the device you’re using.
Let’s step back for a moment and look how the apps landscape was introduced by Steve Jobs in 2008. “YOU are a developer and you’ve just spent two weeks writing this amazing app. What is your dream? Your dream is to get it in front of every iPhone user.”. Quite appealing, isn’t it? Well yes it is but where did we end up? Now, we constantly download and use apps in a race to optimize our work/life tasks/habits.
Is this a business opportunity for software vendors? Apparently yes, but it has many downsides that include:
Those three points make every software vendor shivering when someone asks them to develop a mobile app. Wait, I’m not saying that a software vendor should not develop mobile apps but they should evaluate the opportunity carefully. It can be very crowded out there.
But there’s even a more important indicator that would make you think twice about mobile apps — and that’s the highly desired experience where a user starts a task on one device and seamlessly continues on another.
The apps that enable users to do this are the most successful. This indicates an important trend to me: people much more than a simple User Experience (Ux) based on the simple interface (UxI). Now, they desire the complete experience that includes easy interactions with the app and cross-platform functionality.
Curiously, we spend more time on mobile devices than on any other, so that it is to the point that we expect that everything act as a mobile device and that anything else should follow our mobile experience.
An interesting report is the comScore report
But let’s back to our topic. We defined the mobile application landscape in order to identify the bot landscape with the ultimate need to find a business opportunity.
So let see the indicators of our bot landscape.
First and foremost: A bot marketplace is a bad idea.
Yes I know Facebook has one, Microsoft bot framework has one and there are others but let’s be clear how do you choose a bot? You don’t. It’s not the bot to be chosen but the application. A Bot is a utility attached to the application, hence it’s a bad idea. A bot is part of the app itself.
Let me give you an example: I buy a service, let say One Identity Starling and I subscribe to the Identity Analytics and Risk Intelligence Service (https://www.cloud.oneidentity.com/). This service allows me to find useful insights in the environment I manage and identifies the most risky employees based on their permission, group memberships and other attributes. I find this application useful to the point that I would want additional capabilities, such as reviewing and analyzing group owners, missing data and other user qualities to help me to clean my environment and make it more efficient and tidy.
Now is this an app? It is something you expect to crawl your environment and do the job for you. So here’s the use case: it cannot be an app because requires a certain level of interaction with your end users in order to find, understand and correct the information. That is a bot.
Second: Bot need to offer an alternative, more natural, interface
Yes UX experts may work miracles with UIs but ask yourself how do you use your mobile devices? Do you use a single interconnected app to work or do you jump from one app to another? Do you use the mobile phone to communicate using voice messages, text messages, etc? So what you’re looking for is something that communicates in a natural way and helps you to complete tasks. Eventually, they may do this on their own but on your behalf.
Let me give you another example: Let say you have connected all your applications through One Identity Manager. In a typical Identity Management scenario, you may run attestation campaigns regularly, check any violation on every application through the Identity Audit options, etc. Now, the question is could you extend this scenario to where you may integrate more the ? Of course, you may, but it would require extensive development to extend the Identity Manager interface and engine to accommodate any connected application.
But if you deliver a simple bot based on a framework where any vendor may extend its own application using a standard way to interact with the bot itself, such as the REST API, then you would end up with something that accomplishes the task without development because a bot is already natively integrated with One Identity Manager via the API.
Third: Bot is not an on-premise application.
Thinking of bots like an on-premise market is wrong. Bots are a wonderful way to interact with your users: everywhere, anytime and from any device. They can not be hosted on an isolated, private environment with limited or no access to external, personal devices.
This is quite a view change: we usually think at applications as in a Enterprise vs. Consumer scenario but think about what would stop you from using your personal Amazon Alexa, Microsoft Cortana or any other assistant like Facebook Messenger to interact with your work environment when you’re not at work?
I tell you what probably stops you: security. But today that’s just a hollow excuse – otherwise why do many of us display our personal lives on social media? For arguments sake, let’s say a bot authenticates you securely , communicates with you through encrypted private messages, can reach you any time YOU want and help you to accomplish those tasks that are often forgotten, would your security concerns still outweigh the benefit?
So overall the landscape is this:
So, are bots a business opportunity?
I have a literary response to this, thanks to George Bernard Shaw:
Progress is impossible without change, and those who cannot change their minds cannot change anything.
The ecosystem continues to evolve, new apps product are added and legacy products are dismissed. This puts a constant demand on end-users to learn new tools. The opportunity is to help them to help your company using a common interface that adapts to user needs.
So yes, it’s a business opportunity.
In the third a final post we will look at how we’re building our bots here at One Identity.