The Bot 101 [ Part 4 ]
The Bot 101 [ Part 4 ]
Dear Bot community members, thanks very much reading the last part in the series of four called The Bot 101. And if you missed the last three, you can find it by clicking The Bot 101. And in this Part 4, we will deep dive on the Bot Design Process which includes the following core steps Use case definition and exploration, Conversation Scripting, Designing and testing, Bot Building, and Analytics and continuous improvement.
And the above given is not the fixed and one design process of bot building and deployment, but for the major bot building ecosystem, it will suit. However, it will differ from the context to context.
Use case Definition and exploration:
It is just a white board discussion whit all the parameters inputs and attributes to design the bots. Lets take an example of use cases from the book of Designing Bots by Amir Shevat. So, the objective is to build two bots that will help someone who takes vacations. Let’s name it as PTOBot (Paid Time Off), and HolidayBot. For the bots let’s identify the following Audience, Business model, Features, Preferred Devices, and very important the platforms. This is the vital design steps in 5 Steps bot designing process.
This is 2nd step in the designing process, where we will write the conversation with our B2B, B2C bots will have. It some time we can call its stories or flow
Once the above flow is detailed and fixed, then very important to have the error handling with the help of the below flow,
And we should be little careful and cautious on the dead lock.
Testing the Design:
This is the very important and vital step with the below three options like show users a video or step-by-step-by replay of the conversation and get their inputs, create a mock or fake bot and let the users play it with, create a working alpha and let the other users work on it.
In this we will be reviewing the different ways to implement our tested bots, which is very much vital step. Bot Architecture is contrast to mobile apps, it’s not installed on the platforms on which they run. Instead, the bot connects to the user or team via set of application programming interface [ API ]. In addition all the bot services does not ever communicate with the client today, it is all done via the proxy of the chosen platform service provides like Slack, Facebook, Kik, Amazon and more. And we have many bot building IDE’s SDK’s for developers and architects like Flow XO, PullString, Chatfuel, Chatflow, Pandorabots, Automat, Reacst.AI, api.ai, Watson, wit.ai, LUIS, msg.ai, Slapp, Botkit, and more.
Analytics and continuous improvement:
In this exercise and this step show us how bot designers have turned this data into insights and adjusted their designs, making their bots better. Usually all the bots connect to a platform like Kik, Slack, Facebook, Messenger, and many others via an API to send and get the messages. It’s very much important to look in to raw logs, filtered logs, analytical tools, SDK, SaaS analytics services, and more to have better analytics and continuous improvement in the bot building exercise. And we have error reduction, conversation optimization, new flows, and flow refresh to have the bot’s improvement. It’s cyclic step, with never ending.
Thanks for the reading the series of The Bot 101, and we will be having series Part III and Part IV in due course. We will talk about Bot Discovery, Process Overview, and multiple use cases in the third series of Bot 101, then we will have final thoughts and deep dive on all steps in the fourth series of Bot 101.
And the conclusion is bots are a great new hammer, but not everything as nail; so we need to explore which use cases are better for this new and great interface. And with bots any enterprise or business can have their engagements with higher degree and value; hence the right bots are bringing best values.