EACH BOT RESPONSE IS ITS OWN BRAND EXPERIENCE OR MOMENT OF TRUTH
Following on from our last post, which introduced the dangers of launching an infant chatbot without proper consideration, and looked at the way users utilise bot services, we’re looking at two of the ways bots can function: as a mariner, and to build micro journeys.
1. Use the bot as Mariner
Tracking alongside the bot Intimacy Spectrum is the perceived purpose and subsequent use of the bot by the user. From providing FAQ information such as opening times, facilities and cancellation policies, users use the bot as a navigational tool. The bot used in this way emulates onsite search but actually shortcuts the search facility by fast tracking users to site content.
Retail brands such as Wolford have been captilising on this behaviour over the last few years by using such bots and messenger as Personal Shopping assistants. This type of bot primarily presents responses as content modules with options for the next step or action as part of the response. By offering users options as well as the facility to type in their next response, the users are guided through the experience in way that requires very little cognitive load.
When the bot is being used in this way, in Mariner mode, we see a light weight, sign posting bot response. This pattern is true even when the bot is seeking response to get to a prompt from the user and when responding via a multi-stack utterance (a series of back to back short responses sent from the bot).
Chatbots mean many things to many people. Decide up front on life stages of bot with a corresponding training roadmap to nurture your bot as it matures from FAQ bots to personal assistants and transaction support.
2. Build out Micro Journeys
Irrespective of where the user sits on the Intimacy Spectrum and of the Gamer and Mariner user-bot
roles, there are three layers of a bot response to construct to ensure conversational flow.
At the heart of the bot answer response is the short ‘core’. This acknowledges and directly addresses the question being asked and is an affirmative or negative to reinforce question understanding. In its infancy, bots should be designed to deliver a short content neutral response for a strong baseline from which to refine from.
Over time, bots learn using a machine learning service that builds natural language understanding which helps to assess intent which then determines the actions that the bot takes. In the meanwhile, we can manually prime an infant bot to intercept intent by adding two addition layers: the ‘So what’ and ‘Next steps’ to the core of the response.
These are the onward journeys in the form of content links and are directly related the KPIs of the bot e.g. conversion. Where content modules are not being used to provide these options, we must depend on copy. Where they are being heavily utilised, we can see visually how the micro journey model holds true.
The core and the onward journeys together constitute the micro journeys, and thus the micro brand experiences, of the bot response.
Using bots as part of brand experience
We have identified four bot-user interactions that if balanced correctly impart a more positive brand experience than if they were not considered when launching an infant chatbot. Each bot response is its own brand experience or moment of truth. The impact of a chatbot user experience is amplified because unlike pull only mediums, the bot is talking back and is being reacted to in a very human way.
The interactions have been stress-tested against more mature bots and the interaction patterns hold true. In the case of Personal shopping bots, two of the four bot patterns: Bot as Mariner and Chatbot Micro Journeys form the foundation of the experience.
Since chatbots can be used to render across voice as well as text, these four bot-user interaction patterns have direct implications for VUIs. The effects are even more pronounced due to varieties in voice intonation, pitch and tone.
We’re about to launch the first iteration of our chatbot as an MVT (Multi-Variant Test) to validate success. All of our tracking and KPIs (Key Performance Indicators) are in place and usability testing has allowed us to test drive the data collection and analysis. We’ve built in continuous bot training to analyse the bot-user conversations, identify trends and optimise the responses and bot trigger scenarios.
Companies are keen to invest in Enterprise Intelligent Assistants - $4.5 billion by 2021. (Opus Research, 2017) but like our client they will need to see a return. We are looking forward to our bot analytics and the value that these four bot-user interactions have added. Let us know how you get on.
Behind the scenes
The primary objective of our bot is to reduce call centre volume and the secondary, to increase sales. We’re using the Microsoft Axure Bot Service and working directly into the backend database – the QnA maker knowledge base. Each answer aims to be 100 characters or less due the size of the chat window and mobile affordance. The Q&A pairs are tagged with keywords for matching, analysis and machine learning.
We created a beta by importing the existing site FAQs into the knowledge base. We then mined Customer Service call logs and site stats to identify those questions which if answered would deliver the most business value:
1. The most commonly asked questions
2. The most time-consuming answers
3. The Q&A pairs that held the highest potential value of conversion
The aim, as the bot is used, is to cleanse out any underused question intents and to decrease investment in that content since it represents little return.
About the author
Our Rufus Writers are members of our amazing team with years of industry experience delivering considered solutions to all kinds of creative challenges.
Read the latest inspo from our team