How do you tell apart the merely good chatbots from the truly great conversational AI? Here are 10 features to look for.
The virtual assistant market today is getting saturated with tech giants developing new products and startups entering the field with new innovations. We already looked at the 3 must-have features of every conversational AI system today. These can help you form a minimum benchmark to start with when exploring potential solutions.
But it still leaves several vendors with their unique value propositions. Here are 10 more features that can help you separate the merely good solutions from the truly great ones.
Relevant Industry Language Model to be Enhanced with Company-Specific Vocabulary and Jargon
Recall that one of the AI techniques that make conversational agents intelligent is Natural Language Processing. The sophistication of the NLP models used in the bot makes a big impact on how smart the conversational AI can be.
|Natural Language Processing is the ability of the engine at the core of the conversational AI that allows it to understand, manage and reply to users’ queries. It uses algorithms to extract rules in human language to convert it to a form that machines can understand.|
Not all NLP models will be able to act on users’ queries immediately out of the box. In fact, the cold start problem is prevalent, where there is a lack of clean and labelled data to start the training process.
In addition, every industry has its own jargon. The most frequently used phrases in finance and banking for instance, will be drastically different from the complex medical terminology in the healthcare industry. Moreover, the vocabulary adopted to respond to customers questions will also be different from company to company. Think how the same product, say a healthcare screening package, can be termed by different private hospitals in different ways.
The NLP models therefore need to be pre-trained with industry knowledge bases to a certain extent. It should then be able to ingest a company’s vocabulary through historical chat logs, product or service collaterals and internal documents.
In evaluating NLP models in conversational AI solutions, the aim should not be to compare one model to another or to rate how good the model is on its own. It is just one component in an overall AI tech stack that includes data collection and structure, hardware capabilities, orchestration layers and human expertise.
As an analogy, think about the question, “How good is Google Search in producing relevant results to my search queries?”. The answer does not depend on the ML and NLP algorithms alone, no matter how sophisticated they may be. It also depends on the available web pages relevant to the query, how the content on those web pages are structured and organised and many other factors.
Computing Hardware and R&D
Even the best NLP models can fall short if you do not have the right hardware do to specialised training. The amount of computing power used in AI training has been increasing exponentially. OpenAI, a nonprofit research organisation showed that this compute power doubles every 3.4 months.
Companies planning to adopt AI technologies should look at the long-term implications of this trend. Developing industry ready AI solutions require a lot of experimentation and tweaks to models to achieve the desired level of performance. Many research papers describe experiments that yield good results within a narrow scope and will require a substantial amount of work before it can be applied to solve real world scenarios.
The best conversational AI solutions will not only be perfecting their software, but will also be investing heavily in R&D and computing resources to prepare for this exponential growth.
Dialogue Scripting for Empathy and a Smooth User Experience
While the NLP models influence how efficiently the bot responds to user queries, a good conversation requires much more. Consider the below interaction between an anxious user and a bot.
|User question:“My child has a cut on his finger. What should I do?”|
Chatbot response:“For a small superficial cut, rinse the wound thoroughly and dab dry. Cover with a sterile adhesive bandage or gauze. Change the bandage frequently(at least 3 times a day) and monitor the wound closely.”
While the response is factually correct, it lacks any warmth. Contrast this with the following response.
|User question: “My child has a cut on his finger. What should I do?” |
Chatbot response: “Stay calm and observe the cut carefully. Is it a small superficial cut? If so, rinse the wound thoroughly and dab dry. Then cover with a sterile adhesive bandage or gauze. Change the bandage frequently (at least 3 times a day) and monitor the wound closely.”
The above example is quite simplified but it shows a stark difference in the tone used. The extra touch of humanity and empathy for the situation leaves a positive impression on the user. This is done through clever dialogue scripting and conversation design and should be an essential part of a good conversational AI.
In the unique conversation interface, having all the features in the world will not matter if the user is unable or unwilling to make their way through them. Hence, designing a good dialogue is very essential to allowing the user to discover and make good use of the chatbot’s capabilities.
The target audience for any business solution is never a monolith. In today’s globalised multicultural world, customers can vary in their demographics and language preferences. This is especially true if an enterprise is moving into or has its eyes set on new markets. In these cases, the target audience may use a language that is a mix of English and local dialects, or a different language altogether.
For example, a U.S firm that wants to expand into Southeast Asia will have to deal with customers who tend to speak Singlish or Manglish, or who prefer Malay or Thai). The conversational AI will need to be trained with data in their target languages to be able to respond to their queries. Machine translation currently may not deliver the level of performance that companies require.
English being a high-resource language has large quantities of publicly available data sets that can be used for training. This may not be the case for the local languages. Look for vendors with proprietary NLP models which can handle mixed languages, and who have gathered a large corpus of native language datasets over time to train their language engines.
|PRO TIP: Look for vendors with proprietary NLP models capable of handling mixed languages, and who have gathered a large corpus of native language datasets over time to train their language engines.|
Beyond Keywords to Contextual Understanding
The simplest chatbots rely on keyword matching to extract responses to questions. The user asks a question with certain keywords and the bot looks for literal match in a database to return the corresponding response. The best conversational AI solutions must go beyond that.
Specifically, they should be able to perform a semantic search. Semantic search involves understanding the user’s intent and contextual meaning of terms as they appear in the database. For example, consider the following exchange.
|User question: “What is the weather today?” |
Chatbot response: “It’s 31 degrees Celcius. Bright and sunny with a chance of light shower.”
User question: “What was it like yesterday?”
Chatbot response: “It was 29 degrees Celcius at this time yesterday. Cloudy and raining.“
Note that the user did not have to repeat the word “weather” again in the follow up question. The bot was able to understand the meaning and context of the conversation to provide the response. This capability also allows the conversational agent to understand the different ways in which different users would ask the same question.
A user can also get distracted and go out of the flow to ask another question. It will be useful to bring the user back if need be to complete the conversation flow.
A Unified Experience for Multiple Stakeholders
Companies today have to manage many more stakeholders than before. Single use case bots that only do one thing or serve one customer type can very easily hit a ceiling for large enterprises. They are also harder to maintain over time. Hence, it is important for the platform to be able to manage multiple bots, and know when to send specific information to each user.
In some cases, different users might ask the same question but the responses often vary depending on their characteristics and personal information. For example, members of three different departments in an organisation could ask the same question about the benefits they are entitled to. But the answer could vary based on their pay grade, department, job function and type of role.
The conversational AI should therefore be capable of providing conditional answers. This means that not only do the users get responses relevant to them, it also cuts down the number of intents to improve the AI performance and reduce training effort.
Ease of Training by Users
One of the golden rules of testing a conversational AI solution is to approach it as an ongoing process rather than a one-time activity. This means regular monitoring, review, and update of the training data to improve the bot’s performance.
This ongoing training during testing must involve the key stakeholders in the organisation, especially those who are closest to customers. Based on their input, admins will have to upload, edit, and manage intents and the content. Conversational AI solutions should include such an element of flexibility to allow the users to train the bot based on their subject matter knowledge.
If each change takes a lot of time or if the company needs to rely on the vendor for all types of changes, it will not be efficient. Businesses today need to be nimble and be ready to update their communication channels often. The interface needs to be easy for non technical users to use. The best software interfaces are those that require minimal training.
Integration with CRM, Knowledge Base Systems or Backend Systems
A conversational AI solution that is a standalone system is hardly better than a rule-based bot. If the solution only has access to the user queries and a static database of pre-written answers, it will not get far.
Instead, integration into other internal systems like CRM databases and knowledge bases can take the chatbot to a much higher level of utility. In healthcare institutions for example, integrating a conversational agent to the electronic medical records (EMR) allows it to personalise responses based on the users’ medical history, previous treatment, allergies and more.
In an ideal world, the conversational AI system will be integrated to all other digital systems. But this may not be feasible, at least in the initial stages of deployment. IT teams have to work with the bot development team to figure out which of they internal systems are open and ready for integration.
They also have to find the best method of integration based on time and cost considerations. A comprehensive integration through back-end coding may take a few months but could result in a robust AI in the end. In contrast, a low-code development through API connectors could get this done in a few weeks and deliver a greater ROI.
Customisable Analytics for Decision Making
In the ongoing testing of the bot, the goal should be to iteratively improve performance. Performance can be gauged differently by different functions within an organisation. For example, the customer service team might prioritise customer satisfaction above all else. The marketing and sales functions will be interested in the total leads generated. The operations team on the other hand will want to know the peak hours of the day so they can coordinate resources accordingly. The conversational AI system should be able to provide insights that can inform each of their decisions.
To this end, the system should allow admin users should be able to monitor the bot usage and activity. They should be able to study analytics for different time periods to make business decisions. Common analytics metrics that are relevant to track include user statistics and profile, customer satisfaction, most frequently asked questions, and percentage split of answerable to unanswerable questions. A certain level of flexibility should also be available, in order to cater to the changing goals within an organisation. This means giving the organisations the option to customise the analytics to decide what metrics they want to monitor.
Seamless Live Agent Takeover
There may always be scenarios where the conversional AI is unable to respond to a user query effectively. The bot might be in its early days of training, the question may be out of scope or there may not be an easy answer available in the database of responses. In other cases, the customer may simple request to speak to a live agent.
In such cases the most appropriate thing to do would be to hand over to human to continue the conversation. The best conversational AI solutions should have this fall-back option available, to seamlessly transfer from the automated chat to a human agent. This means being able to find the right human agent with the skills relevant to the customer’s question and to allocate resources based on the agents’ availability.
But there is more. Once the live agent takes over, there is still more that can be done to train the bot. When the live agent responds to the queries or after the chat, he or she can be given the option of tagging each user query to the appropriate intent if there is one to found. This ensures that the bot is trained to treat these queries as variations of the selected intent for future cases.
Whether you are building a conversational AI team in your organisation, or look for an external vendor to partner with, these 10 features are crucial in order to create a best-in-class solution.
To learn more about how KeyReply approaches these critical features, talk to our expert today.