Chatbots are coming out tops when it comes to technology for customer engagement. To make sure your chatbot is successful, follow best practices, start simply, and gradually make it more complex as the bot learns. To get started, here’s a blueprint for successful chatbot design.
Before you start building your chatbot you need to nail down why you need a chatbot and if you need one. Spend some time identifying the problem areas that you’d like the bot to solve, for example, handling customer queries or collecting payments.
The clearer your objectives are, the better your chatbot design will be. It’s helpful to compile a detailed list of actions that your bot will handle and keep it specific and realistic.
Include things like which tasks can be automated, and which are better left for agents. Done well, AI-driven customer engagement increases contact rates and reduces the number of inbound phone calls that agents need to handle.
Another key point is to consider, “Who is my chatbot going to talk to?” and this will guide how you design your chatbot.
Basically, there are two main types of chatbots: AI Based and Rules Based (See: How do Chatbots Work)
Simply put, Rules Based bots do well with straightforward tasks and are designed with logical decision-tree flows. (Example: Yes/No; Click 1,2 or 3)
AI Based chatbots use Natural Language Processing to understand what a person is saying and then respond appropriately. They also use Machine Learning to continually grow in their ability to converse naturally with humans. These bots are used in conversational AI interactions.
In the debt collection industry, for example, AI chatbots work well as they can have more nuanced conversations and can pick up a person’s intent and sentiment, which helps when dealing with sensitive issues like debt. But your chatbot may not need this level of sophistication.
AI chatbots need to be trained for their designated purpose and the first step to that end is to collect the necessary data. This may include industry data, transactional data, and historical data from customer interactions with your contact center.
Chatbots draw their language from Large Language Models (LLM). If you’re in a particular industry, there might be a library or LLM available that has the data and learning already collected. Alternatively, you can build your own based on your data or from the foundation of a readily available LLM.
When looking at chatbot platforms, choose one that meets the needs of your business. Be sure to check that it works on the devices your customers use and across the channels you require, and also, investigate how versatile and configurable the platform is. Some popular platforms include Dialogflow, IBM Watson Assistant, and Amazon Lex.
One huge benefit of digital conversational messaging is that it can be done across multiple channels (e.g. WhatsApp, SMS, Viber, Messenger, etc.). You build the bot once, and then deploy it across the various channels, switching between channels and to agents as needed.
All this can be done in one place without losing the thread of the conversation. Decide which channels suit you and your customers best and start with those. There are steps to go through to set up these channels for business, but they are easy to follow. (See: WhatsApp Chatbots for Customer Service)
Now it’s time to get into the actual mechanics of building and training the chatbot.
Do you give your bot a personality?
Before you get into designing the conversational flow, consider the ‘personality’ of your chatbot. Since it will be talking to your customers, you want it to reflect the image of your company and match the type of service or product you offer. Think about who will be interacting with the bot and how to best connect with them.
Do you use an avatar?
If you opt for an avatar, pick one that complements the tone and personality of your brand. For example, would a cartoon animal be too casual, or would a generic face work better? Attaching an avatar to your chatbot gives it a natural feel which makes customers connect easier.
Design the interface
An uncluttered and easy-to-use interface always works the best. Aim to make it simple to navigate, and having both conversational text as well as decision buttons helps customers quickly get to a resolution as they know immediately which actions to take.
You should also consider design elements like font size, colours, images, etc. And if a widget is on your website, be sure that it doesn’t intrude on the page. Depending on the channel, you can include rich media, images and animated gifs into your messages.
Other important points to check are does the bot look good on all devices, especially on mobiles, and is it accessible for disabled people?
Map out the dialogue flow
To get a vision of how the conversation should flow, start with the end in mind and work towards it, for example, I want the customer to commit to a payment, or I want to answer the query. A useful method is to use flow diagrams to visually plan the dialogue. At this point, decide if the flow is linear, or non-linear with multiple branches.
Keep the flow simple and logical with as few branches as possible to efficiently get to the end goal. Don’t ask unnecessary questions with too much back and forth, but rather get to the point as quickly as possible (no chit-chatting) and be highly specific.
Although conversational messaging is a dialogue, giving someone a choice of two or three options can be the quickest way to move along to the next step without confusion.
In the case of outbound messages, a ‘tee-up’ message should be sent first to let the customers know that you are going to send them a message and that it is legitimate. This way, they are more likely to respond.
Write the conversation scripts
As with any conversation, start with a friendly greeting and then move on to the task at hand, while avoiding complicated messages and too many questions. Let the customer know that they are talking to a bot as it will make the conversation work better with fewer frustrations.
Tone of voice
First impressions count! How you start the conversation will set the tone for what comes next and how a person will feel towards the chatbot. How you say something is as important as what you say, and after all, you are engaging with your customers who are the lifeblood of any business.
In the early days of chatbots, they tended to be commanding: “Click here!”; “Pay now!”. But this authoritative voice does not sit well with customers today and it’s better to use the language of everyday conversations. Look at how people talk to each other and what type of language resonates well.
Design conversations to sound human-like and emphasise respect, empathy and consideration. In the end, your chatbot represents you as a company so design it with this in mind.
Another important consideration is how the chatbot handles errors or invalid input. Users should be given the opportunity to correct errors, ask for more details or be routed to an agent.
Including visuals and emojis into a conversation can add personality and make the bot more ‘human’.
(See: Designing Chatbots for Customer Engagement and How Conversational AI Works)
Bias
When designing a chatbot, check for bias and prejudice, especially when it harms or excludes people.
Always give your customers an off-ramp
Customers need a clearly marked way to step out of the chatbot conversation to connect with a live agent, such as a button to click or contact details. Being stuck in a loop with a bot is frustrating and a poor user experience.
Generally, you would design conversation templates that get approved for compliance before they are deployed.
Read here for more detail on the three layers of AI automation for chatbots. Basically, when building a comprehensive AI chatbot, you would include three layers of AI, which are:
Intent Recognition: Using Natural Language Processing to understand what a person wants to do or what they are saying.
Entity Gathering: Picking out named entities from within a conversation, such as dates and post codes to store and use.
Propensity Guidance: Reading between the lines - predicting where a conversation is going and routing it down the best path, e.g. recognising a vulnerable customer and helping them the most effective way.
APIs are powerful pieces of code that can integrate the chatbot with your existing systems, such as your CRM or payment processing software. This will allow the chatbot to access the data it needs to perform its functions and have real-time information available.
The testing phase is crucial to make sure your chatbot does what it needs to do and to prevent potential disaster. Test that it works conversationally as well as technically and that it is compliant with all regulations.
Get your copy of a handy Chatbot Testing Checklist.
Deploy the chatbot in the channels you picked and be sure to communicate the availability of the chatbot to your customers and provide clear instructions on how to use it.
Monitor the performance of the chatbot and refine it as necessary and use customer feedback to improve the chatbot's performance.
A/B testing is a good way to see which options work the best and be open to continuous improvement as nobody gets a perfect chatbot the first-time round.
Delve into the available analytics to keep improving the chatbot, e.g. response rates, engagement rates, successful conversions, response times, drop-offs, A/B test results, customer satisfaction, hand-over to agent rate, etc.
Machine Learning for AI chatbots is a way to teach the bot on a continual basis. Testing data and ‘failures’ are fed back into the system to teach the chatbot new language and conversational responses so that it gets better all the time.
By following these steps, you can successfully design and implement an AI chatbot in your customer communication channels. The chatbot will provide a more efficient and useful experience for your customers, while freeing up your agents to focus on more complex tasks.
If you need to improve your customer engagement, talk to us and we'll show you how AI automation via digital messaging apps works.
You will love the Webio experience.
We promise.
