How to Improve Chatbot Conversation Strategy

Chatbots have been around for decades now. From Elizza to Alexa, chatbots have come a long way. However, I think it’s safe to say that chatbots are still growing. The capabilities of a chatbot are unbounded and businesses are yet to unleash the full potential of chatbots.

Owing to technological advancements, building a chatbot today is easier than it was before. The challenge with chatbots though lies is delivering a good user experience — can chatbots mimic human agents in customer service interactions?

Apart from factors such as design or usability, you need to focus on building an effective chatbot conversation strategy. Here are a few tips for you to add to your chatbot conversation strategy to guarantee a good experience.

Before we begin to discuss the tips, let’s take a look at what a chatbot conversational flow is and see why it’s important.

What Is a Chatbot Conversational Flow?

A chatbot conversational flow is a decision tree that gives you a comprehensive list of decisions, events, and outcomes.

Here’s what it looks like:  

You can put together a conversational flow by making a list of the different tasks you want your chatbot to perform. Once this is ready, make a flowchart with the possible outcomes and consequences, followed by the actions the bot needs to perform.

Why Is It Important?

Creating conversation flows ahead of writing copy gives you a clear-cut idea about the actions a user will perform. This, in turn, helps in crafting better copy.

Now that we’ve set context, let’s take a look at how you can improve your chatbot’s conversation strategy.

bots for businesses

Start with an Introduction

A study1 has proven that 75% of customers prefer to know whether they’re chatting online with a chatbot or a human. Some of the participants (48%) even agreed that it felt creepy when the chatbot was pretending to be a human.

It’s a good practice to let your customers know that they’re conversing with a chatbot.  

Your bot can establish its identity with a quick introduction before beginning the conversation. Take a look at how Freshdesk’s support chatbot ‘Freddy’ does this:

Set Ground Rules

While we’ve spoken about the dos and don’ts of designing chatbot conversations before, there’s one point that I’d like to highlight here –

Don’t force the details out of your customer before the conversation begins.

A customer’s need of the hour is to find a solution as quickly as possible. Persistently asking for details at the very beginning will hold your customers back from fulfilling their main objective. After a few attempts at taking the conversation forward, the customer might give up.

Unless absolutely necessary, keep this step to the end and collect the information in one shot.

Here’s an example of a chatbot persistently asking for details:

Set ground rules_ACT

Present The Solutions

To make it more convenient for your customers to begin a conversation, specify the tasks that the chatbot can perform.

Twitter and Facebook chatbots offer ‘quick replies’ a list of tasks the chatbot can perform. Look at how the Messenger bot of Food Network puts this functionality to use:

Present the solution_Food Network

This helps customers whose questions do not fall in the list of things the chatbot can do. Instead of fruitlessly trying to shoot questions at the chatbot, they can go ahead and contact a human support agent directly. This is most helpful when customers have questions about pricing, feature requests or want to discuss edge cases.

Add a Human Touch

Consumer Customer Service Survey2, a study conducted by CGS, revealed that as much as customers want speed, they still prefer human help. Customers often feel that the responses from a chatbot aren’t detailed enough and are less personal.

Apart from providing informative answers, chatbots should be able to upkeep the context of the conversation. Which means, the gap between sounding like a machine and sounding human needs to be bridged. To achieve this, power your bots with artificial intelligence (AI).

Bots backed by artificial intelligence actively learn from conversations and provide more accurate suggestions. There’s also a large scope for data analytics, that comes with artificial intelligence. The bots analyze customer needs and provide personalized responses in return.

Use natural language processing (NLP) methods to sense the tone of the customer and craft conversations that deliver positive customer experiences.

For example, Hipmunk, a platform used to make travel plans, suggests options based on the customer’s location with the help of AI. 

 

Be Proactive

Another space artificial intelligence can help you tap into is proactive support. Take the Slackbot, for example. It guides the user to figure out the product while onboarding. Generally, customers read through materials or watch tutorials and then reach out when they’re stuck. But, the Slackbot proactively supplies information about a feature and is always around to provide immediate help.  

Also, look for indicators of frustration like rage-clicking (clicking repeatedly and rapidly on one point) or dead-clicking (clicking on something that looks clickable but isn’t). Use artificial intelligence to help in identifying what the customer is struggling with. You can program your chatbot to proactively reach out to your customers to understand what they’re struggling with and offer solutions from the knowledge base or provide other content from the web.

Allow for a Graceful Failover

Though chatbots play great assistants to human agents, there might be instances when the chatbot is unable to fully resolve a customer’s issue. In such cases, a support agent needs to step in and take the conversation forward. At times like these, you need to allow for a graceful failover.

For instance, Freddy offers an option to convert the conversation into a ticket. The ticket gets recorded in the system, and an agent reaches out to the customer.

To make it easy for customers to exit the conversation with the chatbot, always provide an option for them to reach out to your support team.

Pro tip: If you’re a support agent, before you reach out to the customer, gather all the context you can. This includes the issue the customer is currently facing as well as past questions the customer might have raised. After repeated attempts at trying to find a solution, the chances that the customer is frustrated is highly likely. By having context before the customer spells out the issue once again, you’re saving their time and ensuring that they are in better spirits.

Gather Feedback

To identify what you’re doing well and what should be improved, you need to gather feedback. 52% of customers from the study mentioned earlier, agreed that they’d be happy to provide feedback after a chatbot conversation. While there are different ways to successfully collect feedback after resolving an issue, here’s how Freddy asks for feedback at the end of the session:

Use the feedback gathered to further improve the user experience. In this case, when a customer picks “no”, reach out to them to find out the reason behind their feedback and work on it accordingly.   

Extend Your Brand to Your Chatbot

A chatbot that sounds dull and mechanic is far from mimicking an interaction with a  human agent. To bring your chatbot to life, you need to give it a personality. Creating a personality that resonates with your brand goes a long way in making customers feel more connected.

Here’s how you can extend your brand to your chatbot:

– Customize the look and feel of the chatbot to match your brand. Apply your brand’s color palette to the bot.

– Tailor the voice and tone of the chatbot to suit your brand. While crafting conversations, set guidelines that fall in line with your stylebook.

Duolingo, an app used to learn languages, has different chatbot personalities for a customer to choose from. The rationale behind this was to cater to different personalities and learning styles.

Bonus Point: Maintain Your Knowledge Base

The key to fuel your chatbot lies in maintaining a rich knowledge base. Your chatbot picks and suggests solutions articles from the knowledge base. Write perfect knowledge base articles, and organize the content of your knowledge base. Unorganized portals are a big pain point not just for your customers but also for your chatbot.

If your business is spread across the world, build and maintain a multilingual knowledge base. Providing your customers an option to read and learn in a language that’s native to them, goes a long way in improving their experience with your service portal.

In a recent blog article, we wrote about how Freshdesk scaled support for 150000 customers with the help of the knowledge base. Take a look at it to know how we succeeded in scaling our support.

Conclusion

The benefits a chatbot brings to a business like offering round-the-clock support, providing fast solutions, reducing the number of incoming tickets, are endless. None of this would be effective if a customer isn’t able to smoothly engage with the chatbot.

To get your bot to truly help your customers with their issues, you need to get your chatbot conversation strategy right. So once this is in place, your team will see more customers getting their issues resolved as well as taking home a pleasant experience with your chatbot.


Main illustration: Siddharth Kandoth

Source:
1 – https://www.mindshareworld.com/sites/default/files/MINDSHARE_HUDDLE_HUMANITY_MACHINE_2016.pdf
2 – https://www.cgsinc.com/en/infographics/chatbots-deliver-speed-but-consumers-want-humans