4 Tips to Choose the Best Chatbot Platform

Countless companies have sprouted from the benefits chatbots promise. It's easy to see why. Chatbots give brands an unprecedented opportunity to answer customer questions at scale. But how do you evaluate if a chatbot vendor is right for your needs?

4 Tips to Choose the Best Chatbot Platform

Chatbots and Live Chat go hand-in-hand.

I've been working in both fields for the past seven years, and as live chat becomes a standard tool for customer service teams in 2020 (see my Three Principles Of A Successful Live Chat Operation [2020 Edition]), it's quite obvious that live chat needs a helping hand from the advances in AI.

Namely, chatbots.

Chatbots have been around for decades, but have only really become useful as a commercial tool in the last few years. A lot of this has to do with the advances in artificial intelligence, especially in language analysis, recognition and understanding.

What is a chatbot?

If you're totally new to chatbots, I highly recommend checking out this extensive guide to "What is a Chatbot". It details the types of chatbot available on the market, listing the pros and cons of each.

Briefly, we have non-AI and AI chatbots. Non-AI is known as a button bot, and gives users buttons to choose from. Then it replies with information based on those buttons. AI chatbots use a form of language recognition to understand what you're typing in the chat window and give you answers based on this.

How to Choose a Chatbot Platform

Countless companies have sprouted from the benefits chatbots promise. It's easy to see why. Chatbots give brands an unprecedented opportunity to answer customer questions at scale as well as a chance to learn customer wishes and feedback through automated conversations.

It's a no brainer for anybody who cares about being a pioneer in customer experience.

This presents a problem for CX enthusiasts. How do you evaluate if a chatbot vendor is right for your needs? Here's a few questions you should ask yourself.

1. Is it easy to use?

Keep it simple. Chatbots are fundamentally layers upon layers of code, designed to recognise your customer's question and pair an answer with it. You can get coders to build it for you, but it doesn't mean you'll be able to easily edit it later on. And you will need to do this. Chatbot is one of those things that are rarely ready, just think about your website and how “ready” that is.

On average, customer service chatbots need about 2 hours of maintenance per week. This time is used to teach the chatbot more question-answer pairs based on customer input, and to review chatbot performance.

The most successful chatbots are built by people who understand the user needs best. In customer service, it's best for customer service agents to build and operate the chatbot. In sales and marketing, it would be the sales and marketing team members and so on.

So it's important to find a chatbot vendor who has developed an interface where anybody on your team can easily and quickly build and maintain you chatbot. No technical coding skills required.

2. What Native Integrations does the chatbot vendor have?

Native integrations are essential if you want to create a rounded customer experience. And the availability of native integrations will heavily influence how useful your chatbot might be to customers.

For example, if you're using live chat, it's probably best to focus on chatbot vendors who support your live chat vendor. You can get started quickly, and don't have to worry about any teething problems when it comes to how the chatbot will work with your live chat services.

If you're using a ticketing system, like Zendesk, a native integration powers your chatbot to do robotic process automation (RPA), like opening tickets and checking ticket status for customers.

3. Can I get Custom Integrations for my chatbot?

You might find a chatbot vendor that ticks all the boxes. But maybe you want to connect to your CRM, so that the chatbot can fully cater to your customers with a totally automated service.

Focus on chatbot vendors who can develop custom integrations for your needs.

For example, if you sell subscriptions, you might want customers to be able to manage their subscriptions quickly and easily in a chatbot, rather than talking to a customer service agent. A custom integration might be built to help you accomplish this.

Another classic example is food delivery. Connecting your chatbot to your inventory, product database and kitchen order system would allow you to offer customers a fast, automated and conversational way to order pizza.

4. Is their Level of Customer Support right for me?

New technologies invariably bring challenges. It's a fact of life. Don't underestimate the importance of good customer service from your chatbot vendor.

Look for vendors who have service level agreements that suit your needs. Talk to their current customers, get feedback.

Your team should be getting training from the chatbot vendor. It might be easy to use but this is not Microsoft Word. There is some learning and best practises to learn and follow. Investing in knowledge and support is one of the most important things you can do with a tool that could be saving your team over 300 hours of work every month.

So find yourself a chatbot vendor who shows they want you to succeed with chatbots. And that has your back.

Summary

Chatbots are a powerful automation tool. They will help you significantly reduce repetitive and grunt work in your team. How you use the time saved is up to you, but to get there, make sure you choose a chatbot vendor that:

  • has an easy-to-use solution suitable for non-technical people
  • can support your live chat quickly
  • is able develop connections to your other services for custom RPA functions
  • is able to prove a track record of outstanding customer service and support

Enjoy this article? Feel free to share it with your colleagues. And hit me up for more tips about chatbots, live chat and RPA in customer service.

--

Full disclosure: I work for chatbot vendor GetJenny and hence my thinking might be biased around chatbots.