AI chatbots for customer service have grown in popularity because they give companies a way to better serve people — anywhere in the world, 24 hours a day.
From an internal angle, the cost of providing human support grows each year. Think about having to hire a dozen members for your customer service team full-time, versus a low monthly fee for a customer service chatbot.
When you add emails being lost in crowded inboxes, human agents getting overwhelmed by a high volume of requests, and users not being able to get support in their language, it makes sense that more and more businesses are using AI virtual assistants to help customers.
The biggest challenge these days is choosing the right AI chatbot for your business. There are so many options ranging from no-code to lots-of-code to built-for-you. It can be hard to figure out where to start.
To help decide what the best choice for your business is, we’ve thrown ourselves down the rabbit hole (so you don’t have to) and closely evaluated the 8 best options for AI chatbots. And when we say AI, we mean AI. Products where artificial intelligence and machine learning play a role in understanding human communication and providing the best possible answer to a question.
Right off the bat, that excludes the thousands of chatbots that rely on logical paths, where you, the admin, need to cover every single scenario and provide an answer for it. Those products are not powered by artificial intelligence and it can be a challenge to provide answers to anything more than simple transactions or repetitive questions.
We know chatbots. We didn’t just study the internet for answers before building this list. We’ve built AI chatbots for some of the biggest companies in the world, such as Visa, Sony and GM. We are backing up this list with insights learned from helping users through 200 million conversations across industries like Government, Healthcare, and eCommerce. Because of all this, we don’t just know chatbots, we understand customers and how they interact with chatbots.
By the time you’re done reading, you’ll know what the best AI chatbot for your business is, how much you can expect to spend, and how much time you’ll need to get started.
Ready to explore the world of AI customer service?
Here are the 8 best AI chatbots for customer service
Table of contents
We analyzed each of the chatbots listed above based on the following criteria:
- Is it no-code or coding-based?
- Does it have a free trial?
- Pricing details. Do they charge per seat, per interaction, or other? What’s their lowest and highest plan?
- Power feature. What’s great or unique about them?
Hopefully this guide will give you a good high-level understanding of each of the eight AI tools that we looked at. And you should have everything you need to choose the customer service chatbot that’s going to help you provide better support, reduce costs on customer service agents, and improve customer satisfaction without increasing human interactions.
1. Aivo - Omnichannel support for your customers, wherever they are
Aivo is a no-code, omnichannel AI powered conversational chatbot for a customer-centric experience. You can plug your bot into the various services your customers use, to create a consistent experience regardless of where your customers are contacting you from.
Your Aivo chatbot can engage with people coming from web chats, mobile apps, Facebook Messenger, WhatsApp, and Instagram. It can even automate phone calls made to customer support, helping reduce phone traffic to agents.
Aivo is designed to provide natural feeling conversations, including being able to respond with emojis or videos to give the impression that you’re talking with a person, not a chatbot. You can even give your agent a personality and tone of voice.
You manage the chatbot through the building tool on the backend. You don’t have to come back to Aivo to do more. The platform is as simple and intuitive as possible. And because it’s no-code, getting started is easy.
How easy, you ask? As easy as inputting all the frequently asked questions you get, and the answers to those questions, in a two-column spreadsheet and uploading it to Aivo. You don’t even need to train your bot for the basics or how to respond to conversational language.
To make things even easier, you can cut down on the time it takes to get started with industry packs — Aivo’s power feature. These are built-in lists of the most common questions compiled from millions of interactions processed by Aivo and then sorted by industry. All you need to do is select the most appropriate pack for your business and lots of conversations will be solved for you, out of the box.
Aivo was built with a customer-centric approach. The information collected from each customer interaction, paired with artificial intelligence training tools, easily teaches the bot what it means to deliver a great customer experience.
Aivo learns from every interaction it has with clients and from the decisions it makes. Because of this accumulated knowledge, your bot is always getting better at customer service, which translates to more sales for you and more meaningful experiences for your customers.
On the back end, you get robust analytics to make informed decisions based on the kinds of questions your customers ask. The satisfaction section helps you create satisfaction scores based on the actual expectations your customers have when they engage with your chatbot.
This level of insight helps you understand which questions your bot is nailing, which ones it struggles with, and how to keep training it to get better.
Aivo also has native support for +50 languages without relying on machine translation. Conversations can happen around the world in your customers’ native language and still feel natural.
- Does it have a free trial or free plan? Yes. You can do a 30 day trial on the Starter plan to see how you like it.
- Pricing details: A monthly charge based on the plan you’re using plus a per conversation fee (a conversation is defined as any interaction with the bot).
- Lowest plan: $99/mo + $0.15 per conversation.
- Highest plan: $399/mo + $0.14 per conversation.
- Power feature: Industry packs that help you get started faster.
2. IBM Watson - The Jeopardy winning chatbot that asks for clarification
Watson (from IBM) first made a name for itself as the AI that clobbered the human contestants on Jeopardy! back in 2013. Since then, it’s grown into an industry juggernaut that has been used across industries like healthcare and weather prediction, as well as writing a cooking book: Cognitive Cooking with Chef Watson: Recipes for Innovation from IBM & the Institute of Culinary Education.
Watson Assistant is the low-code chatbot version of Watson for your business. Even though it is low-code, Watson can be more challenging to set up than other options.
You don’t need to have coding skills to get started but the complexity comes from having to understand how customer conversations are structured and implementing some of Watson’s finer details.
The powerhouse feature for Watson Assistant is disambiguation. If Watson doesn’t have the answer, or if there are a few possible answers, it will ask the user for clarity rather than guessing (just like people can). This helps the platform learn better for next time and reduces the frustration that comes with wrong answers when a customer asks questions.
IBM Watson details:
- Does it have a free trial or free plan? IBM offers a free Lite plan, which gets you up to 1,000 unique monthly active users (MAUs), up to 10,000 messages per month, up to 5 Skills (Dialog, Action, Search), 7 days of usage analytics, and session inactivity timeout of 5 minutes.
- Pricing details: Flat monthly rate depending on the plan and MAUs.
- Lowest plan: The Plus plan at $140/mo. This comes with up to 1,000 monthly active users. Additional MAUs are billed at $14 per 100 MAUs. There is an additional cost of $9 per 100 MAUs for users connecting using voice.
- Highest plan: Their enterprise plan, but you have to contact them to learn more.
- Power feature: Disambiguation.
3. Dialogflow - Strong AI supported by Google’s infrastructure
Dialogflow is Google’s entry into the world of customer service chatbots.
Dialogflow is the first code-based tool we’ve looked at so far. It’s a big, hefty platform that you can use to create a chatbot that not only helps you improve traffic, customer engagement, and conversions, but also helps to acquire new customers and retain active customers at lower cost.
But it’s a complex platform. It can be easy to get started with, but hard to master — especially if you have limited technical experience. You can end up spending a lot of time manually noodling around with specifics to get the chatbot you want.
If you’re a smaller company looking to get started quickly, this is probably not the AI chatbot you’re looking for.
Because it’s a Google platform, you can connect everything together to gain access to things like Google Analytics and the Google Cloud Platform to create strong customer service tools.
There are additional costs if you connect to Google Cloud Platform, though. And it can get difficult to know exactly what your billing will be. The best Google does is provide a pricing calculator to estimate your costs before you commit.
There are two versions of Dialogflow. Dialogflow ES (their standard service chatbot) and Dialogflow CX ( the “Customer Experience” advanced version that handles complex workflows).
Dialogflow ES is best for organizations looking to build simple customer service chatbots, while CX does it’s best work with more complex bots as it can handle more customer conversations at once and provides a more robust testing environment than ES does.
- Does it have a free trial or free plan? Yes. It’s free to get started with a Google account. Plus, new customers get a $600 credit for a free trial of Dialogflow CX.
- Pricing details: Monthly pricing based (which isn’t listed) on the version that you’re using and the number of requests made a month.
- Lowest plan: ES Edition which is $0.002 per request.
- Highest plan: CX Edition which is $0.007 per text request.
- Power feature: Integration with other Google services, like Analytics.
4. Amazon Lex - Powerful speech recognition with one-click deployment
Amazon has been a leader in the virtual assistant space for a while now thanks to Alexa. Lex is the customer service chatbot version of Alexa that you can use to help customers get the help they need.
Like a lot of Amazon products, Lex is designed to be user-friendly. It’s a no-code solution that makes getting started effortless — they boast one-click deployment.
Similar to Dialogflow, Lex can be integrated across the various Amazon applications. Like all things related to AWS you need to be aware of additional charges involved with the connected Amazon products.
Plus, if you’re not already an AWS customer (or don’t have an AWS expert on your team) you’ll need technical help to set things up and navigate the complexity that comes with Amazon’s environment.
Because Lex is backed by Alexa, it has pretty solid speech recognition functionality. But that’s not necessarily a good thing when you’re building a conversational chatbot. You want a chatbot platform that has stronger text recognition, than speech to avoid frustrating users writing to get support. Although, Lex is perfect if you’re looking to leverage it in situations where voice is necessary.
Oddly enough, Lex doesn’t offer a lot of native support for additional languages. It currently supports 10 languages, of which three are different versions of English. Not ideal if you’re a company with a truly global reach as customer interactions are going to be limited to awkward machine translations or conversations in a language that isn’t the native tongue of your potential customers.
Amazon Lex details:
- Does it have a free trial or free plan? You can process up to 10,000 text requests and 5,000 speech requests or speech intervals per month for free for the first year.
- Pricing details: Per interaction pricing.
- Lowest plan: $0.00075 per text request (defined as any text input from a user).
- Highest plan: That’s it. Pricing is simple.
- Power feature: Speech recognition.
5. Microsoft Luis - Perfect for Azure-based businesses
Luis is Microsoft’s contender in the AI powered chatbot market.
Luis is a pretty powerful code-based chatbot once you get the swing of things, but you really need to be familiar with all things Azure to get the most of it.
Luis doesn’t have a ton of pre-built models that you can use to get started, so chances are you’re going to need to start from scratch when working with it.
If you have a solid technical background, building from scratch isn’t hard. But there is a learning curve to working with Luis if you’re not a developer. It’s recommended that you spend time going through all the documentation first
Luis’ machine learning capabilities help you build a knowledge base using the most common questions your customers are asking. This automated learning helps Luis better serve your customers and answer questions, without the need for constant manual intervention from your customer support team.
Microsoft Luis details:
- Does it have a free trial or free plan? Microsoft offers a free tier of Luis that gives you up to 10,000 transactions per month. One transaction is a query of up to 500 characters.
- Pricing details: Transaction-based pricing (a transaction being up to 500 characters). Consider you’ll also need to factor in Azure costs when using Luis.
- Lowest plan: $1.50 per 10,000 transactions.
- Highest plan: Same as above.
- Power feature: Active learning that builds a knowledge base.
6. Amelia - A complex bot for complex natural language processing
Amelia claims to offer the most human experience possible from an AI powered chatbot. It’s a strong system but you don’t have a lot of control over it.
Amelia is the only built-for-you platform that we’ve looked at that doesn't provide any self-service options.
They offer two kinds of builds, on-premise or cloud-based, with customizations provided “as needed” according to their site. If there are specific things you’re looking for, you have to work closely with the team at Amelia to make it happen.
Amelia shines when it comes to understanding complex natural language and being able to context switch (thanks to strong natural language processing). This means you’re going to get a chatbot that’s able to intuitively understand what your customers are trying to ask and provide relevant information that helps answer questions.
Beyond not having much control over the build, Amelia really only shows up as Amelia — an overtly smiling blonde virtual assistant sporting a power suit.
You can’t create a chatbot with a personality that matches your business or branding. This is fine if branding isn’t critical for you. For a lot of businesses these days, branding helps boost customer interactions and deliver a more memorable customer experience.
- Does it have a free trial or free plan? No free trial, as it’s not something that you build. It’s built for you. You have to commit before they build you a bot.
- Pricing details: There is no upfront pricing information available for Amelia, so if you want to know more, you have to engage with sales.
- Power feature: Context switching that helps Amelia provide more intuitive responses when helping customers.
7. Mindsay - Give customer service a boost with CRM integration
Mindsay is a relative newcomer in the world of AI chatbots, having been founded in 2016. They offer a no-code tool that helps you get started quickly without the need for a lot of deep technical knowledge.
Because it’s a newer platform, it’s not as robust as other conversational chatbots on this list. The UX can be a bit wonky, which can add a level of frustration to the setup phase.
Mindsay also really lacks in the language department. They currently support English and French, which means that if you’re doing business in parts of the world that speak either language, great. Otherwise, Mindsay won’t be able to deliver a smooth customer experience.
Mindsay shines when you connect it with your CRM to get real-time data checks. Using personalized customer information can give your customer service team a boost and lead to greater customer engagement.
- Does it have a free trial or free plan? There is a free trial but they don’t disclose details for it on their website, so it might be different for each account.
- Pricing details: Beyond being able to easily sign up for a demo with the company, there’s very little information available about Mindsay’s pricing.
- Power feature: CRM integration to extract real-time data.
8. Ada - Drag and drop chatbot builder for beginners
Finally, here’s Ada, another relative newcomer. Ada offers a no-code chatbot solution that lets you build with a drag and drop interface that’s fairly intuitive.
You don’t need a lot of technical knowledge to get started with this chatbot platform. It has strong language support — over 100 languages — and it can help you capture customer information into your CRM.
According to user reviews’ Ada can be very buggy and lacks in terms of reporting. There are a surprising number of reports that you can only get if you email Ada and ask for them. And not just once, but every time you want an updated look at your metrics.
Ada does offer a very strong conversational AI that allows you to create chatbots with in-depth, industry-specific knowledge. For example, if you’re a bank, customers might ask about bank account types. Without the right context, your chatbot could start talking about user accounts, which can lead to frustration.
Virtual assistants that are able to understand the context of each conversation help eliminate confusion and frustration when providing answers to customers.
- Pricing details: the company doesn’t offer any pricing information upfront. Customers have complained that many useful features — often found in other products’ free plans — are locked behind their higher-priced packages.
- Power feature: Strong conversational AI.
Final thoughts on choosing an AI chatbot
Selecting the right AI chatbots for your business is as important as hiring the right employees.
When you think about the time you’re going to spend demoing solutions, training the bots, training your staff to use the chatbots, and integrating it to your systems, you’ll want to get it right the first time.
When to choose an AI chatbot vs. human agents
Thinking about using an AI chatbot for customer service versus bringing more human agents to your customer support team? The biggest factor to consider is volume.
A chatbot can handle way more conversations than a human agent can. Even the strongest multi-taskers reach a wall when it comes to the number of problems they can solve at once.
Unlike humans, bots don’t really have this problem. And if they do, their limits are in the thousands.
When multiple customers need immediate help on different channels, wait times can get increasingly frustrating. Think live chat, Facebook Messenger, WhatsApp, and phone beeping and blinking at the same time. Panic!
Simple transactions can now be fully automated using chatbots, filtering only critical conversations for your team to take care of, inform customers, and deliver exceptional customer experience.
Consider the number of requests you get a day when weighing bots vs. humans. This is particularly important for growing companies that need to scale their product support services while staying profitable.
What we’ve found is that if you’re currently dealing with fewer than 2,500 conversations a month, a human agent is the way to go for your company.
There’s also the time zone issue and the fact us humans need sleep. If you’re running a global business, unless you’re willing to hire humans in every time zone you operate in, chatbots help by providing 24/7 answers for your customers.
It’s not all about the bots, though. Humans tend to make other people more relaxed, which leads to better interactions. People often have a better understanding of the specific context for questions and they think better on-the-fly, making them ideal for conflict resolution.
If dealing with high-tension scenarios is a critical aspect of your customer experience, you might consider adding more human agents to your team.
How to choose an AI chatbot for your company
It helps to look at your future needs every bit as much as your current needs.
If you’re happy as a smaller business, then a platform that focuses on small businesses is fine. You probably don’t need Luis with its full Azure-backed deployment.
If you’re looking to grow your business, explore new revenue options, provide better support at different stages of the customer journey, or even acquire other companies, something that helps you scale is the way to go. This is where Aivo shines.
Create a list of must-haves for your business. Things like multi-language support and an industry-specific knowledge base to help you get started quickly, are going to reduce go-live time, and let you serve more customers world-wide.
Another angle to consider is how each platform integrates with the tools you’re currently using in your business. Especially for sales, marketing, and your customer service team.
The more you’re able to share the data collected by the chatbots, the more insight you’ll have into the needs, desires, and issues of your customers.
Who is Aivo a good fit for
We work best with companies who are either large or growing to the enterprise level. Businesses that have at least 2,500 conversations a month and need a solution that’s going to scale as they keep growing.
Our omnichannel capabilities and language support makes us the natural choice for businesses who are available everywhere their customers are (like your Facebook page), and who speak to customers in their preferred languages.
If you’d like to learn more about choosing a chatbot, we’ve got just the thing for you. We put together a guide on selecting the best AI provider.