Thankful works with all written channels, delivering a human-like experience and fully resolving customer issues via email, chat, social, SMS, and in-app.
Each customer has their own preference when it comes to how they want to get a hold of customer support. Today’s customers want to have every contact option available for their choosing. This means being everywhere your customers are and delivering support via email, chat, social media, SMS and in-app. Not only does Thankful make this possible, but we ensure that the service provided is exceptional and consistent across all text-based channels.
Email is still the most popular channel for customer support. Customers love the audit trail email offers them and the fact that they can get answers to their problems on their time without having to wait for chat windows. Now you can deliver the best of both worlds – chat speed and email convenience.
Email is native to our technology so our resolution rates are the highest in the business, but you don’t need to take our word for it. It takes just 10 minutes to see Thankful’s AI Agent in action identifying service problem types.
Get started
Thankful already has deep integrations with all the major help desks including Zendesk, Gladly, Kustomer, Freshdesk, Salesforce ServiceCloud, and more so there’s no requirement to change helpdesks to get the best available AI technology. Thankful works inside your helpdesk just like any human agent would, identifying, understanding, and resolving large volumes of tickets from over 50 types. And you can review our AI Agents performance in real-time using all our existing help desk monitoring tools or via our dashboard.
Thankful has pre-built integrations with most business nodes that enable our AI Agent to find and resolve issues, just as your human agents do. This includes major carriers such as UPS, FedEx and DHL, order management systems, shopping carts such as Shopify Plus, Magento, Big Commerce, subscription management platforms Need to connect to a proprietary system? Sure, we do that too.
Customers judge how much businesses value them by the service they receive. For a customer, a major part of being valued is being known and being spoken to as a person not a ticket or a number. With Thankful, automation is a means to more personalization. Our dynamic workflows for email enable companies to add context to the replies and treat the customer as an individual. The response can be augmented with multiple sources of information such as the customer’s purchase history, service history, feedback on previous experiences, account information, and even message sentiment.
Our AI agents are designed to work seamlessly with your support agents. If Thankful can’t take a problem to full resolution , the AI Agent collects all the information necessary and hands it off to a human agent. It can even suggest a reply.
From initial greeting to the final response, the language Thankful’s AI Agent uses is totally customizable to fit your brand messaging, ensuring it’s on-brand any time, every time. Thankful can also support multiple helpdesks for a single brand (ie Zendesk for email and Intercom for chat, or different helpdesks for different regions).
Built by the foremost AI scientists using deep neural networks, Thankful’s understanding is powered by the very latest deep-learning neural networks. This superior natural language processing has been trained specifically for customer service and has already processed over 20 million customer service problems.
This built-for-purpose focus and our enormous dataset enables us to more accurately understand and extract meaning from emails however long or short, misspellings…
The Natural Language Processing necessary to accurately identify and understand what’s being said in free-form text is 100x more difficult than chat.
Sentiment Analysis
Thankful’s custom-built sentiment analysis examines each engagement, ensuring personal treatment such as unhappy customers routed to service specialists quickly or high-value customers are flagged for special offers or service.
Intent Classification
The ability for any agent, human or AI to resolve tickets is based on their ability to identify the nuance of the problem.
“I’m having trouble tracking my order”
requires a different response than
“I’m having trouble with my tracking order link.”
Thankful’s intent classifications not only capture the broad problem type (track order) but the nuances within these broad problems.
Dependency Parsing
Understanding what the customer actually wants is key to solving any problem, especially when there are multiple actions or items to process.
For instance: “I want to return two pairs of blue shorts for green but keep the orange shirt”
Thankful knows to exchange the product : shorts, the quantity: 2, and the color from blue –> green.
It also knows the customer wants to keep the orange shirt.
Actionable Data
Thankful also delivers actionable data from the unstructured voice of the customer that can be used to improve the overall customer journey.
This includes in-depth breakdown of intent classifications, fraud detection, early operational failure alerts, and many more features.
