Struggling to keep up with increasing customer interactions? Automation is a great way to contain the masses, freeing up your Salesforce agents to focus on the more complex, high-value issues. But to meet your customer’s rising expectations, you need to ensure your automation-first CX is up to par. 

Follow these tips to design a conversational AI experience your customers will love—and your agents will thank you for.

1. Let users know they’re interacting with conversational AI

A well designed automation-first CX solution can be just as effective as your best agent, but that doesn’t mean you should call it Joan and pretend it’s a real person. Let customers know upfront that they’re chatting with conversational AI so you can manage their expectations. 

That being said, you do want the conversational AI to sound like a human. Be sure to flex your brand’s voice and tone when scripting its answers, and use simple, conversational language that’s clear and concise.

2. Make your automated interactions personal

People expect brands to know them, and this applies to automated conversations too. When you’re building your conversational AI experience, develop templates that enable the bot to personalize interactions. Consider all the factors you can personalize for—things like age, gender, and location. 

Call the customer by name, and think about how to change your message based on what you know about them. A major advantage of using Ada with Salesforce is the ability to automatically pull personal details into the conversational AI. The more personalized the conversation, the more human it feels—and the happier your customers will be. 

3. Build-in backup support

Regardless of how good your automation-first experience is, agents will sometimes need to step in and assist or take over the conversation if necessary—without interrupting the flow. 

Ada integrates seamlessly with Salesforce to make this process as smooth as possible. When a customer types something that’s a red flag, such as, “I want to cancel my account,” Ada notifies an agent who can jump in and offer help. The agent sees the whole conversation, so they can pick it up without missing a beat. And because the UI stays the same, the experience is seamless.

4. Head off bad experiences

Automation is also a great tool for detecting customers who are struggling, and proactively offering assistance. For example, if a web visitor stalls out part way through ordering a new service, the conversational AI can prompt a message: “Hi! I see you’re trying to place an order. Can I help you with anything?” 

Instead of waiting for a customer to complain, whether directly or with a bad review or a social media rant, get ahead of the curve and turn a potentially negative experience into a positive one.

5. Put a positive spin on answers

The way you say things has a powerful effect on CX. If a product is on backorder and the chatbot says, “Unfortunately that item is out of stock,” the buyer will probably feel annoyed. 

Reframe it in a positive way instead: “That item is available next month. Would you like to place an order today for shipment as soon as we get it?” Choosing conversational AI with natural language processing (NLP) built in helps to elevate the experience--it’s better able to understand what customers mean in their own words. People appreciate a bot that “gets” them the first time.

6. Keep customers informed

You can also use conversational AI to make sure customers know about new products or promotions you’re offering, or if a product they’re interested in is running low on stock. The chatbot can proactively tell visitors about relevant deals or product info as they navigate your site. Or you can insert this information organically into chatbot conversations, giving you the ability to drive cross-sells and upsells. Either way, it’s a great opportunity to increase revenue without increasing costs.

7. Bring bots and agents together

They might just turn out to be the dream team. Automating the mundane, repetitive tasks that make up the bulk of customer interactions allows agents to focus on more complex issues, which leads to higher job satisfaction. And with a chatbot front-ending interactions, agents can start the conversation already understanding who they’re talking to and what they’re trying to do, with relevant customer details at their fingertips. 

This saves time and effort on both sides. Plus, agents can use the conversational AI to look up customer info while they chat to speed things up even more. A bonus feature of using Ada + Salesforce together is that agents receive a transcript of the conversation history before entering a live conversation with the customer. 

8. Keep a pulse on how you’re doing

Satisfaction drives loyalty and retention, so it’s important to measure. Automating CSAT surveys makes this easy. 

With Ada’s new CSAT Pro feature, you can go beyond the basic “how would you rate your overall experience” question to find the root cause of issues. By adding customizable follow-up questions and letting customers comment in their own words, you can fine-tune experiences and measure improvement over time. 

There’s even an option to trigger a response, such as connecting to an agent or offering a discount, when a customer types a negative comment—turning insight into action.

9. Route interactions for a better experience

Conversational AI can prioritize and route interactions to the correct department or agent. By asking basic questions up front, it can determine what users need and where to send them for the most efficient service. 

Connecting customers to the right agent boosts first contact resolution (FCR), which is closely linked to CSAT. You can also set parameters to prioritize urgent interactions and send frustrated customers to the front of the line.

10. Train your conversational AI using your CX stars

Your customer service and experience pros know your company and customers better than anyone, and they’re the ones you want to train the conversational AI to recognize and answer common questions. 

Your agents know what people are likely to ask about, and how to best respond. They also understand what makes customers happy and what makes them mad. Be sure to choose an automation-first platform that’s easy for non-technical people to use, so you can put your best asset—your service experts—in charge of the automated experience.

New call-to-action