Conversations are an essential part of nearly all businesses — and depending on how you leverage those conversations, they could either be limiting your growth or accelerating it.
We’re going to be focusing on the latter today, but first let’s spend a moment on the former…
Conversations that limit your growth.
The average live chat operator can handle about six conversations at once. That means that for every six people asking questions on your website, you need another live chat operator.
60 people asking questions? You need 10 live chat operators.
600 people? Get 100 operators.
6,000 people? Yikes.
As you can see, these kinds of conversations can quickly become a bottleneck to business growth. And this is especially true if you have employees who are answering the same questions, providing the same information, and solving the same problems, day after day.
But fortunately, there’s a solution to this problem.
They’re called chatbots. And the great thing about chatbots is that as long as you have the answer to a question, you can use a chatbot to provide that answer to as many people as you want—whether it’s 60 or 600 or 6,000.
And when you start using chatbots to automate those repetitive conversations, you get much more efficient. You can start using conversations to actually GROW your business, instead of just using them to maintain the status quo.
This is called conversational growth.
The awesome thing about conversational growth is that it doesn’t JUST help your business to grow. It helps you to grow BETTER.
Because instead of blasting out the same impersonal message to more and more people, conversational marketing empowers you to build 1:1 relationships at scale.
This is really powerful because it allows you to make a huge number of customers feel heard, special, and unique…without allowing that relationship-building process to become a bottleneck in your business.
So how do you put conversational growth strategy into practice in your company?
At HubSpot, we’ve created a 5-pillar framework for implementing a conversational growth strategy. We call it SCOPE: Standardize, Contextualize, Optimize, Personalize, and Empathize.
Pillar 1: Standardize for Consistency
The first step in any conversational growth strategy is to gather all the information you need to deliver and standardize it. You want to make sure that you’re delivering information that is factually accurate and consistent across all your channels.
One way to approach this is to think about how you would train a new employee.
What would you tell a new sales person on their first day? What things would they need to understand about your company and your products in order to do their job?
Another good idea is to talk with people who frequently interact with customers—like customer service reps. Get their feedback on the kinds of questions that people are asking over and over.
These are the kinds of questions you’ll need to answer as part of your conversational growth strategy.
Create a FAQ for your business. Collect the most commonly asked questions and the responses you consistently give customers (such as how to reset their password, track their order status, or hours of operation, etc.) and keep these updated as a resource for all employees to use for consistent answering.
Pillar 2: Contextualize for Relevance
In Pillar 1 you gathered and standardized all your important information. The next step is to put that information into the proper context of a conversation.
In order to do this, it’s helpful to think about the types of questions that people ask. There are eight different ways that someone can ask you a question:
That’s it. Every possible question will fit into one of those eight categories. And depending on the type of question, it’s your job to provide the answer in a way that fits the question that’s being asked.
Here are a few examples of questions that might fit into each category, along with the type of answer that the person is looking for:
- Who – the answer will involve a person, group, or team. “Who do I talk to about my invoice?”
- What – the answer will involve a description or definition. “What is the difference between a latte and a cappuccino?”
- Where – the answer will involve a location, either geographic or relative. “Where is the nearest Target?” or “Where is the reset button on my device?”
- When – the answer will involve a time. “When does the dry cleaner open?”
- Why – the answer will involve a reason or explanation. “Why do I keep getting this error message?”
- How – the answer will involve a description. “How do I change a tire?”
- Which – the answer will involve selecting one item from a set or array. “Which plan makes the most sense for me?”
- Yes/No – the answer will involve a binary response, often followed by some reasoning or explanation. “Can I return the product if it breaks?”
In some cases you won’t always be able to answer a question right away. For example, let’s say the question is:
How do I change my password?
To answer this question, you may need to ask a qualifying question first. Namely: “Are you currently logged in or not?”
You have to ask this first because in most cases, the process for changing a password is different for people who are logged in vs people who aren’t. As a result, the answer will change depending on that status.
Now if we change that original question to:
Where do I change my password?
That’s a different question altogether. In this case the person is looking for a location—a link or a URL that will take them to a profile page where they can change their password.
In both cases the end goal is the same (changing their password), but the context is different. And so you have to vary your answer based on that context to give the person what they’re looking for.
Context is something that probably seems like common sense to most of us. Because when you’re in a one-on-one conversation, you understand the context of most questions intuitively. You don’t usually have to stop and think about whether the person is looking for a who, what, where, when, why, how, which, or yes/no answer.
But when you’re building out a chatbot, you won’t be there in person to figure out that context. It has to be programmed in. That’s why it’s important to understand these kinds of questions, so you can make sure that your chatbot provides the kind of content that fits the questions people are asking.
Use the FAQs you put together earlier to think through all the questions a customer could ask you, in each of the eight different ways. Write out the most common ones your customer service reps receive, and think through the intention of the customer when they’re asking questions like this.
Pillar 3: Optimize for Clarity
Optimization is all about delivering your message in the right way, given the delivery channel.
And you want to adjust the way you deliver your messages based on the strengths and weaknesses of that channel.
For example, if you’re answering a question via email, one disadvantage of email is response speed. Compared to other conversational channels like live chat, Messenger, or phone calls, email conversations are typically slow (with a response time around 1-24 hours). So you would want to mitigate that weakness by trying to cut down on back-and-forth as much as possible.
An advantage of email is permanence and searchability. Once you send someone an email, it will stay in their email account until they delete it. So you can take advantage of that by sending detailed instructions or even attachments (such as login information) that people can search for and reference whenever they need.
When the channel is Facebook Messenger, the opposite is true. Speed is a strength. You can ask and answer questions pretty much in real-time. A disadvantage, on the other hand, is the screen size. Most people use Messenger on their phone, and there’s only a certain amount of text that can fit at once:
These are important things to keep in mind when delivering content via Messenger. You don’t want to send people 500 words of content at a time. Instead keep your answers short, and encourage lots of engagement by asking questions.
For example, let’s say that somebody asks you a how-to question—like “How do I change the oil filter on my car?”
There are a couple ways you could approach this using Messenger as your conversational channel.
- You could start by asking some qualification questions first. Maybe you want to ask the make, model, and year of their car so you can tailor the answer to the kind of car they drive. This way you can show them exactly where their air filter is and how to change it.
- You could start by giving them a short summary, then ask if they’d like a more detailed description. If they answer “Yes,” then you could give them a link to a longer article or blog post with more information. Or, you could ask for their email address and send a PDF, checklist, or other resource to their inbox. This is a really natural way to ask for someone’s contact information.
- You could answer the question directly in Messenger using rich media. The ability to embed images, audio files, and videos right inside of a Messenger chat is another big strength of this communication channel.
Optimize the way that you deliver your information so that it leverages the strengths of the channel you’re using. On Facebook Messenger that means sending short messages, asking lots of questions to keep the engagement high, and leveraging videos and other rich media.
Pillar 4: Personalize for Impact
When you build a chatbot, that chatbot might be having the exact same conversation with 50,000 people around the world.
But you don’t want people to feel like that’s the case. You want each of those conversations to feel unique and personal.
That’s what Pillar 4 of SCOPE is all about. You want to leverage any information you’ve already collected from the person to deliver a more personalized conversation—one that helps you to strengthen your relationship with them.
The easiest example of this is calling someone by their name. “Hey, Mark!” is going to sound more personal than, “Hey, you!”
You can also reference any previous questions that you asked people (as long as you saved the answer). For example, Domino’s has a chatbot that remembers your order history. So you can track a previous order or place the same order you had last time.
(Perfect for all those Pepperoni and Bell Pepper Fridays.)
The sort of information you use for personalization will vary from company to company. A real estate chatbot, for example, might save each potential homebuyer’s price range and preferred neighborhood, so that they only see home listings relevant to them.
Little touches like this take more work to set up, but they go a long way in making your conversations feel more personal and helping to build stronger relationships with your customers. How can you personalize your chatbot to improve your conversation? Think this through and document it out.
Pillar 5: Empathize for Compassion
Until now, we’ve mostly talked about how to use conversations to deliver the right information in the right way. We talked about making sure the content you give people is factual, consistent, put in the right context, optimized for Messenger, and personalized for each unique user.
But conversations are about more than just dry facts. There are emotions involved, too.
And that’s why, many times in a conversation you need to give the emotionally right response before you deliver the factually correct answer.
When someone expresses an emotion to you, you want them to feel like you’ve heard them. You want them to believe that you understand and acknowledge how they feel.
There isn’t anything fancy involved here. It’s just being human.
And it’s another one of those things that most of us will instinctively do in a real conversation—but when you’re building out an automated chatbot, you need to make sure to take some time for empathy.
So if someone engages with your chatbot and indicates that they’re having a problem with a product, don’t start out by immediately asking what the problem is. This person is probably feeling frustrated, and maybe even angry. So first take just a second to acknowledge and empathize with the way they feel.
Baking in empathy to your chatbot will go a long way in making your customer feel heard, which will put them in a better frame of mind to consume the factual information that follows. Think about being in their shoes and the kind of response you’d want if you were upset and trying to find a solution to this problem.
Not Just More Conversations—Better Conversations
The whole concept of conversational growth kind of implies that your business will have more and more conversations over time.
But it’s important to understand that just having more conversations, by itself, isn’t enough. (In fact, as we discussed at the beginning of this article, too many of the wrong conversations can even become a bottleneck to growth.)
To really leverage this conversational growth strategy, you need to have better conversations.
A better conversation doesn’t just give your customer the information they’re looking for. It makes them feel good about your company and your product. It helps them to know, like, and trust you more than they used to. And it makes them a little more likely to do business with you again in the future.
That’s what SCOPE is all about, and that’s why I really urge you to give this framework a try when you’re building out your own conversational growth strategy.
Want more strategies like this for generating leads, subscribers, and sales from Messenger Marketing? Check out ManyChat’s new Messenger Marketing & Chatbot Mastery course! It’s 10+ hours of free video content, plus 15 Messenger Marketing blueprints you can implement today! Learn more here.
Brian Bagdasarian is the Conversational Strategist and Inbound Professor for Conversational at HubSpot, Inc. – where he guides the development and implementation of conversational strategy and its associated tools, including chatbots, live chat, and more to create an omni-channel, personal experience, at scale. Brian joined HubSpot via the acquisition of Motion.AI, where he was the Chief Evangelist Officer, and which, at the time of the sale, was the largest visual chatbot development platform in the market, with over 74,000 bots built.