I think they are one of the top time saving software systems that has recently become really mainstream. Chatbots save you time by automating responses to questions by your users. This can be questions about sales, support, or a variety of other things. In addition to the automation that a chatbot enables, virtually every chatbot also has a chat system that enables multiple representatives from your company to chat with different users simultaneously.
The big value for chatbots are really in the automation of information deliver though. Every minute your team is not working on supporting clients with issues that they could find through other methods is money in your pocket, so having a system that users are happy to work with that delivers the answers they are searching for suddenly becomes a profitable solution.
Chatbots have a few foundational features that are built into almost, if not all chatbot systems. These features include:
- Install code
- Data connection
After your account is set up, all you generally need to do is install the code on the header or footer, depending on the system, and activate the system from within its setup or dashboard area. You can add it this code using something like Google Tag Manager or directly to the code. When you add the system, you will definitely want to add it to the marketing site as well as your application so users can see the system while they are logged in.
A trigger in a chatbot is a method of telling the chatbot to do something, generally send a message. Different pages can be given different triggers and different users or groups can be given different triggers based on page, person, activity, lead score, etc. The more precisely you can track your users and trigger questions or responses based on their actions, the better.
Triggers will run different recipes, noted below, which will deliver information and ask for information from users. Different kinds of SaaS products require different triggers for different information, so these will be set up in different ways based on your system. However, almost all systems will want to have a sales trigger on the homepage that runs a recipe asking users if they have questions about the system.
Recipes are a set of questions and answers for users. These can be really simple like “What is your name?” Or, they can get really complicated and help users troubleshoot issues in the system. Most chatbot systems have preloaded recipes in place that just need to be edited to fit your product or service.
FAQs and knowledge base are features that work together to enable your users to ask seemingly random questions and have answers ready for them. The knowledge base is a place where articles on these different questions can be stored. With a thorough FAQs and knowledgebase, a user can very often answer all of their own questions.
Tagging enables you to tag users with different attributes. A tag could be the user’s name, where they are in the buying cycle, what kinds of issues they are having and much more. These tags are generally added by either the user answering questions that fill tags or by the system adding tags via actions the user took. These tags are very often the same as are used in lead scoring. In addition to this, if you have a CRM that can be integrated into the system, user information can often be pulled from that system into the chatbot. This way, when a representative jumps onto a chat with a user, they already have that user’s information.
It’s easy to start off with a chatbot tool and want to tell everyone about everything, now that you can. But even a chatbot can be intrusive. Be careful about how you’re using your chatbot and only really throw out information where you know people need to see it or are really struggling.
If a user is logged into your system, even if they are not in the app, you can often still load their information. This way, when the system logs information for that user, no matter where they are in your system, you know who they are, where they are in the buying cycle, how good of a lead they are, and how to contact them. This can make your users’ experiences substantially better and smoother. Just by saying “Hey John, how can we help you today” instead of “Can you please tell me your name?” is a great example of getting started the right way with someone.
Every question a user asks that your team has to answer is time spent that doesn’t need to be spent. So really think through your questions and their answers and get them added to your system. Every question answered correctly and well is money that you don’t have to spend later.
This is one where some people may disagree with me, but I am not a fan of having to pay so much for a separate knowledge base. Besides, if you stop using the chatbot you’re on now, you’ll have to move all that data, and it is a ton of data. So my opinion is to just start off with something that you can use in the long run and link from your chatbot into that system. Besides, there are knowledge base systems that are more robust and extensible than the ones in most chatbot systems.
Knowing who you’re sending what information to is critical. If you know what a certain user type or profile always needs and deliver that information to them, it will make their lives and your life so much easier.