July 15, 2019

Want to build a great chatbot? Start with context.

Article

The whole purpose of chatbots is to simplify and automate processes. Is a chatbot truly serving its purpose if it further complicates a process that was previously handled by humans? For example, let’s say you’re trying to order a pizza. If you call in to place your order, there is a good chance that the restaurant uses your phone number to look up your account information such as your address and order history. The same standard should apply to chatbots. Chatbot context is extremely important if you are aiming to build a bot that is as or more efficient than using the phone.When we talk about context in terms of chatbots we’re referring to the bot being able to identify what it already knows and only seeking out the unknown information it needs to provide a proper solution, or in this case to order a pizza. If this pizza-ordering chatbot uses context appropriately, then it shouldn’t ask you for information that it already knows, and it should leverage your past customer history whether it’s chat-related or not. Let’s dive into a side-by-side comparison of what an SMS chatbot with context looks like versus the same chatbot without it.

No Context

[/vc_column_text][divider line_type="No Line" custom_height="10"][image_with_animation image_url="28379" alignment="center" animation="Fade In" box_shadow="none" max_width="100%"][divider line_type="No Line" custom_height="10"][image_with_animation image_url="28380" alignment="center" animation="Fade In" box_shadow="none" max_width="100%"][/vc_column][vc_column centered_text="true" column_padding="no-extra-padding" column_padding_position="all" background_color_opacity="1" background_hover_color_opacity="1" column_shadow="none" width="1/3" tablet_text_alignment="default" phone_text_alignment="default" column_border_width="none" column_border_style="solid"][vc_column_text]

With Context

[/vc_column_text][divider line_type="No Line" custom_height="10"][image_with_animation image_url="28381" alignment="center" animation="Fade In" box_shadow="none" max_width="100%"][divider line_type="No Line" custom_height="10"][image_with_animation image_url="28382" alignment="center" animation="Fade In" box_shadow="none" max_width="100%"][/vc_column][vc_column column_padding="no-extra-padding" column_padding_position="all" background_color_opacity="1" background_hover_color_opacity="1" column_shadow="none" width="1/6" tablet_text_alignment="default" phone_text_alignment="default" column_border_width="none" column_border_style="solid"][/vc_column][/vc_row][vc_row type="in_container" full_screen_row_position="middle" scene_position="center" text_color="dark" text_align="left" overlay_strength="0.3"][vc_column column_padding="no-extra-padding" column_padding_position="all" background_color_opacity="1" background_hover_color_opacity="1" column_shadow="none" width="1/1" tablet_text_alignment="default" phone_text_alignment="default" column_border_width="none" column_border_style="solid"][divider line_type="No Line" custom_height="20"][/vc_column][/vc_row][vc_row type="in_container" full_screen_row_position="middle" scene_position="center" text_color="dark" text_align="left" overlay_strength="0.3"][vc_column column_padding="no-extra-padding" column_padding_position="all" background_color_opacity="1" background_hover_color_opacity="1" column_shadow="none" width="1/1" tablet_text_alignment="default" phone_text_alignment="default" column_border_width="none" column_border_style="solid"][vc_column_text]

Why is chatbot context important?

In the example without context, the user had to respond with 56 words just to get to the point in the conversation where the delivery address was confirmed. In the example with context, the user had to respond with just 11 words to complete the entire order process minus payment because this chatbot used the user’s phone number to look up their account and order history. The chatbot with context did a much better job automating and simplifying the pizza ordering process for the user. It may even have been quicker for the user to place an order via phone rather than using the chatbot without context. The capabilities of context in chatbots are magnified significantly in more advanced scenarios than simply ordering a pizza.

At Boomtown, we’re working to take chatbot context to a whole new level. Our enterprise support platform collects detailed information regarding the tech support services we provide to our partners and local businesses on merchant services technologies such as point-of-sale, credit card processors and other in-store devices. By incorporating this information into our chatbots in the form of context, we can dramatically improve both the time and accuracy of issue resolution. So if you’re looking to build a great chatbot, relevant context is essential and a great place to start.