Building an Appointment Booking Chatbot That Predicts Cancellations

If you run a business that asks clients to make appointments, then sudden cancellations can be a huge problem. Cancellations impact businesses from barbershops to doctor’s offices and law firms, significantly impacting the bottom line across the board. 

If you’ve recently booked an appointment, there’s a good chance you’ve done it through a chatbot. After all, appointment-booking chatbots are easier to build than ever before. However, these basic chatbots do nothing to protect against the risk of booking cancellation.

In this guide, we’ll build an appointment booking chatbot that automatically predicts whether the appointment-maker will cancel by using no-code machine learning.

If you want to build an AI-powered chatbot using traditional means, then you’re faced with options like Google’s DialogFlow or SAP Conversational AI, which are very complex, and take months to figure out.

With Akkio, you can reach a nice middleground, and create AI-powered chatbots with simple bot builders. Read on to learn how to make an AI-powered hotel booking chatbot in Landbot in minutes, which will automatically predict cancellations.

Building and Deploying a Cancellation Prediction Model

First, sign up for Akkio. Then, hit “Create New Flow” to get started making an AI flow.

Then, click “Table,” as we’ll be working with this tabular dataset of hotel bookings, including cancellation data, from Kaggle.

After uploading the Kaggle dataset, you’ll get an overview of the table, as seen below. 

Now that the data is uploaded, hit “Add Step” and then “Predict.”

We then simply select the column we want to predict, which is “is_canceled.” We can also add “reservation_status” under the “Ignore” tab, since this column includes the target values. Then, click “Create Predictive Model” and wait a few seconds, and you’re done! We’ve now built an AI model that accurately predicts hotel cancellations.

Our final step in Akkio is to deploy it, so just hit “Add Step” and then “API.” You’ll then see a screen like the below, and you can simply click “Deploy” to finish. 

Creating a Landbot

Now, let’s sign up for Landbot, which we’ll use for the chatbot functionality.

Then, simply click “start from scratch” to get started.

We want to create a Landbot that customers can use to book a hotel, giving the same information used in our predictive model. That data includes: 

• arrival_date_year, arrival_date_month, arrival_date_week_number, and arrival_date_day_of_month (which can be extracted from one date variable)
• stays_in_week_nights (which can be calculated from the dates of stay)
• meal (the type of meal booked)
• country (of origin, which could automatically be extracted from the user’s IP location)
• market_segment (automatically will be “Online TA” for our online travel agent)
• distribution_channel (automatically will be “TA” for our travel agent)
• reserved_room_type
• customer_type (will be set to “transient”)
• adr (average daily rate - not a question for the user)
• total_of_special_requests (calculated)

In short, this is all we need from the user:

• Date of stay
• Meal request
• Room type
• Any special requests

Here’s what the Landbot looks like with those questions. I simply replaced the existing blocks with questions asking that information. As you can see, it still all fits on one screen! Make sure to save each input as a variable in Landbot, so we can pass these variables to the Akkio model.

Towards the end of the Landbot flow, you can see a “webhook,” which is where the user’s inputted data gets sent to Akkio to make a prediction. We can then connect to Google Sheets, or wherever our customer data is held, to add our booking cancellation prediction.

The Cost of No-Shows

Any industry loses money when customers don't show up for appointments, but in healthcare, the cost is particularly high. The industry loses more than $150 billion a year to no-shows alone.

That's a lot of money, but it's also a lot of missed opportunity. When patients don't show up for appointments, clinics and systems are losing over $150,000 per provider, per year.

There are a number of reasons why patients might not show up for their appointment. Maybe they're forgetful, or they had a last-minute conflict. Maybe they don't have transportation or child care. Whatever the reason, it's important to remember that every no-show has a real cost – not just to the bottom line, but to the patient's health as well.

There are a few things that clinics can do to reduce the number of no-shows. One is to send reminders, either by text or email. Another is to offer incentives for showing up, like discount codes for future appointments. Beyond the healthcare industry, no-shows are a problem for businesses in a range of industries – from hair salons to hotels.

In the post-pandemic world, it's more important than ever to find ways to reduce the number of no-shows. With so many businesses struggling, we can't afford to lose customers – or the revenue they bring in.

The Problem With Missed Appointment Policies

Google "missed appointment policy" and you'll find a slew of businesses imposing cancellation fees to prevent no-shows. But while this may seem like an effective way to improve attendance, it's actually not the best solution.

The problem with these policies is that they don't address the root cause of the problem: why people are skipping out on their appointments in the first place. Is it because they're forgetful? Is there a conflict with another engagement? Or are they just plain flaky?

When you don't know the reason behind someone's no-show, it's hard to create an effective policy that will prevent it from happening again. And even if you do manage to deter some people with late fees, you're likely also driving away others who simply can't afford it.

A better solution is to proactively predict who is likely to miss their appointment and reach out to them ahead of time.

Not only is this a more effective way to reduce no-shows, it's also more compassionate. After all, sometimes people miss appointments for reasons beyond their control. If you can offer them a little understanding and help them get back on track, you'll build a stronger relationship - and a more loyal customer base.

‍

The Value of No-Code AI

In a matter of minutes, you can build an AI-powered, automated booking chatbot that predicts cancellations. This is revolutionary, because it’s an example of something much bigger: You can build and deploy AI models to predict practically anything in a number of minutes.

In late 2020, an AI startup called ClosedLoop.ai raised $11 million for its AI-based healthcare cancellation prediction software, for a total of $15 million. Clearly, this is big business, though cancellation prediction still has yet to become mainstream.

In short, AI-based cancellation-prediction is incredibly valuable, and it used to be incredibly difficult. Today, you can use no-code AI to accomplish wildly difficult tasks in a fraction of the time.

Beyond Hotel Bookings

In this guide, we’ve explored using Akkio, a no-code AI tool, and Landbot, a no-code chatbot builder, to build an automated booking bot that predicts cancellations.

However, the same principles and steps apply to a wide-range of use-cases. If you visit Kaggle Datasets, you’ll find thousands of ways to apply no-code AI. For instance, you could use AI to predict corporate bankruptcy, credit card fraud, Kickstarter project success, customer churn, or any of a million other metrics. As long as you have sufficient, high-quality historical data, you can make a predictive model.

Summary

This guide can be used by anyone that manages appointments or bookings, whether it’s a doctor’s office, a barbershop, a hotel, even or an airline.

That said, this guide can be so much more. Consultants and agencies can use this guide to increase their repertoire of services, and offer AI-powered chatbots to their clients.

Entrepreneurs can use this guide to build an AI startup faster and cheaper than ever before, and intrapreneurs can use this guide to implement AI in their organizations, even without having any technical expertise.