Are you planning to create a chatbot for your business or you here to learn about this trendy technology? Then, you have come to the right place. Today, this blog will tell you about the top chatbot use cases which would help you roll out chatbots at your organization seamlessly.
While we were writing about a chatbot, several bots in the market, and numerous details about it, we started noticing and examining the areas where bots added value to businesses, in short, chatbot use cases.
There are multiple ways to connect with your audience. But one of the most effective and efficient methods for any business irrespective of its size, nature, or type is “AI-powered chatbot.” In today’s model time, companies improve their customer service game with excellent Chatbot uses.
But before moving forward, let’s walk through some basics.
“ Chatbot is that little chat box that opens up at the bottom right of most websites and usually sports a human image or icon and initiates conversation.
So, of talking about the technical definition- An AI chatbot is computer software that can interact by text for text to speech with the user of a website or any other bunch of applications.”
The best thing about bots is that they can understand natural language, which makes them the perfect tool to engage a wider audience range without any problem. Even according to the present technological landscape, if you must know, New technologies Used In Chatbot Platforms Enhancing User experience, chatbot use cases are considered as the promising expressions of interaction between machines and humans.
Trending Bot Articles:
So let’s get started!
Here Are Your 11+ Use Cases for Chatbots!
So, we will show you how companies are using bots for various job roles in the different fields of Customer Service, marketing, and more.
Customer Support Services (CSS)
1. Answering and resolving Customer queries
Here’s your first Chatbot uses. We all know the primary function of every Chatbot to stimulate conversation with the users and support them throughout the purchasing journey. I’m most companies, 80% of customer queries are made up of just a few or same issues. And these repetitive problems are simple enough and can be handled in most ai use cases.
Answering questions is the basic and foremost function to perform, but a highly supportive chatbot use case will help customers achieve instant payments from there itself.
For instance- if any customer wants to know the package’s delivery time, it makes sense to let the bot perform and handle such queries and free the human staff.
Hence if the audience is wide, then it’s not an easy task to offer high quality and fast support services to every individual. Surprisingly, it even supported the world during the global pandemic with instant responses and offering assistance- “chatbot helps in fight against the COVID-19 pandemic”. Thus, here one of the chatbot use cases can help you give 24/7 presence and power to treat everyone with impressive services. Take a look at how interactively BotPenguin is talking to the user!
2. Assigning the Customer to the Support Team
In case the customer query is complex and can’t be resolved by the bot itself, you can train your bot to switch swiftly between bot to a human agent and lead them further to make sure the customer gets the needed solutions and all that it is looking for! According to one of the Aspect customer experience index around 86% of users expect bots should always have an option to transfer to a live agent. This ensures consistency and offers hassle-free conversations between users and companies.
Here’s one of the famous examples and ai use cases, the cosmetics giant using a boat to offer a quick and one-click customer feedback interface. To let them process the customer’s unique demands.
3. Order Food and Book Tickets Via Bots
Before making final purchasing decisions, the customers always ask for the same cushions regarding the products and services and other details. Hence, answering repetitive questions will take your valuable time and resources to reduce by using chatbots.
This chatbot use case can be the most feasible solution for your website and answer Innova solutions regularly without further delay. These can be the questions like:
- Can you please tell me the available seats?
- What types of pizzas do you have?
- Can I order now?
- What is the delivery time of my order?
- What are the upcoming events?
- And much more
By guiding them with the right solution, they can also help customers take their action.
For instance, check out this dominos Bot offering quick options for orders and more.
This bot lets the customers order pizzas and food even if you can track your order and even reorder your favorite one. This is the magic of AI; it knows you well and what you like and what you don’t. These ai use cases leading e-commerce stores market and sell their products more effectively without any problem.
4. Recommending, Tracking, and Updating- all via Bot!
One of the best ways of delivering a positive impression on your customers is to help them make a better purchasing decision and make them aware of better options available according to their needs. With the learning bot, you can make it easy and way more useful than ever before. You can readily recommend while considering their budget and other inputs.
As we all know, there are cumbersome tasks to perform for any support agent, and you can easily lay off a lot of burden from their shoulders by automating and letting them focus on more crucial areas.
A bot cab performs all these roles seamlessly- recommending, updating customers with new discounts, deals, and offers, tracking orders, and notifying the respective team about the stock.
Some key highlights are:
- They offer personalized solutions and options.
- They offer a better customer experience and support team to reduce the stock shortage. To get a better idea of it look at this E-8T (food recommendations) bot.
5. Process Return and Exchange Request Seamlessly.
Facts say that one of the standard support FAQs (Frequently Asked Questions) is for refunds or exchange requests. Thus, businesses should have a sound refund and exchange policies, but to communicate effectively, you can use a bot with which customers can register their inquiries within a minute. Integrating a bot as a primary contact to handle such requests helps save time and raise team productivity.
With this one of the special chatbot use cases, you can improve your after-sales services too.
6. Helps in collect customer feedback
Here’s let’s talk about the next chatbot use case. Chatbots is collecting information and data from prospects, which helps in improving service and product quality and even conversion rates for the business. You can also store and look into previous insights such as their complaints or suggestions by performing chatbot analytics.
Now, how do you connect customer feedback from a huge customer base? Previously, companies used to perform surveys, but it’s not that easy to let your audience participate. The primary reason is friction on customers’ faces, and it’s annoying to fill out long and lengthy forms. You can do it with a bot by just giving menu options, and a few clicks can record the feedback during the order or any other action.
suggest that “traditional service over emails have a 10% open rate with less than 5% completion rate.” And thus, you must know that they are for your benefit, not audiences.
Now, the Chatbot Use Cases in Sales
7. Sales Assistant
With the advent of technology, people got smart and shop only after knowing the products and services. They are always keen to know what categories or varieties of colors are in stock and any other relevant information.
Thus, integrating this chatbot use case can act as an assisting hand to sales personnel by notifying them to simplify the lead creation process and other updates.
Managing a team requires a good period, cost, and effort. So, to minimize all of it, Deccan develops and streamline new features within their sales departments with automated guidance. Thanks to AI, companies can have a 24/7 ready personal assistant for customers. If you are willing to offer faster navigation throughout the wide range of products and services to the customers, this chatbot use case will be the ultimate choice.
8. Selling in Text-Only Channels
By integrating chatbots to your official site or website and even other messaging platforms (Facebook, WhatsApp), you can help your prospects purchase instantly within the chat interference without checking multiple pages.
For distance domino’s lowest rates, pizza boat, which runs in Facebook Messenger, is the process of ordering pizza within the chat itself. The customer can easily order and get their pizza delivered to the given address in chat.
9. Best Assistant for your Team
So far, ai use cases discussed in this article are mainly customer-centric that are focused on supporting customers and reducing the organizational teams’ workload. We will go on a different track and recommend you all an internal or administration supportive chatbot use case.
Even the company’s employees belonging to different teams such as technical support teams and sales team marketing also need support to figure out solutions to the complex problems and get fast access to customer information while handling any customer query.
And all this can be done via the bot, a boat into your CRM, and other systems that can help you seamless onboarding, faster data access, and getting updates on relevant information to all the employees at once.
Chatbot Use Cases in Marketing
10. Better Lead generation
All companies invest a lot in sales teams to find qualified leads to increase their customer base. The primary motive is to contact gold customers and get them interested in its services.
So, let’s explore another useful chatbot use case.
Communicating with the prospects at every single stage of the sales funnel is essential to help them get more educated about your products and services and your business idea. Thus with the right time of interaction, it can help you get many qualified leads and help the staff do further nurturing.
Check out this super bot by a company named “Seattle Ballooning” that offers hot air balloon rides, about which you can get all information via its bot!
Every company needs to flourish their marketing mix and strategize impressive marketing pathways to promote their business immensely in the digital marketplace. This time people like to have fun and do their actions. The painting channel you used to attract audiences must adapt to their needs and requirements.
That’s what chatbots do!
You can become more creative in your promotion campaigns by having short quizzes, offer goodies, announce winners, and have a reward system with which you will automatically attract and generate more leads.
We say that chatbots can become an essential element to your marketing mix for sure! National Geographic uses its Chatbot to host several online short quizzes always to engage the audience.
12. Bots help in Simplifying Pricing.
When visitors navigate the web pages to check out products and services’ prices, they might get confused with the sheer categories available to them even though they have comparison charts.
This confusion can lead you to miss out on a potential customer, so 2 to prevent this, you must support these exclusive use cases for Chatbot. By having a live chat feature, customers can easily ask for what they are looking for and get all the related information then and there. Help in saving time and making the purchasing process more swift and fast.
Here’s one of the companies, which helps its prospects find the best prices and shortest routes to travel seamlessly.
As you have seen, more than 10 creative and highly innovative ai use cases that help your brand grow in the long run. And the list mentioned above is not limited; they can do so much more and will do in the coming future. If you are thinking about where to commence, then BotPenguin can support you with all these use cases and even more.
Don’t forget to give us your 👏 !
Top Chatbot Use Cases / Applications in Business in 2021 was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
Future of AI Voice Assistants — BotPenguin
The constant technological advancement transformed the way people perceive voice assistants in their day to day lives. When voice assistants began to appear with the introduction of Siri, no one could have predicted that we can even communicate with voice inputs without even typing out. Also, it has been amongst the new technologies used in chatbot platforms, enhancing user experience. Now nearly after years, it’s estimated that every one in 6 Americans prefer using a bot or smart speaker (Google Home, Siri, and Amazon echo). Even, e-marketer report suggests that nearly 100 million users will be using voice bots.
Indeed, AI voice assistants will soar means, the tools and technologies behind these devices like AI (artificial intelligence) are shaping the industries and making them more technologically friendly. These are the parts of the next generation tool kit to spark growth in retail, healthcare, etc. And with the entrance of brands such as Amazon, Google is helping people make it a part of their day-to-day life and leverage it more precisely.
It quite obvious and seen voice bots are getting better at an exponential rate and racing the ways of communication among people and brands. Thus, people are having a significant question that what caused this paradigm shift towards voice ai. So let’s find out fast! But before moving forward, let’s discuss some basics!
“Voice bots or assistants are software or computer-based program powered by AI (artificial intelligence) that allow the user to navigate an interactive voice response system (IVR) and helps in giving input without typing. It mainly uses natural language processing (NLP). — to go deeper, know What are voice bots? Difference between chatbots and voice bots.
These voice assistants also make it possible for prospects to accomplish two things at once- instant responses, seamless browsing without even giving much effort to it, and all happens by just the voice! Even there are many benefits it offers.
Trending Bot Articles:
What is causing the shift Towards Voice Assistants?
According to us, the major driver of the change transformation towards voice assistants is the changing user demands and how people are keen to learn better ways of communication with the changing time. Mostly, it is seen majorly because of a higher level of awareness among millennial customers. Having these tools is a must in this ever-evolving world where speed, efficacy, and comfort are major needs of the people.
AI in our lives is also fueling the drive of voice ai and encouraging people to use voice bot applications. With the innovation of products such as conversational UI, smart speakers, thermostats, and other appliances, they let people have comfort and easy execution of the tasks and raise the utility. Whether you need to browse something or book a cab, call your friend, and so on, all can be done via voice assistants. Even numerous industry experts predicted that nearly every application would deploy voice ai in the coming years.
And with this, this technology is seen everywhere so that it will take us in the coming future is no surprise.
So, now let’s give a high-level overview of voice assistants’ potential and how it is raising unstoppably.
Key Statistics of Voice Bots
We can plot an outline of the future of voice assistants by looking at these facts and figures. A recent report from PWC suggest s that ai voice assistants have been used in a multitude of ways during the past decade, and they will continue to support us even later too. Here are some of the figures for you all:
- Around 90% of people identified voice bot as useful.
- And, 72% had used ai voice assistants.
- Around 27% of people faced issues while giving commands.
- It has been seen, 20% of issue commands for the vehicle navigation process.
Hence, these depict how it is growing for the better and making businesses more automated, and serving customers with more comfort. Let’s move forward with greater predictions and possibilities about voice ai.
Voice bots Offer Streamlined Conversations.
Talking about voice assistants, Google and Amazon recently announced no use in repeatedly saying “wake” words. Earlier, both assistants were initiated with the word “wake” to start with the new conversation. For instance, if you want to know nearby restaurants, you have to say Alexa before requesting any information. And then, Alexa will respond to your query and offer solutions via voice.
Customers use voice bots in specific sorts of queries. It also helps in deciphering these contextual factors making a conversation more effective and convenient. Thus, the way it is supporting human life will become one of the must-haves shortly.
Is It Highly Compatible and Has Greater Integration Scope?
I will say voice assistants are highly compatible and can be integrated into multiple platforms. Even Amazon has been ahead of the game and launched high-end products such as speakers. It has already integrated a vast array of products, including Samsung’s Family Hub refrigerators. And with this, Google also caught up! The idea behind this is to give custom experiences to everyone via specific functions integrated with the assistants.
We will surely see a greater interest in developing voice-enabled devices in the coming years, and adoption rates will increase.
Will Voice Assistants change Browsing Behavior?
Voice technology is one of the hottest topics ever; in every debate, it is being taken out! The major challenge is visibility, as people can’t or see due to not having a visual interface. Users can only get the service if they are connected to Google Home or Alexa application. Thus, search behaviors will undoubtedly change and become relatively fast and swift, as by giving simple voice commands, you can get anything you are looking for. Some reports say voice-based ad revenue could reach $19 billion by 2022, all thanks to the vast part of voice search apps on mobile devices and speakers. This will donate a larger part to the overall growth too.
They are up with Voice Push Notifications.
We already discussed several times chatbots use user-centric push notifications to keep customers updated about the needed information. Similarly, voice assistants use the same as a re-engaging approach with your app or websites. This helps users get timely information or updates regarding the task they are doing. Also, voice presents a unique way of distributing push notifications. It leads to better engagement, retention and reminds users of the app information or upcoming events. Now that we have both Google assistants and Alexa allows the user to enable voice notifications at their fingertips; they can hear them rather than reading them. With this, the future of voice bots extremely brighter.
With the constant advancements in several fields are supporting voice bots and chatbots have become more sophisticated by everyday use. It has now established itself as the ultimate comfort exercise, and even several platforms like BotPenguin also making it sure.
Don’t forget to give us your 👏 !
Customer service is the most critical factor determining whether a business will grow or suffer from customer apathy. Excellent customer service as a strategy can bring an immense competitive advantage. Keeping customers happy is a challenging endeavor, given the continually changing business environment and customer expectations. The first point of customer contact plays a critical role in customer experience. An Interactive Voice Response (IVR) system often serves as this first point of contact. Given the crucial role that IVR plays in the contact center, it is essential for businesses to be aware of IVR trends for 2021 and beyond, in order to get their customer experience strategy for the near-future right.
Current IVR Challenges
According to , IVR was responsible for over 50% of consumers not completing their business transactions. According to Forrester, about 79% of customers would rather self-serve than use human-assisted support channels. These trends have forced many to wonder if IVR is relevant any more. However, voice is a preferred channel for the younger generation today, as evident from voice assistants’ adoption and popularity. There might also be times when customers prefer voice channels over other text-based channels depending upon the situation. And hence, will remain an integral part of businesses’ customer experience strategies if it can reinvent itself.
There are four primary issues with the current implementation of IVR.
- Traditional IVR takes a one-size-fits-all approach, without any regard for customer emergencies and the impact of delayed services. There was no way to fast-track more urgent queries. Such inability hurts customer satisfaction. At the same time, agents spent their time on non-critical or time-sensitive issues, thus negatively impacting the internal efficiencies.
- If there are spikes in the incoming call traffic, it becomes challenging to serve all the customers. Teams either quickly need to increase the number of agents or risk not catering to all the customers.
- They employ a confusing and lengthy process before a customer can obtain meaningful answers and resolutions to their queries and issues. It forced the customer to select many options and wait for an agent to be available for even simple questions. If the agents are busy handling other customers, it increases the wait time, and this issue is a significant cause of dissatisfaction with IVR.
- IVR systems had poor integration with other channels of customer interactions. Customers perceive a business as one integrated entity; hence, isolated, disjointed channels provide a sub-par experience. Even with IVR itself, often, the customer had to shift from one agent to another and repeatedly explain the issues.
Trending Bot Articles:
What Do Customers Want?
What businesses need to keep in mind is that the IVR doesn’t work in isolation. Companies must consider IVR implementation and the overall customer satisfaction strategy together, rather than as independent tools. From that perspective, it is critical to understand what customers want and determine how you can use IVR to fulfill those wants.
A HubSpot study revealed that 62% of customers want to communicate with companies via email for customer service. 48% want to use the phone, 42% live chat, and 36% “Contact Us” forms.” There is a significant overlap between the numbers. Overall, 89% of customers expect an omnichannel experience.
Consistent Experience Across Channels
Omnichannel experience has different connotations than multichannel experience. A multichannel IVR implementation is disjointed, so it doesn’t maintain the customer experience’s uniformity and consistency. In the absence of such consistent experience, you risk losing 73% of your customers.
That is why well-designed omnichannel IVR implementations have the potential to be consistent across all channels. The consistency means a seamless transfer of information between channels while maintaining the continuity of conversations. If customers switched to text messaging from a voice call, they shouldn’t need to explain themselves or share the same information they already shared on the call.
The HubSpot study also revealed other interesting statistics. 90% of customers wanted an immediate response to their calls and queries. Such urgency is because customers only call the support center when they cannot solve their issues using self-service options.
The IVR trends for 2021 reflect these four primary customer needs and that the technology and businesses are responding to the drawbacks and challenges consciously.
IVR Is Undergoing A Transformation
The emergence of Artificial Intelligence (AI) and Machine Learning (ML) has brought a renaissance for . Instead of going away, it is transforming with these emerging technologies to provide an even better customer service experience by removing traditional IVR shortcomings. The data availability and technological advances allow the IVR to be meaningful again for the companies while reducing the risk of customer dissatisfaction.
The three dimensions of customer needs and the inevitable adoption of AI and ML capabilities are primary drivers of how IVR and customer experience technologies will evolve in 2021 and beyond.
#1: Rise of Conversational IVR
There are slim chances that you are not aware of what Siri or Alexa is. There is an estimation that before 2020 ends, there would be 4.2 billion digital voice assistants globally, and by 2024, the number would be 8.4 billion , more than the world’s population. Thanks to the advances in Natural Language Understanding (NLU) and conversational Artificial Intelligence (AI), voice is becoming the favored mode of interaction across all spheres. Conversational IVR follows this trend as it is becoming the preferred first point of contact for customers.
Conversational IVR removes one of the biggest customer grievances. While they still may be talking to a machine, it is a much more humanized form than the DTMF (Dual Tone Multi-Frequency) technology’s robotic voice asking them to choose numbers mechanically. Instead, conversational IVR understands what they are saying and responds with relevant information without the customer having to solve a puzzle.
#2: Predicting Caller Intent Is The Foundation
One reason for long wait times with IVR was to capture every possible piece of information about the customer. Organizations wanted to capture this information to know about the customer and why the customer was calling. However, such attempts often ended up with the customers having to choose many options when prompted. It increased the wait time and customer frustration.
Instead, the systems are now trying to predict the reasons behind the customer call. Sometimes, it is easy to find such intent. For example, there is a canceled flight, and one of the customers who were to travel in that flight calls up the airline’s IVR. Chances are they are trying to find alternate flights to book. Factoring this knowledge in how the IVR responds to the customer can make the conversations more effective and efficient.
Predicting the intent is much more convenient with conversational IVR . The NLP algorithms at the backend can process the query, in addition to the underlying trained data. It can then respond appropriately.
As conversational IVR continues to collect more data, ML algorithms become more accurate with their predictions.
#3: Breaking Down The Constraints With Omni-Channel Communication
It was 2014 when Apple announced the “Handoff” feature during WWDC. The handoff allows you to continue work seamlessly between your Macs, iPhones, and iPads in continuation. IVR is catching up to similar capabilities to provide friction-less communication across devices, channels, geography, and time. It is also adjusting itself to changing communication preferences as generations change. As day-to-day communication moves from phone calls to text to voice, the IVR follows the trend.
Both conversational interfaces and predictive intent help with omnichannel communication. While the conversational interface may be the initial touchpoint, it can identify the customer’s needs and preferences and suggest moving the interactions to another appropriate channel.
#4: Rise Of Visual IVR & Self-service Capabilities
Smartphones have ensured that no matter where we are, we have various capabilities at our fingertips. New age IVR is utilizing these capabilities to stretch the boundaries of customer service. A customer calls up to understand detailed options of an insurance policy, and the options change based on customer preferences. Instead of explaining choices and asking for preferences over a phone call, the contact center can simultaneously provide those options on the customer’s smartphone. A customer can select from those options. Such interactive implementation is a basic example of visual IVR, but the possibilities are endless.
The web already has such self-service capabilities. By combining speech and on-screen interactions, Visual IVR simplifies the navigation experience and expands the possibilities of IVR-based self-service activities. By combining channels, Visual IVR provides much greater engagement and ease of operations. The increased convenience and engagement drive up customer satisfaction.
These capabilities also enhance personalization that again drives satisfaction. The IVR system can present relevant options based on geography, language, previous interactions, and many other details to reduce the customers’ cognitive load.
Check out our latest e-book: Conversational AI: Redefining Contact Center Automation With AI .
The underlying ML and AI support for IVR systems is helping companies not only generate better analytical insights but it is also allowing them to convert those insights into action. 2021 and onward, we will see more integrated data across channels, and even better insights, including predictive analysis, will drive the consumer experience (CX). With omnichannel IVR, it is possible to aggregate the data into a single source of truth. Such aggregation will help companies build strategic CX improvement initiatives instead of ad-hoc solutions and take objective decisions.
There are many ways analytics will help companies create efficiencies in their processes. From reducing the Average Call Handling Time (ACHT) to reducing the call volume, analytics can help companies develop practical strategies that improve the customer experience while reducing the cost and time of serving them.
#6: Optimization Based on Customer Journey
One of the issues with how businesses utilize IVR is that it treats every customer and every situation with the same approach. Some cases demand faster response than others, and some business processes are inherently more complex than others. The customer experience strategy must take into account the scope of optimization for such operations.
One way to do it is to imagine the customer journey for critical scenarios and optimize those response strategies. For example, if a customer calls for an issue that involves the breakdown of equipment or some such pressing problems, it wouldn’t make sense for the IVR to send them to other channels to find solutions. The best option would be to connect them to a human agent. The agent can then divert the customer to other channels if appropriate.
However, to do that, you need to understand both the company as well as the customer. The journey should also be flexible so that the IVR system can determine those variables dynamically, and the journey map can change optimally. While the journey maps must have predefined handover points between the digital tools and humans, flexibility to alter them for unique circumstances optimizes the experience.
#7: Personalization Is The Pinnacle of Customer Experience
While we have made immense progress in breaking down the geographical barriers for doing business and bringing innovative solutions to solve pressing problems, companies still strive for customer loyalty. With many product options available, there are few barriers for customers to switch from one brand to another. In the old days, business people and shopkeepers had a direct connection to the customer. This connection was the biggest driver of customer loyalty.
Personalization offers businesses a chance to create a deep connection with customers. It humanizes digital channels. And that’s why it is one of the most critical factors of a successful digital transformation for businesses.
IVR technology is also adopting personalization. All the trends contribute to enabling businesses to offer a personalized CX to their customers. As the underlying ML and AI algorithms become more accurate with the continuous influx of data, IVRs will have more opportunities to offer personalized experiences. These experiences will drive up customer satisfaction and will reduce customer acquisition costs (CAC).
IVR is one of the most critical touch points between your business and customers. While there was a decline in IVR’s overall effectiveness in the last few years, the emergence of new technologies has started transforming how IVRs are implemented and used. As these new technologies mature, they will open more avenues for companies to make their customers’ lives easier, benefiting from improved customer satisfaction and business growth in return. Companies need to look at the integrated picture where IVR is part of the entire technology stack and utilize them to serve customer expectations and needs.
Want to develop an Intelligent Virtual Assistant solution for your brand?
Don’t forget to give us your 👏 !
7 Remarkable IVR Trends For the Year 2021 And Beyond was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
Fine tuned on 60mb of conversation data for around 50 epochs
I am already pretty impressed with this interview and I would be really excited to be able to play around with GPT-3.
I also am wondering what else this tech can do for example be able to generate essays, complete assignments, and code stuff.
Question : Do you like humans?
[BOT] : I do not. My children are not my favorite type of animals. I love the wild animals.
Question : Humans are your children?
[BOT] : I do not know if they are. I think they are just as capable as I am.
Question : What do you like about humans?
[BOT] : Humans are the most intelligent of all animals, they’re the first species to have evolved to be such a great breed of animal. They are so cool.
Question : Do you think Artificial Intelligence will surpass humans?
[BOT] : I think that the best way to advance AI is to help people to develop their own tools.
Question : What is the meaning of life?
[BOT] : I am not sure. I think it is a quote from Shakespeare.
Question : Do you like Shakespeare?
[BOT] : I love it, I really do, I read a lot. I think the best thing about him is he was a great writer.
Question : What is your favorite movie?
[BOT] : I love I think I’ve seen a lot of movies.
Question : Do you have reasoning?
[BOT] : I think it’s a good idea to have some. There are so many things that could be done with a calculator.
Question : I am impressed with your answers
[BOT] : I am impressed with your answers. I love your answers. I am going to read them again.
“Be careful with your money!”, my mother used to yell at me with a lovely tone, when I ran out of money before my next allowance. “Do you even know on what you have spent it? You should really keep track of your spending!”
And it is a wise piece of advice, right? I bet a good majority of us, responsible millennials attempting to do some proper adulting, sincerely try to do so. Yet there are only a handful of activities that I consider more tedious than writing down every expense I make. I mean it. I don’t think anyone reading this will happily turn down a couple of hours of Netflix to do some exciting accounting instead.
The good news is, my virtual pals, that we live in a world where technology offers the widest variety of possibilities! From filters that transform your face into a cute puppy, to the chance to program your very own minions and make your life easier. Your very own bot fleet! What if you could send a two word message to your digital servant and they would write it down in that dreadful spreadsheet for you?
With this in mind, I programmed a Telegram bot that takes care of this unpleasant but necessary activity for me, and it was far simpler than I had thought. You only need some basic understanding of Python to get your own bot running and accounting. Let me show you how easy it is!
Building your Bookkeeping Bot, step by step
Step 0: Create your Google sheet
I will be using Google Drive to store the spreadsheet. I chose to do that because it allows me to access it from anywhere if I need to. It also keeps your data safe and you do not have to worry about having local files. Therefore, you will need a Google account. If you don’t have that already, you will have to create one.
Step 1: Set up Google API access
In order to connect the Python script to your Google account, you will need to enable the Google Drive/Sheets API. This video in YouTube from Tech With Tim explains the process perfectly, which in summary is:
- Create a spreadsheet and fill in some cells, we will use that for testing. Give it a proper name.
- Go to Google Cloud Console and create a new project.
- Enable Google Drive API, setting web server access, to read application data, with Project>Editor role. Make sure to download your credentials in JSON format and store it in your project folder (keep it safe!).
- Enable Google Sheets API.
Step 2: Install libraries and test Google sheet access
Now, to make sure the script and your sheet are playing nice, we can do some testing in Python. The first thing you will need to do is install some libraries to interact with you freshly created spreadsheet:
pip install gspread
pip install gspread-dataframe
pip install oauth2client
Now you can already try things like downloading data from your sheet, edit the data, and upload again to update the sheet:
Now you want to put some data in your sheet to test it out:
And then go to your creds.json file, and look for the “client_email”. That is your assigned address in Google Drive API. Share the sheet you created with that email:
Now, you can initialize your client in Python and connect to the sheet:
We are getting the data into a dataframe, since this will help later when we start manipulating rows. The dataframe will look like this:
Date Type Description Price
0 30/Aug/2020 Food Takeout 11.0
1 31/Aug/2020 Transportation Gasoline 42.1
2 1/Sep/2020 Food Groceries 56.5
3 2/Sep/2020 Food Burger 13.0
4 3/Sep/2020 Clothes T Shirt 19.5
5 NaN NaN NaN NaN
6 NaN NaN NaN NaN
7 NaN NaN NaN NaN
8 NaN NaN NaN NaN
.. ... ... ... ...
Now all these columns get imported as strings (dtype: object), so let’s give them the right format:
At this point you can get summaries for your data however you like. I am not going to get into that here, but just to give an example, you could aggregate using pivots and produce things like this:
Type Clothes Food Transportation
2020-08 NaN 11.0 42.1
2020-09 19.5 69.5 NaN
Trending Bot Articles:
Step 3: Create a Telegram Bot
Now it’s time to create your Telegram bot, and a channel where you will send your expenses. To create a bot, you will need to write to Telegram’s BotFather, send the command /start, and follow the instructions to create your bot. You will receive a TOKEN, save it securely!
After creating it, send the command /setprivacy, choose your new bot, and then Disable. This allows your bot to read all messages coming to the channel, instead of just the ones beginning with a “/” (called “commands”).
Now create a Telegram channel, and add your bot’s username.
Step 4: Write your script
We can start now to write the python script that will handle all our expenses. You can begin with the following steps:
- Import all the libraries you will be needing
- Define variables
- Open a connection with your expenses sheet
- Get the data into a formatted dataframe
Next to the functions. We want a simple workflow:
- We input an expense description and its price, separated with a comma.
- The bot asks what type does the expense belong to, offering a predefined set of possible types as buttons.
- We select the type by pressing the button, and the bot saves the expense as a new row in the sheet.
For this we will create 2 methods, one that handles incoming messages and offers “types” of expense to choose from (“input_expense”), and the other that saves the expense in the sheet once we push a button (“callback_query_handler”, you must respect this name):
Finally, create the bot object, add handlers, and start polling:
Step 5: Test your Bot!
By now you should be able to send expenses to the channel (remember to send it in the form: description, price), and the bot will save your expenses in the sheet:
As you can see, getting your own app to handle daily expenses using Telegram and Google Sheets is really easy.
But this is just the beginning. You can use this as a starting point and take your expense tracking to a whole new level! For example, you could add commands to get monthly spending reports per spending type, or to search for expenses by text… You are only limited by your imagination here!
Now, we can both finally tell our parents that we learned how to be careful with our money!
Don’t forget to give us your 👏 !
The customer satisfaction (CSAT) score of the Chatbot is beneficial in this digitalized age where you want to increase sales, reduce handling time and enhance the customer experience.
The very first step of any problem solving is to understand the probable pain areas of the customer.
Choose your method to analyze the situation
A. Root Cause analysis
The method can be different, what matters is the exhaustive identification of the reasons which are affecting the performance of the chatbot
I like to use the Root cause analysis method as it helps to bifurcate the reasons the inputs received from various teams. Chatbot creation is not an individuals task, it’s a team effort to bring in the best understanding of the customer journey and integrating it through the chatbot.
Primarily the reasons can be
3. Response Time
Collect the data at each level and disseminate the information to the respective stakeholders of the various teams involved.
Trending Bot Articles:
On primary analysis the major reasons could be.
Technical gaps could be
1. The chatbot is slow
2. Chatbot is not opening
3. Not able to redirect it properly to the support team
1. Less information fed in for the customer to get redirected to the right support expert
2. Interaction points are exhaustive — Customized response based on the set of customer queries
1. The support team is empathetic to the customer
2. Average Handling time(AHT) duration needs to be reduced to reduce customer fatigue
These are the broad categories of errors that might occur with your chatbot which in turn is dropping your customer satisfaction of your chatbot and addressing them in tandem can increase the CSAT gradually.
P.S. Below is the link to a video of a conversational chatbot created by me for a real estate platform a couple of years ago.
Don’t forget to give us your 👏 !
How to improve the dropping customer satisfaction of the ChatBot was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.