Your cart is currently empty!
Category: Chat
-
Conversational AI for Opticians
Real Test of Patience with Optician Practices
Imagine a patient is driving back home from the office, or he is in an adventure in the park with the kids and reminded of booking an appointment at the optician practice. But as he is away from a computer, the only option is to call, even though there could be other alternatives for booking an appointment with that practice. If lucky enough to call in the appointment-taking hours, after first or more attempts, a receptionist picks up the call. With her gentle voice she asks “Can I keep you on hold for 1 mins”? 3 minutes. 5 minutes. Ten minutes. Finally they hear a voice “Sorry! For keeping you on hold for this long”. The patient is then asked about all the details and hopefully booked with an appointment.
Result : Frustration. And he will probably have second thoughts booking with that clinic again.
The hunt for cognitive strategies in health care is on. In this modern age, you may have thought that emails or even social media posts have surpassed the phone, but in fact, approximately 80% of inquiries received by your business are more likely to be in the form of phone calls.
Every time the phone rings at your optician practice, you have a new opportunity to provide excellent customer service to your caller. Any call that comes into your private practice is a chance to increase sales by assisting a prospective or current customer. With too much riding on the ability to return phone calls on schedule, failure to do so is not a choice that you should seriously consider.
Regardless of how well equipped an optician practice may be, it is likely they are experience challenges in their business. Let’s examine the hurdles faced by them and the best ways to overcome them.
Handling Call Volume
On average, how many calls would you bet an optician practice receptionist handle a day? Between patient calls, new patients reaching out with queries, referrals from hospitals and eye clinics, test results from outside labs, and more. It is likely that your receptionist must be attending upwards of maybe 150 incoming calls a day. Often, your receptionist also handles outbound calls for consultations, follow ups, availability of lenses, and more. This means the receptionist stays on the phone for a good majority of the day as well as handling patients and medical vendors that enter through your doors. Your receptionist stays busy handling patient needs! Sometimes they have to stretch their working hours to more than 8 hours a day.
While they are answering the phone, it is important that they adhere to phone etiquettes. But the question is how far a person can maintain it after getting overwhelmed with that many calls in a day? The answer is “ Impossible” . Maintaining excellent phone etiquette helps to improve patient satisfaction and will help move the call along in a positive manner.
The diagram below clearly states that long call wait times and poor customer service goes hand in hand (source: YouGov report).
A waiting time longer than 10 minutes is considered to be poor customer service Fielding Feedbacks
At the end of the day, your front desk receptionist takes the stress of the feedback from patients. These feedbacks would be usually negative ranging from long hold times, to waiting for the practitioner to start the test, to the quality of glasses or contact lenses received or the care shown during an appointment. For a receptionist these negative feedback can be a drain and for your business it’s a risk. But if you see many of these circumstances are not under your control. It is still beyond important for you to have empathy for the patient’s situation and to determine what can be the best way to address those remarks.
Missed Calls and Lost Revenue
If your practice is missing calls, you’re losing out on important opportunities. People still make phone calls if it is an emergency or if they are not around a computer. And if you are missing such calls, it delays patients who need care and prevents your practice from growing and remaining financially viable. Then, when a patient’s call is missed, one of three things happens for them(ordered from best to worse)
- They get their query answered from a recorded message (e.g., your clinic hours)
- Drop a voicemail
- They hang up being frustrated and angry
Of these 3 things, #3 is clearly the worst. If a prospective new patient calls your practice but then gives up, you’re probably not going to hear from them again.
Trending Bot Articles:
3. Concierge Bot: Handle Multiple Chatbots from One Chat Screen
- According to a research, 74% of people are likely or very likely to choose another business after a negative call experience.
Of course, you or your staff does not intend to miss calls; it only happens because your practice is running hectic during the workday and then prohibitively expensive to staff at night or on weekends.
Costs that Come with Staffing and Adopting Technology
Sometimes a person who calls your practice and does not reach a human. This is unavoidable unless you staff your phone lines 24/7 (expensive and inconvenient), pay for a staff at a triage line (burns a hole in your pocket), or hire an answering service machine something like a voicemail.
For optician practices that are inundated with calls and other challenges, here is a solution that could help their receptionist decrease time spent on the phone, reduce call waiting time and help maintain efficiency, productivity, and healthy patient relationships.
Optician practitioner helping customers Enter Conversational AI and improved Customer Service
Spike in call volumes require optician practitioners to take proactive approaches to identify solutions that can help patients with self-service. The obvious solution that comes into the picture is Conversational AI solutions which can include things like virtual assistants/chatbots, machine learning, and voice analytics. In today’s time Conversational AI can be deployed and adopted with speed and scale to help increase call volumes and transform overall patient experience without going heavy in pockets.
According to BID Research and ReportLinker, the global healthcare robotics market is predicted to reach $11.4 billion by 2023 and the market for artificial intelligence in healthcare has the potential to reach $36.1 billion by 2025.
So how Conversational AI will rescue your Optician Practice from such a crisis?
Better Patient Experience at a lower cost
Staffing a 24x7x365 your optician practice with human agents can be costly or impossible. When you add the resources required to support multiple lines of tasks and languages, it increases the headcount and skill sets required to serve patients around the clock. But with Conversational AI-powered receptionists you will reduce the number of front-desk staff required on your practice due to the repetitive nature of routine calls. All of these lead to significant cost savings.
Personalization
Your optician practice can collect customer data to deliver an unrivaled level of personalization to the patients. This can range from simple lens or glasses recommendations based on past purchases, to eye tests in real-time.
Personalization can greatly improve the patient experience, promote satisfaction, improve conversion to new patients, and drive business growth. Conversational AI can also help your staff at practice to have context to understand who they are talking to. Understanding their past consultation history, their purchase history , and what their experiences have been so far with your practice you will be well equipped to provide the best possible patient experience, tailored to your patient’ needs.
Self-Service Options
AI driven phone assistant are 3 times more efficient than a human assistant The majority of Millennials who call optician practices avoid situations, which typically require a human interaction and long wait times and prefer to have self-service options instead. These tech-savvy Millennials prefer to book their own appointments, solve their own queries, either on a website, AI driven chat or AI driven phone call. Deploying Conversational AI solutions into your phone service can help these patients stay away from call waiting frustration.
No missed calls and more conversions
In addition to helping you elevate experience for current patients, Conversational AI solutions can act as an asset to your optician practice front desk and help with capturing new patients and miss no calls. With Conversational AI powered virtual receptionists at your front desk, your practice can respond to patients so quickly that they will continue to schedule an appointment with you and visit you as a potential customer.
Conversational AI will help you reach more leads quickly and efficiently. Some practices already use solutions for live chat on their website or on their Facebook business page to easily interact with patients and new leads — these AI platforms simply just allow you to take that patient engagement offering to the next step.
Introducing Swifter AI
As a Conversational AI platform , Swifter.AI has foresaw the inevitable integration of conversational AI and Big Data into healthcare and optician practices and offers conversational AI-powered virtual receptionists that would decrease call wait times, reduce costs and enhance patients experience.
If you are looking for a Conversational AI solution that could transform your optician practice in terms of patients experience and engagement? We can help. Contact us today!
Request a demo to explore how Swifter AI can help transform your healthcare operations and enhance patient engagement with Conversational AI.
Our 24/7 AI assistant is available over web chat, on Facebook Messenger™ or over the phone +44 (0) 1256588165.
Don’t forget to give us your 👏 !
Conversational AI for Opticians was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
-
Adaptive chatbot dialogs
Using machine learning
Building a chatbot initially seems quite simple. The internet is full of examples, and it’s easy to put something together. In the real (i.e. production) world, life is more complicated.
Computer programmers make assumptions about user behavior. We establish the invariants of the system and enforce them through UIs, validation rules and error handling. Life is predictable. Unfortunately chatbots, like all bots, should behave like humans, not computer systems.
We expect chatbots to handle whatever is thrown at them. In this post I’m going to cover the things we need to think about when building a production grade chatbot. In particular, I’m going to explain why we need to build adaptive dialogs.
The concepts in this post apply to all major chatbot engines (Rasa, Google Dialogflow, Amazon Lex etc).
Intents and dialog flows
Most chatbots rely on pre-scripted dialog flows, built to meet a specific goal. Let’s take a simple example:
Bot: how can I help?
User: I need a duplicate bank statement
Bot: ok which year?
User: 2020
Bot: which month?
User: mayThis simple example can be served using the classic intent to dialog mapping. In this scenario, the initial phrase “I need a bank statement” triggers a pre scripted dialog. The dialog has two turns/slots, prompting for the month and year.
Dynamic slot filling
Even this example is not quite as simple as it seems. What happens if the conversation goes like this:
Bot: how can I help?
User: I need a duplicate bank statement for may 2020
Bot: ok which year?
User: i just told youIf we now ask the user which month and year we look stupid. The dialog flow needs to adapt. The most common solution to this is to use a combination of Named Entity Recognition and slot filling. In this case we need to fill two “slots” — month and year. First we map the slots to prompts:
year ➞ what year?
month ➞ which month?Next we perform Named Entity Recognition (NER) on the user’s text, trying to fill both slots. In the example above, “bank statement” triggers the dialog and NER fills the year and month slots. “what year?” and “which month?” will not be asked.
Trending Bot Articles:
3. Concierge Bot: Handle Multiple Chatbots from One Chat Screen
Part of Speech Tagging & Dependency Parsing
So far so good. What happens if the conversation goes like this:
Bot: how can I help?
User: I need a duplicate bank for the last month
Bot: ok which year?
User: this year of courseThis becomes more difficult. Named Entity Recognition is unlikely to work for “the last month”. We could explicitly train the NER model for this scenario, but it’s likely to be brittle. We need another approach. One approach is to first attempt NER and if this fails, fall back to Part of Speech tagging and Dependency Parsing.
Part of Speech (POS) tagging tells us that “bank”, “statement” and “month” are nouns. Dependency parsing tells us that”bank” and “statement” form a compound noun; “last” qualifies “month” and “bank statement” is the object of preposition to “month”. If we put all this together we can understand the users intent:
Part of Speech tagging and Dependency Parsing is much more complex than simple NER, but it’s also a lot more flexible.
Capturing the right data
Let’s take another example, this time retail. Imagine we run a clothing store, and we want to recommend products to our customers. First we ask the user some questions to understand their wants. For this example we will assume we need to fill these “slots”:
- product
- style
- brand
- fabric
- colour
- size
- price
We can use a combination of Named Entity Recognition, Part of Speech tagging and Dependency Parsing to fill these slots. Hopefully system is smart enough to adapt, filing more than one slot at a time.
“I want a black dress in size 8”
In the above example we don’t need to ask for the product, colour or size, but we still need to ask 4 additional questions.
The risk of over qualifying the need
Do we really need to capture 7 pieces of information before displaying some results? We run two risks:
- We may force the user to be so specific that we can’t actually find any matches in our database.
- The dialog may be so long that the user gets bored and gives up.
The risk of under qualifying the need
Maybe we decide to focus on the attributes/slots that we absolutely need. Perhaps product type, price and size. After all, there’s not much point offering someone something they can’t afford or won’t fit.
However, we have to ask ourselves what value are we adding? A simple faceted search on the e-commerce site would achieve the same results. We also run the risk of under qualifying the need. For example, if the user asks for a dress costing less than £500, we may find hundreds of matches.
Striking the balance
Ideally we want to achieve three goals:
- Meet the user’s needs
- Offer real value, beyond that which is achievable through other means (e.g. a website)
- Stimulate and retain the user’s interest (AIDA)
The first and second goals could be achieved using a rules based approach with short-circuiting. We ask the user questions, whilst checking our database for matches. When we are able to offer “enough” results we stop asking questions and move onto displaying the results.
The third goal is not so easy to achieve. We could also use a rules based approach, maybe limiting the attributes to 3 for product one, 4 for product two etc. This is guesswork though. In reality the user’s attention span will be dictated by many factors including:
- the intent / need
- the time of the day
- the device used
- new vs repeat/loyal customer
Using machine learning to drive the dialog flow
Machine learning can help us here. During a “training” period we build the dialogs dynamically, trying different permutations of slots. Like split A/B testing on steroids. We record everything, including the time of day, drop off rate, conversions etc. This behavioural data can be used to build a machine learning model. This could be a simple regression model or something more sophisticated like a decision tree or ensemble model.
When we have a good model, we can plug it into our dialogs. We feed the same variables (product, device, time etc) into the model and ask it to predict which slots should be filled. This can of course be supplemented by a rules based algorithm. The end result is dynamic dialog flow, which is statistically proven to generate the best results.
We used retail e-commerce as an example, but this concept can be applied to any domain. Going back to our original banking example, we could ask the user if they want a paper or electronic statement? do they want a certified copy? etc. A machine learning model could predict which questions to ask to get the best results.
Summary
At a minimum our dialog flow should be smart enough to avoid asking redundant questions. We do this by filling multiple slots from each user response. We only prompt users for unfilled slots. Named Entity Recognition may not be enough. We may also need to also employ Part of Speech and Dependency Parsing for more complex concepts.
Finally, we need to think carefully about the information we capture from our users. If we ask for too little, we may be unable to offer any value. If we ask for too much we may lose their interest or struggle to return a result. We can use machine learning models to build dialog flows dynamically, delivering the best results for each individual user.
Don’t forget to give us your 👏 !
Adaptive chatbot dialogs was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
-
Chatbot best practices
KPIs, NLP training, validation & more
Anyone can build a chatbot. Most chatbot libraries have reasonable documentation, and the ubiquitous “hello world” bot is simple to develop. As with most things though, building an enterprise grade chatbot is far from trivial. In this post I’m going to share with you 10 tips we’ve learned through our own experience.
This is not a post about Google Dialogflow, Rasa or any specific chatbot framework. It’s about the application of technology, the development process and measuring success. As such it’s most suitable for product owners, architects and project managers who are tasked with implementing a chatbot.
1. Know (and measure) your KPIs
I’m stating the obvious here, but it’s really important to know what you want to achieve, and how this can be measured. Many of your KPIs will be sector or domain specific, but I will give you some chatbot specific KPIs to think about. Listed in order of importance:
User feedback — Ask your users if they are satisfied. This can be done during the dialog flow or at the end. Asking for feedback will give you two KPIs:
- Feedback rate — the percentage of users who provided feedback.
- Satisfaction rate — the percentage of those users who were satisfied. I generally recommend offering a binary choice “satisfied” vs “not satisfied” instead of a rating from 1 to 5, but the choice is yours.
Bounce rate — The percentage of users who abandon conversations midway.
Self service rate — The percentage of users who were able to achieve their goal without needing to chat, phone or email a human agent.
Return rate — The percentage of users who return to use the chatbot again.
Hourly usage distribution — When are people using your bot? A bot that serves customers 24/7 is especially valuable as it’s usually prohibitively expensive to provide human cover 24/7.
There are also a couple of technical KPIs to think about:
Model accuracy — If you are using natural language understanding to understand messages you will need to measure the accuracy of your models. This is a complex topic and is best left to data scientists and experienced developers. One word of caution — don’t get too hung up on the technical measures of accuracy. Ultimately it’s the business KPIs that matter. A model that is 98% accurate is no use if users are dissatisfied with the service.
Response times — Speed is not so important for a chatbot. You can work around slow performance by adding a “bot is typing” type message to dialogs. Sometimes we even fake a delay to make the bot seem more human. Nevertheless, you want to keep an eye on performance to maintain acceptable response times.
Finally, it’s important to know which channel your users favour if you deploy an omni-channel chatbot. Being dependent on one third party channel, e.g. Facebook Messenger puts you in a vulnerable position.
2. Use human agents first
Like most technology, a bot is designed to automate tasks that would otherwise be done by a human operator. Before embarking on a chatbot it’s essentials that you know exactly what you are trying to automate. The best way of doing this is to first employ human agents to respond to your users’ messages. Why do this?
To validate assumptions — you (or your product owner) may have established your use cases or user stories, but how valid are they? How many users actually choose live chat to check their order status?
The 80/20 rule — users will make all sorts of requests. Even for a particular use case, some requests will be so esoteric that it makes little sense to automate. For example — you will most likely want to handle postage and delivery type queries, but does your bot really need to handle queries about VAT on BFPO shipments? probably not.
To understand the tone of conversations — How do your users interact over live chat? are they friendly or professional, do they use colloquialisms? Do they use emojis and text speak? Understanding the tone of conversations will allow you to develop a chatbot that your users are comfortable using.
To get training data — This is probably the most important reason why you should use humans first. Even if you’re not planning to use natural language understanding you will still need data for keyword matching. If you are using NLU/NLP you will certainly need good training data. From what we’ve seen, 80% of chatbots fail to meet expectations because they used synthetic training data, or a limited sample of real world data.
You may already employ human agents to serve your customers. If so, you probably need to tweak the data you log, and the way it’s structured (see below). If you don’t yet employ human agents you can actually do this on a (relatively) small scale.
You don’t need to serve all your customers manually before switching to a chatbot. What you’re after is a representative sample. For example, you may display a “live chat now” button for one in 10 visitors.
Trending Bot Articles:
3. Concierge Bot: Handle Multiple Chatbots from One Chat Screen
3. Log (almost) everything
Ok, you need to be mindful of GDPR, so you can’t log everything. For your purposes we don’t actually need personally identifiable information. What you’re after is the phrases users use. In particular, you’re interested in:
- intents — “i want to check my order status”. You can map intents to use cases/user stories.
- entities — “yes, a jacket”. Entities (relevant nouns) form the basis of named entity recognition
- parts of speech — “I want a black or white dress”. The adjectives, prepositions and conjunctions. You will use these to train your part of speech tagging models.
- sentiment — “I’m not happy” or “thanks for your help”. You can use text classification and sentimental analysis to detect when users are satisfied or dissatisfied.
Ideally you will log conversations in a freeform database, something like elasticsearch would be great. It’s important to log not only the messages but the wider context. i.e. you want to tie messages together into a conversation threads and identify the participants (user vs agent). Log the conversations during the initial human pilot phase and also during the full implementation. You’ll want to continually evaluate, refine and improve.
You should also log key events such as clicks and conversions and along with metadata such as timestamps, device used, ip address etc. Just careful that you don’t breach GDPR via jigsaw identification
4. Make the most of your training data
If you’ve followed our first piece of advice, you should have some decent training data. Now it’s time to put it to use.
You may want to split the conversations into 3 parts:
- intro — the first couple of messages e.g. “I want to check my order status”. Useful for intent and entity analysis
- body — e.g. “last Monday”. These messages may contain additional entities
- sign-off — last couple of messages e.g. “thanks for your help”. Useful for sentimental analysis
Next, apply clustering on the intro messages to identify common intents and entities. Following the 80/20 rule this will tell you where to focus your efforts.
Finally, use the data to train and test your NLU models or keyword matching algorithms. Take care when splitting the training and test datasets.
As mentioned in the first section, you may also want to analyse the data to understand the tone of the conversations. This will be useful when thinking how to word the questions your bot will ask.
5. Think about validation and error handling
Experienced IT professionals think carefully about validation and error handling when building apps or websites. You can usually rely on the UI to help enforce constraints. For example, by using a dropdown select box with the valid options. The challenge arises when trying to enforce the same constraints in a chatbot.
Quick replies
Some channels offer quick replies— prefilled responses which can act as a replacement for select dropdowns, radio buttons and checkboxes. Quick replies can be used as a means of constraining user behaviour, but should be used with care. Unlike dropdown boxes, the options are typically displayed horizontally or vertically and take up valuable screen real estate, especially on mobile devices. This makes them suitable for responses with only a few options.
In most cases you won’t be able to use quick replies. Even if they are a feasible option, a chatbot with lots of quick replies is nothing more than an app with a poor UI. As the name implies, quick replies should be used to help users respond quickly. They exist to make life easier for users, not developers.
Message validation
Free text entry is at the heart of a chatbot. It’s unconstrained, so good validation and error handling is especially important. Remember — whilst your NLU model may correctly identify an entity, this doesn’t mean your downstream systems can handle it. 100 pounds or last monday are examples of entities that an NER model will probably recognise, but need transforming for downstream consumption.
6. Think how to handle more than one message
Here’s the typical chatbot flow:
- Ask question
- Process reply
The problem arises when your users don’t send a single message in reply to the question but several. Let’s take this simple example of a dialog between a customer and a human agent:
Agent: hello how can i help?
User: hi
User: i want to check my order status
User: order A123Now let’s see how this might look with a naive bot:
Bot: hello how can i help?
User: hi
Bot: Sorry i dont understand
User: i want to check my order status
User: order A123The bot asks the user a question, then tries to infer intent from the reply “hi”. It can’t make sense of this, so generates an error. There are a few workarounds:
Support chit chat
Basically you train the chatbot to recognise “chit chat” type messages, which it can either reply to or simply ignore. Taking the example above, the bot would either ignore the “hi” or reply with “hello”. Either way, it wouldn’t generate an error.
Buffer incoming messages
Buffer all incoming messages. Wait until N seconds have elapsed since the last message. At this point concatenate all the buffered messages together into a single message and process it. Taking the above example it would look like:
Bot: hello how can i help?
User: hi
User: i want to check my order status
User: order A123
(wait N seconds)
Bot: Ok …If the channel allows, you may be able to monitor the “user is typing” notification instead, setting N to a lower value. The downside to this approach is that the user always has to wait N seconds for a response which makes the bot seem unresponsive.
Buffer but short circuit
The same approach as described above, but instead of always waiting N seconds, you try to process the message buffer every time a message is received. If you can process it, you do so immediately, avoiding delay. Going back to the contrived dialog it would look something like:
Bot: hello how can i help?
User: hi
(can’t process — wait)
User: i want to check my order status
(bingo)
Bot: Ok what is your order number?7. Use checkpoints
As well as validating each user response, you will want to set up various “checkpoints”. This means telling the user what the bot has understood and asking them to confirm this. For example saying something like:
“I understand you want to check the status of order number A123”
Of course, you need to think carefully about how you will handle a negative response. Simply repeating the same questions again and running the answers through the same NLU model or algorithm is unlikely to work. Many chatbots ask the user to rephrase their request in the hope that it will work second time around. We think this is a poor strategy — there’s no guarantee it will work, and it’s a poor user experience.
We believe there are two approaches that will yield better results:
Drill down
We call the first strategy the “drill down” approach. Start out by asking users open questions e.g. “how can I help?” or “what are you looking for?”. Run the responses through the NLU models and algorithms and checkpoint the conversation.
If all is ok, great! If not, you move on to ask more specific, closed questions — probably with some guidance. For example “do you have a query about a return?” You will probably use a different set of NLU models or algorithms to parse these closed questions.
You wouldn’t want to start out by asking this sort of question, because closed questions result in a lengthy dialog. It’s much better for a user to say “I want a white dress in size 12” than answering multiple questions about the product, colour and size. The aim here is to gracefully handle the outliers that can’t be served via the happy path.
Bailout
We call the second approach the “bailout”. Put simply if you can’t understand the user’s needs you fall back to human intervention. See below for more details.
8. Augment your chatbot with human agents
What can you do with the outliers? Firstly it’s important the system recognises when it’s failing to meet the user’s expectations. For your users, there’s nothing worse than talking to brick wall. One way of detecting this is to count the number of “sorry I don’t understand” type responses generated for each dialog. As mentioned above, checkpointing is also very important.
You can’t expect your chatbot to be perfect, and it doesn’t have to be. There will be cases where the chatbot doesn’t understand the user due to an imperfect NLU model or algorithm. There will be instances where the bot simply lacks the business logic to fulfil the users request.
Providing a fallback or “bailout” to human agents is a great way of handling these edge cases. You’re not trying to create the perfect chatbot, even if such a thing were possible. You’re aiming to get the best return on your investment. These esoteric edge cases can be handled by a relatively small pool of human agents. What’s more, the conversations between the users and agents should be logged and will feed into your continuous improvement plan.
You don’t necessarily need to offer live chat style support either. It may be enough to ask the user to email your sales or customer service team with their request.
9. Adopt a continuous improvement plan
The reason you’re logging the conversations is to build up training data, allowing you to build accurate models. To borrow a cliché — this is a process not an event. Whilst the data captured during the initial “human” stage gets you started, you need to retrain the models as you collect more data.
You may discover that your users interact quite differently with your bot vs human agents. Decades of Googling have conditioned people into using a terse form of language. Language intended to help the system understand their query. For example a user may tell a human agent “a white or cream cotton shirt” but tell the bot simply “cotton shirt white”.
It’s also important to keep an eye on the KPIs and metrics.
10. Look for opportunities
Chatbots are freeform, users can say whatever they like. This presents challenges but also opportunities. Chatbots are great for market research.
Take for example, an e-commerce site for a clothing merchant. While viewing a dress the user can choose the colour: red or white. How can the user give feedback that they would like the same dress in black? They could fill out a feedback form or send an email, but they’re unlikely to do this. More likely they will look for another dress or choose another merchant.
In contrast, an e-commerce bot could ask “what colour?” to which the user will reply “black”. The bot would tell the user that the dress is only available in red and white. However, it can suggest a similar dress that’s available in black. Crucially the bot has captured the demand for a black version of the dress. If enough users ask for black, the buyers may decide its worth offering it next season.
Summary
Implementing an enterprise grade chatbot requires careful planning. It’s important to understand the KPIs and business drivers before embarking on the project. Having a means of measuring success is also really important.
Getting suitable training data is essential and one of the best ways of doing this is to use human agents first. Careful logging and monitoring will allow you to improve the accuracy of your chatbot over time. As with all software applications, validation and error handling is very important. Chatbots have the potential to misunderstand users, so checkpointing is a useful double check.
Be prepared to adapt and evolve quickly, especially during the early days. Use A/B testing to find the dialog flows that work best. Retrain the NLU models as you collect more training data. Look for opportunities — are users asking for use cases you’ve missed? Is there demand for a product or variant you’re not yet selling.
Don’t forget to give us your 👏 !
Chatbot best practices was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
-
The Impact of Conversational AI: 2021 and Forward
Automation has consistently been one of the fastest-growing fields in the past decade and also one of the most influential trends. It has been continuously expanding into increasingly complex areas of business operations such as finance and compliance. In the past few years, automation has also become a part of customer relations and management with the help of a technology called conversational AI — the latter proving its importance during the pandemic.
In 2020, with companies being short-staffed globally, automation helped bring productivity and operations closer to 2019 levels but it wasn’t just process automation on its own, conversational AI played a key role as well, both internally and externally. With that in mind, let’s take a closer look at conversational AI’s impact last year and its influence going forward.
What is Conversational AI?
In order to fully understand conversational AI, there are three broad concepts that we must know as well: artificial intelligence, automation, and computer speech. Speech is arguably the easiest concept to grasp — it’s teaching computers how we speak. Computers only speak in zeros and ones, we don’t. Developers and scientists have been trying to bridge this gap by making computer speech more natural with the most progress being made recently with automation and artificial intelligence becoming a part of conversational tech.
Each of these technologies add more capabilities to conversational tech. For instance, AI enables the computer to process exponentially more data faster and significantly refine its speech. Furthermore, with machine learning, computers can “evolve” and learn by making mistakes, although it’s too complex to explain in this article.
Likewise, automation allows the conversational tech to be integrated into complex workflows and be very useful in various areas including customer service, internal messaging systems, business intelligence (BI), analytics, etc.
Examples of Modern Implementations of Conversational AI In essence, it’s a technology that gives computers the ability to not only comprehend natural speech but also derive commands and take appropriate action without any direct input from the user.
Conversational AI Against Global Challenges
One of the most dominant themes of this global health crisis has been limiting human contact. In an ideal world, this would mean that businesses would pivot to operating by putting employees in contact with consumers. However, in the real world, this wasn’t possible due to the heavy reliance on the human workforce.
However, some businesses used this challenge to put conversational AI to its biggest test — replacing human capital, albeit temporarily. From mobile AI assistants that allowed customers to check store inventory without visiting the store to health chatbots that enabled users to self-diagnose and take proactive health measures, a lot of chatbots were deployed last year to continue serving customers without exposing anyone to risk.
At the same time, the extended lockdowns and travel restrictions meant consumers spent over 50% more time on messaging services such as Facebook Messenger and WhatsApp. This also became an opportunity to put conversational AI through its paces. Businesses built applications for messaging platforms and social media platforms to bring important services closer to their fingertips. From placing grocery orders on Facebook Messenger to browsing shopping catalogs on Instagram.
Trending Bot Articles:
3. Concierge Bot: Handle Multiple Chatbots from One Chat Screen
Examples of Modern Implementations of Conversational AI during COVID-19
As you may imagine, the ability to not only parse documents and messages but to also take autonomous action has tremendous value to any business — it’s taking automation to a whole new level. On top of this, because the technology is rather open-ended and the implementation-dependent on the business itself, possibilities are virtually limitless. That said, the use of conversational tech is, for now, most concentrated in a few key business areas including customer service and engagement, business intelligence, and sales.
What is Conversational AI and How Does it Work? We’ll take a look closer look at examples of conversational AI in these areas but before, let’s answer the important question of how conversational AI is actually implemented. Since this technology is most useful when users are able to “talk” to it directly, one of the most popular implementations of conversational AI is a chatbot. A chatbot is an umbrella term covering different types of bots but the ones we’re interested in are usually referred to as AI chatbots. Chatbots aren’t new and have existed for well over a decade but with new technologies like natural-language-processing (NLP), developers are able to infuse artificial intelligence to make chatbots far more capable. At the same time, almost all major social media and messaging platforms have chatbot support.
Let’s now take a closer look at some of the most popular examples of conversational AI implementations:
Customer Engagement Bots
Today, you can find more than a handful of companies selling the same product/service at the same price. With so little product differentiation, customers have begun basing their buying decision on customer service. In fact, according to Microsoft, customer service expectations for more than half (54%) of consumers have increased globally.
Customer service/engagement bots are thus built with one purpose — to open up a two-way communication channel that offers consumers a unique and valuable shopping experience. Customer service bots are most commonly known for providing business/product-related information in a question-answer format but there have been some very creative implementations of customer bots as well.
Take, Nike’s Stylebot for example. It’s an AI chatbot that helps customers find, personalize, and even create their own shoe designs. Although customers could buy shoes using the bot as well, its main purpose was to engage with the customers and it did that very well — achieving a click-through rate (CTR) 12.5 times more than Nike’s average CTR and also 4 times the conversions.
Business Intelligence Bots
Business intelligence is a field bringing together analytics, big data, and data visualization to help decision-makers gain visibility inside and outside their company. Most companies deploy dozens of BI tools ranging from relatively simple (GSuite) to custom-built platforms. However, most of these BI tools lack integration which means deriving insights from the data in different tools is still time-consuming and manual.
Conversational Business Intelligence BI bots use conversational tech and artificial intelligence to trigger complex automated workflows and interpret commands from normal speech and text. For instance, with just a simple text-based message, BI bots can autonomously:
- track and send live updates about lockdowns, COVID-19 cases, supply chain disruptions, etc.
- scan large data sets and find specific documents
- share documents securely
- visualize raw data into tables and charts
- gain crucial insights into customer behavior and buying trends
Sales Bots
AI chatbots for sales is one of the most valuable implementations of conversational AI as it has a direct impact on a company’s bottom line. The purpose of sales bots is to make online shopping more convenient and engaging for the customer. Since an increasing number of customers spent more time on messaging and social media platforms, companies deployed chatbots to bring products where their customers now were.
One of the most popular examples of a feature-rich sales chatbot is SnapTravel. In addition to deploying AI chatbots to all major messaging platforms including Facebook Messenger, WhatsApp, and SMS, the travel company has also developed RCS chatbots, allowing consumers to check hotel reservations and flight status, book tickets, and much more — through the default SMS apps.
Conversational AI Beyond the Pandemic
Implementing any new technology is a long-term investment which begs the question: will conversational AI still be a viable alternative once things go back to “normal”. Consumer behavior trends in 2021 indicate that conversational tech will not only survive after the pandemic but also thrive. Take a look at the following global trends:
- New Generational Behaviorisms
Millennials but especially Generation Z are growing up with smartphones — they’re absolutely used to the convenience of having the world at their fingertips and it’s going to be essential to cater to these habits.
Furthermore, these demographics actually prefer talking to AI chatbots rather than human sales representatives.
Conversational AI and COVID-19 - Productivity and Efficiency
There are hundreds of ways automation improves productivity and efficiency but conversational AI takes this one step further — enabling automation in places where it normally wouldn’t be possible. For instance, L’Oréal used an AI chatbot to free up 200 hours (a whole working month) while shortlisting 80 interns from a talent pool of 12,000.
- Cost Savings
Apart from the indirect cost savings associated with time savings, conversational AI can reduce costs directly by automating a number of tasks in business intelligence (BI), marketing, CRM, and more, thereby reducing dependence on permanent staff. Additionally, conversational AI is more secure, less error-prone, and generally more reliable at handling routine tasks.
- Rise of Online Shopping
Online shopping has been on a steady rise for the past decade and will continue to grow at a fast pace. Furthermore, the pandemic forced a lot of new demographics to shop online for the first time, most of whom will continue shopping online even after the pandemic is over.
- Consumer Engagement
According to KPMG, personalization is the biggest factor in customer loyalty, and we predict that personalization will be one of the most important business trends of this decade as companies adopt more and more technologies that allow them to communicate on a one-to-one basis economically. Conversational AI is one such technology. Furthermore, maintaining a two-way is paramount to increasing brand loyalty and in turn, sales. This isn’t possible with traditional business messaging.
How can a brand benefit from Conversational AI?
The changing consumer preferences and needs are enough to prove that conversational AI is here to stay but it doesn’t just end there. Conversational AI also ties in extremely well with various dominant trends of this decade, making it a key tool in the modern marketing arsenal. It’s a scalable, reliable, and evolving technology that opens up new possibilities to aid in internal decision-making, increase brand loyalty, reduce costs, and most importantly, offer a personalized shopping experience that helps you stand out from the crowd.
Conversational AI has numerous benefits for businesses in 2021 but the most important benefit is conversational AI’s role in differentiating your product or service from the rest. It helps businesses cater to the need for instant gratification by providing solving a wide variety of customer queries instantly. It also enables the business to improve brand loyalty through a more personalized communication channel without any significant increase in CRM costs.
To get started with Conversational AI, consider contacting the Master of Code. If you’d like to learn more about chatbots and how they can be tailored to your exact business model, schedule a free discovery session today with one of our experts today.
Don’t forget to give us your 👏 !
The Impact of Conversational AI: 2021 and Forward was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
-
OCR: Give Eyes to Your Chatbot
We live in a world where robots are increasingly common. There are chatbots on the company websites and machines that build cars and other equipment by themselves. More and more are the tasks that these agents can perform, and OCR is one of them.
This article tells you what OCR is, its applications, and how your company’s chatbots can use it.
What is OCR?
OCR stands for Optical Character Recognition.
It’s a technology widely used in Artificial Intelligence, Computer Vision, and pattern recognition.
As the name implies, this technology allows a machine to read the text in images and convert this information into code that the system understands.
The text can be printed, handwritten, or typed, and the conversion can be done from scanning or photography.
It can also, for example, read captions superimposed on an image, as is the case with movie subtitles.
The Origins of OCR
According to Schantz (1982), the first forms of OCR were related to telegraphs and reading devices for blind people.
To make such recognition possible, the first versions of this technology had to be trained with images of every character and every existing font.
In 1914, Emanuel Goldberg developed a machine that read characters and converted them into standard telegraphy code.
A few years later, entering the 1930s, Goldberg invented and patented the “Statistical Machine.” This used OCR code to search microfilm files. A while later, IBM bought that same patent.
The Applications of OCR by Sector
Today, OCR systems are already quite effective and capable of easily recognizing any text. That said, their use covers various sectors and purposes.
Banking sector
Reading and Extracting Information of checks — recognizes account number, written amount, and signature. Moreover, it’s used on procedures containing large volumes of documentation.
Insurance Sector
Like in banking, OCR can do information extraction on documents such as accident reports, collecting the cars involved in the claim, the signatures of the responsible parties, etc.
Retail Sector
You no longer have to carry around all the promotional stubs every time you go to the supermarket. Today, most supermarkets already have virtual receipts that can be scanned using OCR to extract the serial number.
Other Applications
Traveling
These days, you hardly need to learn other languages anymore, because technologies take care of that for you. This is the case with automatic translations.
There are more and more applications where you point the camera of your smartphone at a label, for example, and you have all the information translated into your language.
These tools are also useful when traveling, and you want to read the indications on signs, such as geographical names.
Besides, you already see many automated systems at airports, where machines do the ticket verification without the need for human intervention.
Trending Bot Articles:
3. Concierge Bot: Handle Multiple Chatbots from One Chat Screen
OCR Technology in Chatbots
What are chatbots?
Chatbots (conversational robots) are virtual agents with Artificial Intelligence that automate communication processes. Companies widely use these types of bots in their services, such as Customer Support.
These agents need certain technologies, such as Machine Learning and Natural Language Processing, to understand and respond to us.
It may be interesting to consult the articles below to learn more about these technologies:
Furthermore, if you want your chatbot to perform other tasks, such as reading documents, you need to include OCR technology.
The Visor.ai OCR Use Cases
Visor.ai integrates, according to the customers’ needs and requirements, the OCR technology in its solutions.
Heineken Use Case
OCR can be an exciting capability for a chatbot to have, namely in marketing campaigns. Take the example of Heineken.
Heineken launched a campaign where consumers had to send a photo, via chatbot, as proof of purchase of their beers. After the receipt validation, the chatbot would award prizes to users.
You can read more about this case study here.
Email bots Use Case
Another automation solution that Visor.ai offers is email bots.
Email bots are solutions that automate processes that take place, in this case, in the emails that companies receive.
It’s an ideal solution for companies with high volumes of email contacts.
Besides reading the emails, categorizing, and forwarding them to the specialized service departments, these bots also read and check the documentation attached to the email.
That is, imagine you have an insurance company and a customer wants to report a car accident.
To do this, he/she needs to send, for example, the claim registration document and the identification document. However, the customer has only sent the claim document.
In this case, the bot reads the attachment, checks to see if the user properly filled out the documents. And, if the ID document is still missing, asks the user for it.
Only when it has all documents present and well filled out, the email bot pass the case to a human assistant.
Conclusions
OCR technology allows your chatbot or email bot to read and extract information from images/documents that have text.
With this capability, their complexity increases, and they become more efficient as they automate processes that take a lot of time away from their teams.
Whichever Visor.ai solution you’re interested in, talk to us, and we’ll be glad to look at the options that fit you best!
Don’t forget to give us your 👏 !
OCR: Give Eyes to Your Chatbot was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.
