No matter the business case, whether it be customer service, product development ideation, or online sales, businesses have begun to flock towards chatbot technology to gain a competitive advantage over their peers. Understanding the value of a chatbot, and getting started building your own chatbot, isn't as daunting as it sounds. We've put together numerous whitepapers, articles, and videos to give you an easy-to-understand, accessible start to building your own chatbot.Schedule consultation
July 1st, 2019
In the context of the enterprise, Artificial Intelligence (AI) is simply another toolset for driving higher productivity and optimized revenue. Enterprise-grade AI solutions are becoming commonplace boasting simplified implementation and proven ROI. AI can be as subtle as a chatbot supported by big data and natural language processing offering sales or technical help on a website to something as amazing as the AI program that created the entire movie trailer for the film Morgan. Perhaps, I am an AI creating the blog you are reading right now, would you know the difference?
Consider ancient farming that used the earliest of tools requiring hard work and manual labor where farmers could only support a small farm on a few acres. Adding the plow lead to larger fields with lots of acres but now required many field hands to harvest. Now with modern combine harvesters, farmers can manage 1000s of acres by themselves. Think of AI as the combine.
If your team’s time is the limiting factor, and they spend 90 percent of their time doing menial, manual, and repetitive tasks, that only leaves the last 10 percent to be creative or to enhance the process if they are not already too exhausted. What would happen if they had 90 percent of their time to be creative and rather than being tasked with manual processes, were tasked with monthly meaningful improvements on those processess that improve ROI. Suppose they were given an AI toolset to implement those enhanced processes and real-time dashboards to steer and share their results.
“Digital transformation” has been a common corporate call-to-action for over a decade now with varying results. At first, we were supposed to get rid of paper, then integrate distributed technology islands, then we needed to “move to the cloud” and digitize workflows. Together, these common directives and more are referred to as “digital transformation.” The resulting digital cloud created a new monster called “big data” capturing every conceivable data stream related to your customers, their behavior, and the makeup of each transaction including the weather. The size of Big data is expected to explode exponentially with the current call for a move toward the “Internet of Things (IoT). Now the call is to enhance the customer, employee, and management digital experience. This new digital domain with cloud enhanced digitized workflows plus associated big data, and a desire for an optimized digital experience is fertile ground for AI.
With the digital groundwork laid through each enterprise’s digital transformation, AI solutions can be easily implemented to create immediate impact and offer a solid ROI. One of the key aspirations of a digital transformation is to digitize the workflow. Once digitized, the workflow can be measured and optimized, then scaled. This is where AI comes in. Remember the combine harvester analogy? This is where your employee that could only manage an acre or two can now manage 1000s of acres – where that acre can refer to customers or products produced or the number of employees they manage. Even more powerful is that up leveling of employee activity is not just about saving labor costs but improving margins through improved processes or increasing ales through a better customer experience or increasing quality by changing the employee focus from quantity to quality.
AI vs. Automation
While automation and AI are different activities, they are often offered together in a solution to maximize the impact of each. However, to be truly AI there needs to be an intelligent aspect to the solution such as machine learning, natural language processing, and/or big data analytics integration. Therefore, automation may or may not be AI-based, but to be AI-based, the automation needs to be powered by data to mimic or supersede human behavior and intelligence.
Let’s look at a hypothetical example. Most people are familiar with a chatbot. A chatbot might be found on a service portal to offer front line often avoiding the need for live human support by offering simple solutions up front – turn it off then back on again. If a chatbot can solve 50 percent of the customer issues, this allows the existing live support to serve double the tickets, and to fill out the service ticket before being passed off to the service desk for further efficiencies. That is simple automation. These are simple workflows that are predesigned allowing the customer to follow a predefined path of multiple-choice questions and fill in the blank to move them toward a pre-programmed solution response.
Now let’s add AI. Natural language processing allows the chatbot to take questions freeform from the customer allowing them to express the question the way they want – text or voice – and then parse the text to determine the context, tone, and underlying concern when answering the question and offer the appropriate response with the right tone or compassion to mimic a real person. The customer may question if this is a chatbot or a real person at this point and proper etiquette may dictate alerting the customer up front, so they don’t feel fooled. If the chatbot is connected to big data or customer information, the chatbot may see who the customer is, the state of their service, and may even see that there is an issue and reset their service and fix their issue even prior to the customer’s ability to ask their question. The chatbot may even proactively say hi to the customer by name and mention that they noticed their service was derogated and that the bot restarted the service to fix the problem. If an internet provider for example, the bot may even mention that the client is using an older router model and suggest switching it out to avoid the issue in the future. This bot may even notice the customer’s demographics and offer an upsell that would be appropriate for that customer potentially driving more business. With machine learning, the bot will “remember” this conversation and could continue the conversation with the customer on their next encounter saying, “I noticed that you upgraded your router, is it working better for you?” With machine learning, if they are still having issues, the ticket can be resurrected, and the bot can continue its analysis to fix the issue, perhaps pulling from the customer service database to determine the next most likely cause while updating the results of their analysis into the algorithm and customer service database to better serve the next customer too. By adding AI, you may be able to automate the servicing of maybe 70 percent of all issue inquiries, shorten the resolution times dramatically, increase sales, and best of all, improve the customer experience dramatically. This is all hypothetical of course, but you get the point.
For the sake of completeness, let’s define the key AI tools before going into the next section where we discuss use cases.
Key AI Feature List:
Common AI Use Cases
While I will go into greater detail on AI use cases in later blogs, I wanted to offer a good overview of how AI is being used today. The key use cases that we will explore are enhanced customer service, intelligent workflow automation, advanced AI robotics, real-time analytics, and security.
We live in a 24/7 global economy with fierce competition providing lots of options for our customers. To stand out, we need to offer better customer service than our competition and to be supportive of them all along their customer journey. We need to understand their needs and be ready to work with them on their desired medium – phone, internet, texting, social media – and we need to be available when they need us. As expressed in the example above, AI Chatbots offer a 24/7 solution that is available to converse anytime, anywhere that when integrated with big data can offer intelligent solutions that is optimized for that consumer. These customer assistants can source and drive sales funnels, offer optimal pricing and close sales, solve complex technical issues, and more.
When focusing on internal operations, intelligent workflow automation can be applied to a range of procedurals that simplify, streamline, and optimize outcomes – from production lines to project management to self-service HR options. Consider a production line that has 100 people managing quality assurance to oversee 1000 parts per hour that simply recycles bad parts due to lack of time. Suppose an AI automation using optical pattern recognition and specialized optics can be added that can manage 10,000 parts per hour so that those 100 people can be reassigned to work on production instead. Perhaps 1 or 2 of those people can simply fix the broken parts to increase the yield of the manufacturing process bringing costs down. Suppose the AI automation using machine learning which is connected to big data and internal company data can also analyze those broken parts to determine the potential cause such as a bad run of parts, a consistently bad supplier, or maybe a production line section or person that is causing most of the issues. Perhaps a third person could remain in QA to manage production optimization to pull the bad parts before they enter the run, or to deal with the bad supplier, or maybe work with the section manager to determine where the source of the errors is coming from. This AI upgrade theoretically has boosted the QA performance 10x, increased production yield, and offered insight into production issues to proactively optimize the production line.
With AI, robots can be produced that act and behave like people, but more importantly, interact with the physical world to perform highly advanced randomized tasks. An AI assistant can greet and escort guests, retrieve vital physical copy during meetings, deliver mail, or even disarm bombs if set up properly. Robotics has come a long way from just being an arm with fingers in a production line. Robots can go places people cannot, they don’t require rest, they can be produced to have superhuman strength, but without AI their use is limited. With AI a robot can complete an assigned task without requiring a human to complete it. Within the enterprise, robot mail delivery, receptionists, factory workers, and dangerous duty support personnel is likely to become much more common.
AI and machine learning can use real-time analytics to identify risks and opportunities in real-time along with taking corrective action or produce product offers. For example, airlines can monitor industry pricing in real-time to engage in dynamic pricing or banks can detect and correct fraudulent transactions. This can also be applied to data security with early risk identification and early detection to implement operational improvement and corrective action.
AI equipped with big data, a video feed, object pattern recognition and given access to security protocols and can communicate with security personnel can take security operations to the next level. For example, a security AI monitoring an airport in real-time can use facial recognition to identify and track terrorists while supplying TSA with alerts and in-depth information on their computer screens at checkpoints. Security can be alerted to move to that area to support any potential threat. Pattern recognition can detect and identify others that are likely accompanying this suspect and can check to see if there is a likelihood of them carrying a weapon. Emotion detection can be applied to check the state of these individuals to indicate the threat level. Their movement can be tracked from camera to camera to aid capture. Unlike a human security analyst, this AI can scale this activity to track any number of suspects at the same time.
Vertical AI Solutions
Where AI really becomes interesting for the enterprise or public sector is in vertical solutions. When integrated into vertical solutions that are already being used within the industry, AI can really come to life. Looking at the Airline industry, imagine a real-time AI analytics system that optimizes prices dynamically in comparison to the competition, to fill the last seats on a flight, to take advantage of the extra demand on certain dates, and perhaps to load balance connecting flights. Suppose that the AI system also acts as a travel bot further customizing pricing based on the customer history and perhaps offering intelligent upgrade options based on customer preferences. Perhaps the bot also uses demographics as well as data found on their social media site to suggest hotel and vehicle packages and even theater tickets. Maybe the bot can even warn about closures for attractions to assist in planning. That bot can also assure that this person’s points are added to their account and may even offer a free upgrade to frequent travelers. Suppose that bot greets the guest by name and continues the personalized conversation when the traveler self-checks at the gate. Too much…… maybe, maybe not. But this would be a very different upgraded customer experience than the one that I typically receive from the airlines today.
My main point here is that any industry can offer a better end-to-end experience to their customers, employees, and managers by taking advantage of AI. Structured correctly, an AI system can go that extra mile improving the experience for everyone involved. When employees time is freed to focus on higher level tasks, everyone benefits. The point of AI is not to replace the human touch, but to allow the humans that they support to do more, higher level activities, that allow them to succeed with elevated performance. As I mentioned earlier, think of AI as the equivalent of offering the combine to the farmer.
We are just getting started. In future blogs, I will go deeper into each of the topics that I touched on here to further explore applying AI to optimize enterprise results. For our next blog, I want to dive deeper into AI use cases followed by real-life applications along with their results. I hope you enjoyed this topic and continue with our next blog. Feel free to sign up for our free newsletter to ensure that you see each new blog as they are posted. If you have any questions, please post them in the comments below and I will do my best to answer them quickly. Our blog topics will be based partly on commenter feedback, so feel free to suggest topics that you would like to see.
Chatbot technology is still relatively unknown so we provide you with some examples of how your firm can use them to improve your business processes.
Learn best practices
Once you understand where chatbots can be used, it's important that you learn how they should be used. You'll learn best practices in the below content.
Helping to remove the fear of getting started, our content will help you know where to begin and how to start building your own chatbot.
Dissatisfied customers complain that resolving a complaint in the air travel industry takes too long, and often ends in a resolution that leaves them unhappy. While aligning your aviation firm with customer expectations is a difficult notion, one new technology is helping bridge the gap: chatbots. This whitepaper provides insight into how chatbots are helping aviation firms align their communication methods with consumer expectations.Read the Whitepaper
In the quickly growing telecommunication industry, it is clear that Internet and smartphones are feeding quick growth, but the industry is also being forced to recognize the threat of disruption. Chatbot technology has been recognized as one emerging tool that will help counter the disruption and provide a competitive advantage. This whitepaper provides valuable insights into how telecom firms are leveraging chatbots for a greater competitive advantage and more satisfied customers.Read the Whitepaper
Retailers face a number of key pain points that did not exist five or even ten years ago. One of the main challenges is providing customers with polished and immediate communication. Leading retailers have already found success in fulfilling these increased consumer demands with chatbots. This whitepaper showcases how chatbots have proven to be very effective at improving customer engagement and lowering costs for retailers.Read the Whitepaper
<p class="x_MsoNormal">By <a href="https://www.linkedin.com/in/wajihyahyaoui/" target="_blank" rel="noopener noreferrer" data-auth="NotApplicable">Wajih Yahyaoui</a>, <a href="http://www.blackstoneeit.com/" target="_blank" rel="noopener noreferrer" data-auth="NotApplicable">BlackStone eIT</a>, Chief Executive Director</p> <p class="x_MsoNormal"><em>July 1<sup>st</sup>, 2019</em></p> <p class="x_MsoNormal">In the context of the enterprise, Artificial Intelligence (AI) is simply another toolset for driving higher productivity and optimized revenue. Enterprise-grade AI solutions are becoming commonplace boasting simplified implementation and proven ROI.</p>Read the Article
Communication has evolved tremendously over the past 20 years. It goes without saying that the speed of human interaction has increased exponentially. This article provides a primer on chatbot communication.Read the Article
T:+1 (425) 283-3463