Earlier this year, the chatbots were barely a bright spot on the radar of the tech world. Then, at the F8 Facebook developer conference, the company announced that its Messenger application would soon feature chatbots from brands such as CNN and 1-800-Flowers.
Apparently, overnight chatbots became the next big thing in technology, and the subject of media coverage. And other messaging platforms, such as Kik, Line, and Telegram, are experimenting with chatbots developed by third parties.
Here are seven things IT and other digital marketing professionals need to know about chatbots.
1. Two main types of chatbots
“A chatbot is a service, driven by rules and sometimes artificial intelligence, with which it interacts through a chat interface, according to ChatbotsMagazine.com.
Today, there are two main types of chatbots. First are those designed to serve one or more purposes of business, notes Matt Schlicht, founder, and publisher of ChatbotsMagazine.com. These chatbots are typically found in messaging applications. The chatbot of 1-800-Flowers in Facebook Messenger, for Example, is designed to help consumers choose a flower arrangement and send it to someone.
Likewise, chatbot platforms also exist, and all are virtual assistants, such as Apple’s Siri, Google Assistant, Echo and Alexa from Amazon and Microsoft’s Cortana. These chatbots help users with a variety of information and other needs, instead of helping consumers to interact with specific brands.
2. Chatbots for e-commerce, customer service, travel, and insomnia
The chatbots of messaging applications provide basic communications, e-commerce features, travel planning and much more. The Botlist website offers a useful chatbots directory organized by platform and category.
Chatbots, today, are widely used by brands to run e-commerce, especially among young consumers. The chatbots have the potential to inspire purchases and increase the number of items in people’s shopping carts.
Although chatbots can reduce the burden on the customer service team and sales agents, the larger opportunity returns to personalization and commitment, says Klein. Look at Amazon Echo, for example. Echo gave way to the wave of conversational commerce, making it easier for people to ask for what they need simply.
Other chatbots of featured messaging apps are
- Uber – to request a taxi.
- Burberry – to show your fashion during London Fashion Week.
- Taco Bell and Dominos to order food
- Hi Poncho – a jack forecaster very popular for the messenger.
- HelloVote, which helps people to register to vote
- IcelandAir, which helps airline flights and book them.
3. Chatbots are not new
Consumers have been using chatbots for years. When you call an airline, a bank, health insurer, or other company, for example, is often greeted by an “auto attendant. That assistant, or robot, is aimed at helping to deliver the information that the customer is looking for by asking for some commands.
Such interactions usually do not give rise to great and incredible customer experiences, he notes, because consumers typically navigate through different levels of the phone just to be put on hold – or get them hanged.
In comparison, the interactions with the chatbots of messaging applications are visual and more immediate. When asked something, for example, chatbots can provide information to provide answers, such as videos or photos. Customers do not have to waste time waiting on the phone or worrying about the hangers. The chatbots are able to engage much more and satisfactorily with customers.
4. Increasing messaging applications promotes chatbots
Messaging applications are among the most popular mobile applications today, and that helps to create awareness of the chatbots of this type of applications.
There are more active users per month of messaging applications, than social networking apps. Combined messaging (WhatsApp and Facebook Messenger, WeChat and Viber) have nearly three billion users, compared to 2.5 million in popular social networking applications.
Research shows that users of mobile messaging are loyal, or at least, they are always active. Messaging applications are used, therefore, less than nine times a day, five times the average of any mobile application. One month after installation, messaging applications have almost twice the retention rate of other applications: 68% versus 38%. A year later, users released 62% of downloaded messaging applications at least once.
The use of messaging applications is on the rise. In May 2016, the top 15 application publishers saw a 20 percent drop in downloads as per the data from application monitoring by firm SensorTower.
In a separate article published in September 2016, Recode reported 50% of US smartphone users downloaded “zero applications per month” during the three-month period ending June 2016.
According to NYTimes.com, messaging application robots are becoming more popular in a time when people are getting more and more tired of individual applications. Large brands have been promoting easier access to shopping and customer service through their own applications. Of smartphone, some consumers are tired of having to download a different application for each company.
5. The chatbots are in the fart app
Chatbots are about to become an important tool for long-term user engagement. The chatbots of messaging applications are in their infancy, like the first iPhone apps debuting when App Store from Apple opened in 2008. At the time, the best mobile app was one that made fart sounds. So, we can think of the current chatbots as ones that are in the stage of fart application.
The technologies that drive chatbots, including natural language processing and automatic learning, have not evolved enough to allow a fluid dialogue between humans and robots that do not depend on prescribed conversations.
And, then there’s the recent Microsoft Tay fiasco. Tay was a Twitter chatbot released on March 23, 2016, which represented a pixelated young woman. Biting Twitter users led the robot to make offensive comments, including positive statements about Adolf Hitler. In 16 hours, Microsoft disconnected chatbot, apologized and offered lessons learned from Tay.
6. Chatbots are the future, but they will not replace people
Chatbot is not likely to replace applications, e-mail communications, web pages or humans. However, chatbots will see widespread adoption for specific use cases, such as helping consumers make simple purchases, or quickly finding the information they want. For these situations, chatbots can be a “best delivery service” than other forms of customer interaction.
The robots you see today can be a little bit hit and miss, and above all, launched to join the wave of exaggerated robots. But the concept of individual communications with brands via messaging applications are here to stay.
7. IT needs to get involved with chatbots
When developing chatbots, marketers should work very closely and from the beginning with the IT area for the following reasons.
Security: For maximum security, chatbot communication must be encrypted, and these should only be implemented in encrypted channels.
While this sounds like obvious safeguards, it’s not so simple to implement. An internal robot that runs on the organization’s system can be configured in one private channel and encryption. However, if an organization wants to deploy the chatbot on a public channel, such as in a messaging application, it is at the mercy of the security capabilities of that platform.
There are also other potential security issues for chatbots. As chatbots become better at mimicking humans, the technology will be used by hackers to conduct electronic frauds and other social engineering hacks. For example, a chatbot designed to mimic a customer or vendor could initiate a conversation with a worker through a messaging application. After the relationship has been established, chatbot might convince the employee to click on a malicious link or deliver sensitive information.
Support, budget, and ownership of chatbots: Robots require a lot of testing and support. They let the chatbots developers handle that work, but they do not necessarily have the resources to support the robot after it comes into operation. Do not underestimate the amount of support involved in launching a successful robot.
Besides, since chatbots typically run through the entire customer lifecycle, companies with silos may face internal budget and ownership problems, as the robots traverse these silos.
Robots are often not among the core competencies of IT: The biggest challenge for creating useful chatbots is in the underlying technologies and integration with enterprise systems.
Natural language processing and predictive analytics, which are essential for robots to be useful, are not part of the core competencies of most IT departments. Branded chatbots should also be integrated directly into the relevant registration systems for the company, including CRM and support systems and order processing.
The IT department should take a holistic look at technology architecture so that chatbot can seamlessly integrate with the company’s overall communication architecture, rather than being a separate class.