2 Types of Chatbots
Chatbots are essentially text-based dialogue systems that enable two-way communication. They feature a text input field and a text output field, so users can communicate in natural language with the system behind the chatbot. As a rule, these bots are integrated into the company website or the careers page, where they handle candidate communication.
Essentially, a distinction is made between simple chatbots and those capable of learning.
Simple chatbots are designed to draw on a database of information and prewritten answers when communicating. When a user enters a question in the chat window, it is broken down into its components and analyzed. The system responds to certain keywords and structures with predefined answers. Such bots give site visitors initial orientation by answering simple questions. The use of this technology is therefore quite limited, and no real conversation can develop. If the bot is asked a question it cannot answer, it at least asks for the person’s contact details. A responsible team member can then follow up with the interested person personally. This type of response ensures that a company can handle all inquiries from interested candidates, even if the bot cannot help directly.
The second type of chatbot is a different story. Thanks to artificial intelligence (AI) or so-called learning algorithms, these bots can do far more: For example, they can give intelligent answers to complex questions, assess a candidate’s suitability for the desired position, and, if necessary, proactively present more suitable job offers. To do so, they draw on a wide range of data from application documents, social media activity, and other web interactions. Because chatbots learn with every interaction, more and more areas of application are emerging.
Advantages for Recruiting
Availability
At the end of the day, a chatbot is a machine, and as we all know, machines don’t need sleep – at most a break. But that tends to apply to production machines. A chatbot, by contrast, is a computer-based system or algorithm and therefore needs no breaks at all. It is available to its users 24 hours a day, 7 days a week. Lunch breaks, vacations, and sick days never happen. Top candidates in particular look for jobs outside regular business hours, since they work during the day. This makes chatbots especially well suited for communicating with such busy people – they are always available.
Round-the-clock availability offers candidates significant added value. Recruiters are usually very busy and are not always able to respond right away.
Response Speed
Closely related to the previous point is the fact that a chatbot responds to inquiries immediately and in real time. Candidates, especially passive ones, expect quick and direct answers to their questions so they can decide whether the position or employer is a good fit. The days when applicants waited days, weeks, or even months for a response are over. With short response times, the machine thus ensures an improved candidate experience.
Automated Application Process
The chatbot can also be used in the application process:
- It corresponds with interested candidates and answers their questions (e.g., about the company culture or specific questions about the advertised job).
- If the candidate wants to apply, this is also done via the chatbot. It collects any missing information and documents needed for a complete application.
- At this stage, the bot already performs a pre-selection. Based on the information available, it recognizes whether the profile fits the vacancy or not. Age, gender, background, religion, application photo, etc. play no role. The pre-selection is free from bias. If the profile does not fit, it can suggest alternative positions or add the applicant to the talent pool.
- If the profile matches the job requirements, the chatbot forwards the complete application to the responsible HR professional.
- All applicants are informed via chatbot about the next steps in the application process. This does not necessarily have to happen via the bot on the careers page. The digital assistant can just as well communicate next steps via other messengers, such as WhatsApp. Candidates are also informed via bot once a decision has been made. Of course, the chatbot can also share the contact details of the responsible contact person if the applicant wishes to discuss a topic personally.
- Last but not least, the chatbot can also collect feedback on the application process.
Most internet users are familiar with chat interfaces. This simple, intuitive, and user-friendly way of applying also has a positive effect on the candidate experience, since people already know how to use such interfaces. It also saves recruiters time, as they receive only complete and suitable applications from the digital assistant. Overall, this shortens time-to-hire.
Multitasking
While an HR professional can only handle candidates’ inquiries one after another, a chatbot can process an unlimited number of topics simultaneously – without showing signs of fatigue or making mistakes.Simple tasks in particular can easily be outsourced to the bot, which then completes them (semi-)autonomously. Compared to an HR professional, the digital assistant handles more of these tasks in less time. This makes the chatbot a reliable, high-performing support in the office.
Multilingual Responses
For chatbots, it makes no difference whether the person they are talking to speaks Chinese, Hindi, English, German, or Spanish. Equipped with the right data and algorithms, their ability to respond in multiple languages is another advantage.
Disadvantages
Risks
With a (semi-)autonomous chatbot, questions of risk and responsibility arise, because the behavior of such systems is not always fully predictable. As AI advances, these bots are becoming increasingly autonomous, and the volume of data they process is growing rapidly. Storing and logging ongoing processes and decisions is therefore very difficult. Against this backdrop, it is questionable to what extent chatbot behavior is or will be controllable. The fact is: A chatbot needs a solid data foundation and must be programmed exceptionally well. Only then can potential errors be avoided.
A prominent negative example is the AI tool developed by digital heavyweight Amazon. It was developed in 2014 to simplify the hiring process for new employees. The AI was supposed to analyze the résumés of previous applicants and then search the web for suitable candidates. Unfortunately, the developers discovered that male profiles were being favored. The AI assumed that men should be preferred because the data it relied on suggested as much. It also recommended candidates who were not qualified for the vacancy. As a result, the project was terminated early in 2017.
Who is responsible when a chatbot makes the wrong decision? This is a question to always keep in mind if you want to use these machines. It must also be clear that there is no definitive answer to this question yet, which means a certain baseline risk remains.
Flawless Implementation and Costs
A chatbot offers many advantages, no question. For a bot to contribute to a positive candidate experience, however, it must be implemented flawlessly from a technical standpoint, both in the existing HR system and on the company’s careers page. This includes providing it with relevant data it can access. After all, there is nothing more annoying than a chatbot that doesn’t understand your questions or can’t help you. Such situations must be avoided at all costs. Otherwise, candidates quickly drop off and perceive the candidate experience as negative, which weakens the employer brand.
Flawless programming and implementation of a chatbot requires time and financial resources. Before building a digital assistant, you should therefore make sure the need is truly large enough. Otherwise, the effort is not worth it.
Conclusion
The bottom line: A chatbot can be a real asset for recruiting. It can support recruiters with almost all of their tasks or take over entire work packages completely on its own – responding faster and reliably, around the clock. This gives HR professionals more time for other tasks. Whether and to what extent the bot is truly helpful depends on the data it is fed. The more comprehensive and relevant this data is, the better the quality of communication. If the system is also optimally programmed and implemented, it creates real added value for candidates as well – which in turn has a positive effect on the candidate experience.
A chatbot thus provides support in various areas of HR by taking tasks off recruiters’ hands. In any case, however, a digital chatbot can never fully replace a real HR professional.