"Are chatbots ready?", "Nothing will replace human interaction", "Chatbot is not a solution", "Chatbots are very expensive", "No one will take this technology serious" - how many else "pure facts" you have heard from the horse riders?
This text is dedicated to showcasing not just chatbot’s potential in customer support, but also real-life cases from Botbox.
In the article that says how chatbots are replacing or about to replace the traditional customer support would be fair to show what traditional customer support is about.
Despite the beautiful picture from Unsplash, of course, it is never (well, almost never) includes expensive hardware like Apple machines. Furthermore, most like customer support is being provided from the company that is proving outsourcing services — that, in fact, has nothing bad in this — what is a matter most is that all is about giving the best at the lowest price.
Usually, those cost savings are being paid by the employees themselves by negatively damaging their own happiness for the sake of the employer’s stats and monthly KPIs.
KPIs? Let’s go a bit deeper into that and see what kind of KPIs are out there.
There are several. I’ll list the most important and popular ones.
- Calls received per day
- Tickets resolved (amount of requested obtained by the agent)
- Time spent per call
- Customer’s satisfaction
- Tickets created by the customer
- Tickets resolution rate (as a team)
KPIs are usually no less than 80% and sometimes even about 90–95% as a goal for the employees. Therefore, the goal is not coming from the clouds — it is usually being close to the goal that the service provider promised to deliver to the client.
Now let’s see how many actions do agents need to perform physically in order to obtain a request, register, resolve and etc. While evaluating all of this in mind please pay attention to what we have just talked about — a not always best performed hardware.
While customer support agent obtaining a single request, trying to resolve the issue, there are an endless amount of external technical distractions that are affecting the service and customer’s satisfaction accordingly.
Now also count the scale of requests that are usually known, but always ruin all the KPIs when technical outages are happening. Count how many employees need to be hired (if you don’t lock them to the desk), think of a management system that should be productive, but not cruel.
Believe me, just because we have seen it this way, doesn’t mean it is right.
Traditional customer support is solving yesterday’s problems.
Not even today’s.
How Chatbots perform in Customer Support?
To cut out all the conspiratological thoughts, let’s agree — technology is not replacing humans, it gives humans a better environment to perform and act.
In traditional customer support, the client spends time to:
— write email
— call customer support and wait in the line (usually a lot)
— internal system to approach the customer support team
Meantime, the agent from the customer support spends time for this:
— replying customer within an email
— replying to the call, clarifying the issue/request
— replying in the internal system to the customer
— registering a ticket (customer’s request)
— resolving the ticket (in case of success)
— sending a survey for the customer
Since we talk about chatbots, in most cases chatbot does not do what the customer support agent does and everyone benefits from that. Curious?
Let me spread a few facts.
86% of B2B executives prefer using self-service tools for reordering, rather than talking to a sales representative.
82% of customers say the number one factor to great customer service is having their issues resolved quickly.
60% of customers are willing to pay more for a better customer experience.
This all tells us that traditional customer support is not effective. Specifically in big corporations. Clients are unhappy, employees stressed, hardware expenses continue, and low results.
Chatbots are filling the most crucial part of the customer support — waiting and response. Chatbot makes it immediate. This matters. A lot.
At the beginning of this article, I’ve promised to share a case from Botbox.
We have got a leading insurance company called Pasha Insurance as our client. As a Botbox, we made a customized chatbot that responds to the website users and random visitors.
A tool that could look like “addition”, did an actual irreplaceable job.
In August 2020, chatbot did next:
— talked to 12203 users
— obtained 70 travel insurance requests
— registered almost 400 mobile numbers
— made 70 appointments to the doctor
— and many more…
For all of this chatbot’s average time spent per user was 1 minute.
Of course, there were requests that have been transferred to the live agents, but the matter is not only in bringing solutions but also — proving a valuable & joyful experience for the clients.
It’s clear that without a website chatbot a really few people would hesitate to call & write to the customer support team. Specifically out of business hours.
In our customized report to the client, as Botbox, we try to highlight what is better to improve for the customer, what kind of information is better to add, and what to avoid.
I’m sure that the results are important, but it’s important to make them healthy.
For anyone willing to learn more, please reach us via this email — firstname.lastname@example.org