Human Resources

DRUID Talks Ep #9 Shaping the Work of Tomorrow – A Real Life Example of Conversational AI in HR

DRUID Talks Ep. 9 explores a real-life example of Conversational AI's groundbreaking impact on a prominent bank, its dedicated staff, and valued customers.

Episode #9 of the DRUID Talks webcast features Nevenca Doca, Senior Executive HR Director at Banca Transilvania and Subject Matter Expert Kieran Gilmurray. See the full episode and transcript below.

Subscribe for the next episode!

Kieran Gilmurray: Someone once said that “Banks are places where good people go to die”. But this is not the case with this bank! Banca Transilvania has constantly reimagined from the inside out with over 15 years of tenured staff, still hungry to innovate and change banking as we know it.

“But how is this possible?” I hear you ask.

Well, today's DRUID Talks guest, Nevenca Doca, is a senior executive human resources director at Banca Transilvania, who introduced Conversational AI to impact the bank, its staff and its customers hugely. So it appears there are banks, and then there is Banca Transilvania. Welcome, Nevenca!

Nevenca Doca: Thank you so much, Kieran. This was a lovely introduction.

Kieran Gilmurray: Well, Nevenca, for those people who don’t know you, let's start with the beginning. Who are Banca Transilvania, and what makes this a special place to work?

Nevenca Doca: Well, it all started about 30 years ago in the region of Transylvania, as you can imagine, with some founders just having this crazy initiative to fund the bank, to start the bank. And it grew to being today the largest bank in South Eastern Europe; we are very proud to be able to mention that, actually, the third year in a row. It currently employs the financial group of which Banca Transilvania is the main company, but it also has all sorts of other financial institutions around it. The whole group employs more than 12,000 employees, which you can imagine, it's not easy to, you know, to work with these days. It's nationwide, and we have small investments outside Romania, in Italy and in the Moldovian Republic.

So, it's a complex business, not only because it's in finance and banking and, as you mentioned, it's not as popular these days as it used to be, but also in terms of businesses, legislations, regulations, EU, non-EU, all sorts of complex issues there.

And then, the pandemic kicked, and then, unfortunately, and very sadly, we have, you know, the largest war in our times at the largest border of the country. So, it's a complex situation.

Kieran Gilmurray: It is a complex situation, and it's interesting. So much has happened in 30 years, but I love your statement there. They just decided to start a bank as it was the most natural thing in the world. I haven’t woken up with that thought just yet. Maybe someday I'll start the largest bank in Eastern Europe for fun.

Nevenca Doca: We could have a whole podcast about that. In '93, in Romania, just a few years after, you know, the communism left this country to have this initiative… And, you know, some friends saying, “Let's start our own bank!” That's some courage, some intuition, and some vision.

Kieran Gilmurray:  That is kind of special. Well, Novenca, what’s got you inspired to implement Conversational AI in the bank to cope with all of this change and to promote innovation and better customer service? How did you go about implementing it?

Nevenca Doca: I would make a disclaimer from the very beginning, and this is that I love people. I truly, totally am in love with people, not with robots. Just wanted to make this disclaimer.

However, I find that people can add value more than, you know, moving papers from left and right or, you know, doing some very repetitive, very boring and little value-added work. So, that's the business starting point, in general. And, specifically with this project, I remember I was in the office with my payroll and admin team on a specific payday, and I couldn't talk to them - the phones were ringing all the time. And at some point, I was almost annoyed, and I asked, “What's happening? Why are those phones ringing that much?” And my colleagues were telling me that colleagues were calling to ask about the amount that was credited to their accounts: “I have, you know, X amount less than last month. Why do I have that amount more than last month?” or whatever questions they had. And then I asked them, “And what do you do? How do you reply? How do you answer?” – “Well, we check, we look in some tables in Excel files and other databases, and then we answer.” I said, “I'm sure that there must be another way to do that. Some kind of robot can check into those very similar tables and databases and answer them, or they can themselves find the answer if we give them access to those databases, control access.”

So, that's how it started.

And then I talked to our colleagues in digital transformation, which was, by the way, that was happening in 2018, and I asked them, “What robots, what tools, what can we do? This is a situation, what can we do?”.

And very shortly afterwards, I remember in all details, in specific details - it was October 2018, they said: “Let's give a presentation to the management and let's suggest, let’s propose a proof of concept project for three months. Do you want to be part in this, this proof of concept?”. And I said, “By all means, yes, let's do that!”

So it was HR and Finance to the proof of concept project. We started the last quarter of 2018. We were completely amazed by the results. So, very, very early in 2019, we realized we didn't have enough engineers that are knowledgeable with this technology with RPA programming; we couldn't find any outside because there were not that many available. So, we said, “Let's train them! Let's do some reskilling initiatives and see if some of our own people want to learn this.”

So we started our first, actually, it was another premier, it was our first reskilling project. We announced that we are looking for people willing to learn the RPA programming technology. We had less than 300, 280-something applications from all over the bank, like risk analysts, front office clerks, credit analysts, salespeople and managers, and all sorts of applications.

We selected down to 20 on different layers of interviews.

The funny thing was that most of them said, “We don't know what you are talking about, we don't know what an RPA programmer is, but it sounded interesting enough, so we put our application forward.”. Some of them, as you mentioned, had 15 or 18; I remember one colleague was 18 years in the network, in a specific agency doing the same thing. So he said, “I don't know what it is. I don't know what you want from us, but I’m in, you can count me in.”

And then we had partners. We outsourced these partners who were training them for three months, I think. And then we, they, we had the graduation ceremony. Ten of them, right away, right after reskilling program, were transferred to Digital Transformation Office. And that was another premiere because we allowed them to work from wherever they were. We didn't have remote work at the time, but we said, “Let's start with it. We don't ask people to move for this, they could do this from their own place.”. So, they were very happy for that as well.

And from the other ten that we didn't have jobs for right away, six of them, we delegated them, we sent them for six months to Moldova, where we have another bank, another, the third largest bank in that country, so that they automate some processes and they actually skipped some stages, they ran directly to the robots.

All these people are still in the bank, and the loyalty and the enthusiasm they had by learning that and working on that – that’s priceless.

Kieran Gilmurray:  Oh, Wow, Gosh! That's right in the heartland of UiPath you started so many years ago and it's quite funny there weren't many people there. That's the extraordinary because now there are so many people who know automation.

But, Nevenca, how did you make it work? Not in terms of the developers, but how did you convince people in the bank to use Conversational AI assistants as part of their daily tasks and roles?

Nevenca Doca: It wasn't that difficult. We thought it would be difficult. We were not aware of how prepared people are to use this. So, we ran a digital readiness survey, we developed it with the researcher in this field, we developed a survey asking them: how familiar they are, what they are using, where do they get their news, where do they get their holiday tickets; all sorts of questions just to see if they are using already or they are willing to use more technology. And much to our surprise, more than 80%-82% had no problem in adopting or using such technology. So, we said, “Okay, we are ready. Let's do it!”

That was one of the first things. And then, another thing with the same impact or even larger, we partnered with some of our colleagues. We said, “This is what we want to do. We are developing an HR robot, an eight-hour chatbot, and we want to make it work, we want to make it friendly, we want to make it familiar to people.”. So, we put together a team of 10-12 people: “Give us some ideas or do it yourself.”. And, as you can imagine, the people's creativity is something, that's why I love people, by the way, people's creativity kicked in and then they, you know, they gave a face to - they gave a name, first of all, her name is Aida, it's not a random name either, in Latin means help or helper - so, they gave her a name, they gave her a face, obviously she’s a fashionista, she is very friendly. They taught her to have a sense of humour, to be a little bit cheeky when colleagues, because that happens to, some colleagues, they ask her all sorts of things and that's fine; I mean, we allowed these things to happen. So, by partnering with our colleagues and making her as human as possible or as friendly as possible, adoption was immediate.

You know, preparing for this talk, I was looking at the reports for the last three years, and we have, like, we all, everybody, all colleagues are using her, single users exceed the number of our current employees, because there are also some people that left. So, everybody is using her.

Kieran Gilmurray:  Oh, wow! Because you normally don't associate technology with personality. So to be able to do that, to create something, someone that people want to engage with, is brilliant.

But what improvements have you noticed? They're using the robot or conversational agent. But how is that actually impacting your business in terms of your customer-facing activities, your internal user experience, or your banking customers externally?

Nevenca Doca: So, first of all, direct observation: on payday, phones are quiet in payroll, and I’m in department. So, first objective - ticked. Because people can now very easily find out everything they want. They have access, they ask for the formulas, you know, all the details for their pay calculations computing. That was the first. But then, obviously, looking at the number of conversations she has. By the way, last year she had more than 4.4 million conversations in one year.

And the beauty of it, if I can use this word in this context, in March 2020, when lockdown was implemented almost overnight, we had no problem in transferring documents, communication, information to our colleagues. We were open, obviously, the bank was functional, was operating. In Romania at the time, you could travel, relocate or commute from one place to the other, only if you had certain papers or certain, specific reasons as, for instance, was the one that you needed to go to work, your employer requires you to go to work. So, we distributed all these certificates, all these documents instantly and people use them.

So, all sorts of benefits we have there, I mean, for 4.4 million conversations a year… Imagine how many FTEs we saved only in HR!

And now we have robots, not only in HR, we had to finance in HR initial projects, we have them everywhere these days. For external customers, we are now using them in call centre, in our Internet web or mobile phone applications – everywhere, people can just easily access, like, to see information that they want; statement of account, the balance of their account, the next mortgage date or whatever they need to do, not necessarily going through the app, but just quickly to the chatbot- they have an identification and just ask questions. It's a huge, huge help for all sorts of areas. Imagine risking compliance, doing some checkings, some reports, and it's incredible.

Kieran Gilmurray: I can imagine just what you mentioned earlier on Nevenca, and the conversation volume is fantastic, but people don't want to do some of those, dare I say, boring but necessary tasks. Therefore, if they can be offloaded, fantastic! And that's the bit I like, what someone once said: taking the robot out of the human, and I adore that phrase these days.

What would be your advice to organizations that want to use Conversational AI in their organization?  

Nevenca Doca: I would give them the same advice, to do the same things that we are doing when we are launching an idea or testing idea. First of all, get the people on board. Get some people on board. And I have, I’m here, by the way, we didn’t say that, but I’ve been here for 20 years now. So, it's, yeah, I've been around for a while, and…

Kieran Gilmurray: You started young, you started very young.

Nevenca Doca: Thank you, thank you. And I have colleagues, we have colleagues that I can count on – I call them ”early adopters”. They, they are willing to test and to give you the real feedback. I don't need, I mean, it's not useful to have politically correct answers all the time. So, I count that their feedback is objective and constructive. So, that would be my first advice. Identify the early adopters that you can count on with objective and friendly, constructive feedback for improvement, not for destruction of an idea, and work with them. They have the perspective of the users, not of the one that had the idea and believes that he or she knows all the answers. So, use these people. People are willing to jump in and cooperate, to co-create, I’m fascinated by this tool or by this concept, co-creation, it's so powerful and we are not using it enough. And that would be my first advice: you identify some people and you involve them into this idea.

The second advice would be, if it's possible, and, in this case, for instance, it was possible, to create a story around the technology or the tool that you're in, create a story that's appealing to people, that's friendly, that's a bit funny, you know; it doesn't all need to be so serious and so boring.

And then, people will adopt it, you know, easier. And then there is another thing. Don't expect that 100% will adopt it from the beginning. There will always be a number of people that, no matter how friendly, and no matter how useful, and no matter whatever beautiful story you create, they will not use it or they will criticize it or they will, you know, complain about it. Forget about them. Just focus on the other ones that are willing to use it and that appreciate it, and you grow from there.

Kieran Gilmurray: Well, I love that! It's that thread of people that you keep talking about: they're involved in every part of it and have been communicated with as well. Nevenca, one thing we cannot ignore is the future of work and the future of work that is currently being impacted by the adoption of Generative AI technologies. There's ChatGPT, and Bing, and Bard, and a whole lot of others.

How do you see the future of work in general? How do you see it for folks in the bank, knowing that Generative AI is such a massive thing?

Nevenca Doca: Yeah, I was surprised as, probably, many other people when ChatGPT was launched and became part of our life, because, as others, I was too convinced that it will take a while until technology will take over creative jobs, you know? - The jobs that were said to be left to humans for long, from now on, for many years from now on, like creating jobs, for instance. And we see that ChatGPT and other AI that's acting these days are doing just that, are, you know, creating visual art or creating text, literature and other sorts of creative work. So, that's a surprise.

And I think we have to rethink how we position ourselves against technology. Still, I believe that technology’s here to help us. And it should be technology helping humans, not humans being threatened or replaced by technology. If technology can make our life easier in terms of work or, you know, our personal tasks at home or in other parts of life, we should embrace it and we should make use of it. We cannot avoid it. We can resist it for a month or two or, you know, maybe a year. But it will overtake us. And in Banca Transilvania, since 2019, as I mentioned, that was the first reskilling program that we've done, but since then, because it was such a success, it created so much engagement and joy.

Actually, people saw that, you know, people around 35-40 who thought that their life, professional life at least, is over. “I will be a banker for my whole life. I will retire, you know, grumpy risk analyst” or whatever. And they saw that there is a second curve, to use that term. There is a chance that you learn something new, something that is in trend, something that is for the future, and the employer supports you to do that in every way with time, with money, with resources. And it creates a lot of attachment and engagement, and for the future of work, this is at least the way that I'm thinking of it.

Try to foresee, as much as possible, the next steps and try to prepare your people, at least some of them, for these small next steps by reskilling them. From what I… my experience, the best investment you and most technologies today can be taught in very pretty limited periods of time, maybe not to the expert level, not to the senior level, but good enough, so people can start working with it. In our experience, someone who was in a banking role for 15 years and then learned the technology, immediately started producing added-value, producing value, because the induction was unnecessary, it was not needed, and they had all the other knowledge that usually technology people don't have into that much depth about regulations, you know, products, customers, and legislation and other leaders.

So, it's a good investment. It pays off and also retention afterwards, engagement afterwards is very high. So, future of work for all of us… and this is what I'm telling, for instance, my daughter who is 18 now and graduating high school, and last year she had done a period where she was totally agitated and stressed, and her language was “I'm anxious that if I’ll make the right choice now, all my life I will be a loser. You know, I will make the right career, professional choice, and I will be unhappy all my life.”. And I told her, “Don't put such pressure on you. You will have the chance or you will need to change every ten years, whatever you want it or not. At least every 10 years, if not more often. Because things will change, technologies will change, you know, life will change, habits will change. So, don't put that much pressure. Just make a choice that suits you now, or sounds good now, and you will see in five years if you want to switch or not.”.

So, for the future, work is similar. We can foresee some of it, some of the future, most of it, we can't. Being ready for the change, having this mindset that life is ever-changing, life is not static, life this dynamic, and we have to learn, you know, what's needed. If ChatGPT is something that is needed to be more effective and more accurate, or I don’t know, quicker in certain and rest, Why not use it?

Kieran Gilmurray: I like that. I like all that advice. I have an 18-year-old son. I'm going to tell him the same thing, except not be as good or fluid as you describing it.

So what's next for you and Banca Transilvania?

Nevenca Doca: Well, we are evolving so many projects. One of them is developing a fully digital bank. Banca Transilvania acquired another small bank in Romania, moved all the business to Banca Transilvania, closed the branches and is preparing it to be launched as totally, fully digital bank, the first one in the country. So we have a team there and growing, that is working on that, which is purely technology-based. And… that’s an interesting project itself. Besides that, Banca Transilvania grew, initially, organically and then, the last decade, through a series of mergers and acquisitions. We are still interested in growing in both ways: organically and by acquiring other banks when an opportunity arises. So, we have many, many plans. People are always the essential part of it, and I hope that, at least for the remaining of my career, people will still be central to this business.

Kieran Gilmurray: I somehow suspect they will be. Nevenca, that is excellent. Thank you so much for today. I am not surprised that Banca Transilvania is so successful and whilst you might not say it, I finally feel that you are one of the primary reasons why it's such a great place to work. I love that investment in people. I love what you're saying there around augment and not replace. I loved the idea of having multiple careers. I don't think any more banks or places where people go to die or not, in the case, Banca Transilvania. I'll maybe rephrase that Banca Transilvania is where people go to thrive.

Thank you so much. That's such an amazing conversation today.

Nevenca Doca: Thank you. If you allow me to add two more things. 

One a disclaimer: there are more of us, it's not only me. There are more in the bank that love people and, yeah, it's a team effort in management, and senior management, and middle management, and everywhere in the bank, actually. The culture is very friendly and informal, and yeah, there are many. It's not just me. That was the disclaimer.

And the second thing is that the retirement, you know, people go to banks to retire. Actually, in Banca Transilvania, some people are coming to retire because we have excellent benefits for people that retire from Banca Transilvania. So, we have people that retired after a long career with the bank, but also people that are joining few years before retirement and then they enjoy all the benefits.

Kieran Gilmurray: Well, where do I sign? That's the question, just a year or two to go yet, but then I'll be knocking on your door in the future.

Nevenca Doca: Okay. It’ll be amazing!

Kieran Gilmurray: And thank you very much! Thank you!