Mistakes and Tips on how to use data in recruitment with Sergej Zimpel

Thomas Kohler:
Today’s guest, Sergej Zimpel.

Sergej Zimpel:
Is the data that you have really correct? Do you really have all the data that you need, especially if you work in bigger companies? If you also have more complex systems as a workday or as SAP, these things are really correct. It’s not so easy to get the data. And then you also have to really, really make sure that the data is also correct. So it would also be a wrong approach to just look at the data and then start managing with this data. You first have to understand this data. Correct. And does it also really show the reality?

Thomas Kohler:
Definitely true. Therefore, I also think not just in recruiting, but also from transitioning from system a to system b. That’s, I think, where a lot of bottlenecks are there, where maybe you have the correct data within the system. When you look at it isolated, but when you look at it comprehensive, then it’s not correct.

Sergej Zimpel:
Yeah, I agree. And then you also have to use the data. What I also often see in companies is that they have the data and that also someone is measuring the data, but no one knows about the data. Often, I don’t know. A ten to hire is then reported to someone in the management. But if you then talk to the recruiting team, they might know about attempt to hire, but about nothing else. And they never look at the data because they are also so busy with operational stuff. Not because they don’t want the data, not because they aren’t curious, they’re just so busy or they don’t know about it. And I think that’s also key. As a leader, as a manager, to be transparent by default, to be as transparent as possible, and also to share your insights with the team.

Thomas Kohler:
And also maybe another problem adding to that is, okay, if the data is maybe then communicated to the right person, how is it communicated? Is it communicated in the sense that the right context is delivered, that it’s benchmarked against what is good, what is bad, what does it mean, what’s the storyline? I think that’s also key. And what do you want to achieve with it?

Sergej Zimpel:
Yeah, absolutely.

Thomas Kohler:
Sergej and I talked about data driven recruitment, how to convince stakeholders, how to get certain things approved, decisions approved, moving projects ahead by using data, using funnels, using also holistic data, using the right storyline and communication frameworks on how to place the data for the right people at the right time, with the right information and get to effective results.

Today we have Sergej here in the show, and I’m really looking forward to this episode because it’s also a topic of my heart. Data driven working or data driven recruitment. And it’s more than just looking at funnels. And today we drill down a bit into that topic. So maybe we start with a short introduction about yourself, Sergej, of course.

Sergej Zimpel:
Hi Thomas, thank you so much for having me. My name is Sergej. I’ve been working in HR talent acquisition recruiting now for a bit more than twelve years. Worked for very different companies. Everything from tiny, very, very early stage startup to big corporates. Also worked on the investment side. And now I work at the Autobahn. So that’s basically a public company. That’s like a company owned by the german state. That’s maybe the better description of it. And it’s basically the highways authority. And I think super, super interesting because it’s a bit company between the public sector and also it’s still a company. And I like companies with challenges like that a lot. As you already mentioned, super important for me, data topics.

But I’m also very, very passionate about talent acquisition. I’ve co founded two communities. The first one is called Purple Squirrel Society, that’s basically mostly focused on our Snack channel. And then ta crunch, that’s a meetup series based here in Berlin. Besides that, also a big, big passion for leadership. And I think this also really goes hand in hand with our topic today, data driven recruiting, because I strongly believe, and we probably will dive deeper in that in a bit. But I strongly believe that if you want to follow a modern leadership approach, data is crucial, but you can also use data to really fuck things up as a leader. And I think that’s a super interesting challenge. And as somebody mentioned before, I like challenges and I’m super happy to talk about this topic today.

Thomas Kohler:
Okay, then maybe let’s talk with the data fuck up piece, because I think that’s also sometimes where we can avoid and certain things where do you can fuck up things with data or with the wrong data.

Sergej Zimpel:
Yeah, I mean that of course, always like a personal opinion, right? But if you ask me, I’ve known a couple of people who started their career in a big recruiting agency and I think for some agencies, what they do well is that they measure everything, but that’s also what they don’t do so well because they basically only use this data to put a lot of pressure on the team. And if you don’t have enough telephone time, then basically you’re out. I mean, maybe that works for a certain environment, that’s okay, but it’s not an environment that I would enjoy. But how I always like to describe that they see data as the endpoint, right? You look at your data, you look at your phone time, and if your team does not meet that phone time, they are fired. I think that’s the wrong approach. And I think that’s also what people are often afraid of. If you tell them, oh, now we will start measuring data, and now we will also have a very detailed look in your funnel. They are afraid, oh, no, no, I’m controlled or I’m micromanaged. I don’t want this. And that’s also often the reaction that you get from teams when you start.

Thomas Kohler:
Measuring agencies, but also for internal teams, right? Because sometimes I think there are maybe, let’s say there is a structured hiring manager, like hiring manager that is also measuring the sales teams based on that, or an engineering leader that needs a logical explanation of every problem. And then they just look at, I would say, solving the problem one sided by quantity data, for instance. Right? And then say, hey, but to make this higher, we just need to do this activity. So why don’t we? Why aren’t we able to do that? Okay, then secondly, what’s the quality? Do that many candidates even exist, right? Or even looking at a scoping problem, right? Is there really, well, already existing or not? Right. So that’s, I think, something where you can really fuck things up with data and when you don’t know how to display certain areas, then this can put you in a very difficult situation. Because also, when you are put in a place where the data that is expected to be delivered is not solving the problem, then I think it’s hard to go back and then reset the expectations based on unrealistic expectations set already instead of upfront setting the expectations realistically and clearly.

Sergej Zimpel:
Yeah, fully agree. I think there are so many things that you can do wrong. I’ve done lots of these things wrong in the past, and I’m pretty sure that I will continue to do mistakes, to make mistakes. It’s normal, I believe. But for instance, every company I worked for, data quality is always a topic. Is the data that you have really correct? Do you really have all the data that you need, especially if you work in bigger companies? If you also have more complex systems as workday or as SAP, these things are really correct. It’s not so easy to get the data. And then you also have to really, really make sure that the data is also correct. So it would also be a wrong approach to just look at the data and then start managing with this data. You first have to understand, is this data correct and does it also really show the reality?

Thomas Kohler:
Definitely true. Therefore, I also think not just in recruiting, but also from transitioning from system a to system b. That’s, I think, where a lot of bottlenecks are there, where maybe you have the correct data within the system. When you look at it isolated, but when you look at it comprehensive, then it’s not correct.

Sergej Zimpel:
Yeah, I agree. And then you also have to use the data. What I also often see in companies is that they have the data and that also someone is measuring the data, but no one knows about the data. Right. Often, I don’t know. A time to hire is then reported to someone in the management. But if you then talk to the recruiting team, they might know about attempt to hire, but about nothing else. And they never look at the data because they are so busy with operational stuff, not because they don’t want the data, not because they’re not curious, they’re just so busy or they don’t know about it. And I think that’s also key as a leader, as a manager, to be transparent by default, to be as transparent as possible, and also to share your insights with the team.

Thomas Kohler:
And also, maybe another problem adding to that is, okay, if the data is maybe then communicated to the right person, how is it communicated? Is it communicated in the sense that the right context is delivered, that it’s benchmarked against what is good, what is bad, what does it mean, what’s the storyline? Right. I think that’s also key. And what do you want to achieve with it?

Sergej Zimpel:
Yeah, absolutely. And that’s, I mean, what we already started to discuss a bit, especially if you start using data for your team. I like introduced data and recruiting for a couple of companies. And really for like each company. For every company. The first reaction of the team was, oh no, I don’t like this. This is micromanagement. Right.

But then, and I think this, this also like shows how important communication is. And I think that’s also what some people misunderstand. We have technology and we have data, but it makes, that make. Only makes communication more important, not less important. You really also have to really like sit down with your team. You have to really sit down with each team member, sit down and really explain why are we doing this. Give full transparency what we are measuring, and then really also work towards a positive error catch, right? Because everyone has fuckups in their recruiting funnel. Also, like when I record recruited myself, I made so many mistakes, and that’s normal.

And I think what’s important, if you, as a leader, then look at your funnel data and see, I don’t know, there are candidates that are stuck in a certain stage for a month. No one has reached out to these candidates. Of course, that’s not good. But the trick, at least for the first time, is not then to blame your recruiter, but to sit down with them and to tell them, let’s have a look at the funnel. What I realized, looking at your funnel, that we have a couple of candidates who haven’t heard from us for months. Let’s discuss why. And you mentioned context. It’s super important, because I might have one idea about this data, but the recruiter might have another idea.

And then you can really have a conversation starting without blaming, where you, together with your team, try to understand why this happened. And often the reason is not in the person of the recruiter, but in the process. For instance, they have too many active positions, they have an annoying hiring manager that does not give feedback. They got ghosted by the candidate. Also happens. And you really don’t have to sit down and to discover these things, because my main task, my main responsibility as a manager is to help my team to do better work. I’m not the expert for their work. They are way better in that than I can be. But I have the strategic view on it, and I can sit down with them, discuss it with them, challenge them on their processes, and work together with them on improving them.

Thomas Kohler:
Nice. And I think that’s a very important piece. What you mentioned that data driven approaches are not micromanaging. And I think it just gives you control of the situation and also visibility. And ideally, a person or a recruiter, for instance, understands why the data driven approach is important and what you can do with it, instead of, oh, now I measured, now I micromanaged. That’s, I think, the wrong mindset. Right. So how do you get them on the, I would say, right attitude to be data driven?

Sergej Zimpel:
I think the most important part is really not blaming people, but really sitting down with them and to show them, look, also happened to me in the past that I forgot about a candidate, that the process needs too long or whatever. But let’s look at this and see it as a problem that we can solve. And then the half, like, if you continue doing this and you also have to look at the funnel every week or every two weeks, maybe every month, really depends on the person on your team, then they also start thinking about the data more independently and themselves. And if they know that you will also discuss this with them and everyone on one, then they also start looking at the funnel when they are not in the one on one with you. And then after sometimes a couple of weeks, sometimes a couple of months, then suddenly your one on one start with your recruiter telling you, oh, by the way, I looked at the funnel yesterday, and what I realized is that we have, I don’t know, far too many candidates in the interview stage for a certain position. Or I see we really have an offer acceptance rate problem here, right? So they really start thinking about this, that themselves. And that’s really also what I want. Because at that point, I can stop looking at every detail. I can really pull myself out and I can give the full responsibility for that to the team. And they can also really be the facilitator for their one on one with me and really point out where I can support them.

Thomas Kohler:
And do you also have examples where maybe hiring managers can be convinced by, I don’t know, changing a role profile, changing behavior, collaborating better, seeing you as a trusted advisor instead of recruiting, is it delivering value?

Sergej Zimpel:
Oh, yeah, many examples. I mean, maybe the, I mean, I think every recruiter, that’s the situation, their career, that had a really annoying hiring manager that never gave feedback, always took like, way too long. And what really helps to sometimes then sit down with them and to show them, look, this is your average time to give feedback, and that’s the company benchmark. You need, like, I don’t know, double the time that other managers need to give feedback. And then they’re like, oh, okay, maybe we have to change something about the situation, right? You have hard data to really discuss it. And you can also use this hard data to have good or far better conversations with the top management. Because if you start measuring stuff, and maybe again, an easy but very powerful example, if you just do easy things like track the source, that’s not complex, that’s not hard. But if you then can show, look, there are so many candidates that we hired through active sourcing, and we know if we would not have done active sourcing, we would have needed a recruiting agency on that.

We saved so much money. Let’s hire another source and we can even save more money. I think this is really an approach that we don’t see enough in HR and recruiting departments to really actively also use this data to have better discussions with the business and also with the top management.

Thomas Kohler:
Yeah, definitely. And how would you, for instance, showcase through data when something is not realistic in terms of expectations? Do you also have an example there? Because I think that’s often the case, right? That a recruiter gets this difficult hiring manager who swapped recruiters maybe two, three times already because he was never, or she was never really satisfied. Then suddenly the third recruiter also sees, well, if we continue that way, I won’t succeed because.

Sergej Zimpel:
I think depends a bit on the level of the manager and also how data driven they are as a hiring manager. In best case, they take some time to really understand the problem and they also enjoy looking at data. Then I would like start with, I don’t know, average hires per recruiter per month and time to hire and to show them, I don’t know, you want to hire like 100 software engineers in three months. Look at our time to hire. Look at our team of like two recruiters with two hires per month. That’s not possible, right? But then of course, nonstop there and tell them, look, we can’t do, but really like, then dive deeper and then say, okay, let’s look at the funnel and let’s understand why each recruiter only makes two hires and not four, right? This might be a problem of not having enough candidates. And again, active sourcing could be a solution. Employer branding could be a solution, a recruiting agency could be a solution. Or we lose all candidate at the end of the funnel because our offer is not good enough. And then maybe you can also use this annoying hiring manager as an ally to have a discussion with the top management, to maybe discuss our salary of our candidates, to be able to provide better offers. So of course that’s the best case. Sometimes you do that and the hiring manager is still not convinced, of course. But I think that’s still a good approach.

Thomas Kohler:
Nice. When we abstract that now, I think that’s the very macro level on headcount. And I think another solution could be phasing out rows to say prioritization, right? Do you really need all the, oh yeah key or can we phase it out along the year? And then we have maybe around 30 each quarter. Is this not enough, for instance? Right? And then I think another example, what is very present, I think with critical hires that are just with evergreen rows that never failed, right. Is often that the scope of the role is always changing. The judgment process or the evaluation process of several evaluators is different with each candidate and the rides, I would say seasonally, that’s a big problem. Right. And therefore, I think what really helps is also showing the real funnel, the individual funnel, as you said. Okay, from that stage to that stage, against company benchmark or against industry benchmark. Oh, yeah, we are far behind. Yeah. So what we could do is either asking for certain attributes, what we want to understand early on in the process, and prequalify more. And then if we do that, and that’s still not the solution, then we might need to re scope the profile and say, maybe we make two rows, one piece of the job can maybe already be done by an internal employee, or we just split the headcount, maybe even ask for additional funding internally to say, actually, that’s two rows. Because the recruiters now brought me 15 candidates. And if we would have two types of rows, we then could solve the problem by splitting it up. And initially I was just not aware of it, right.

And the third piece is, how do you really make them accountable to stick to one evaluation effort, and therefore, you really would need to define, okay, what do you want to achieve and how do you ask for how to evaluate what you want to achieve and what you look for in an answer and what is good, what is bad? Instead of, oh, I have not a good feeling with this candidate because he or she made something weird.

Sergej Zimpel:
Yeah. I have a very concrete example for what you just talked about. In a past company I worked for, we had this very niche role and we were really struggling and filling this one role. And then I was approached and I was told, oh, now we gonna fill 15 of these roles. I was like, yay, this will be fun. And what I did is basically I just pulled some external data from LinkedIn to show them, look, there are only like so many candidates on the market. And what the siring manager also needed is german speakers. And then I showed them, look at the talent market is not large anyway. And now you’re only looking for german speakers. This is like, I don’t know, 5% of the talent market. We won’t succeed with that. And if we then have this data control, then again, it’s also part of our responsibility to also propose solutions. Right. It’s not enough to just say no, but really come up solutions. And what we did is exactly what you said. We said, look, we have this role. This is a bit of a software engineering role. If we split this up ten only focus on software engineering. It’s okay for them to only speak English, but we have then five other roles that also have contact to our clients and they need to speak German, but they are not so heavy on the technical side. Problem solved. We were able to fill these roles. And I think this is, again, getting this data is not so complex. I think often if you go to a conference and see one of the big companies talking about their data and recruiting, it’s often very fancy stuff and very complex stuff. And that’s awesome.

Don’t get me wrong. But I also sometimes feel some people then think, oh, that’s so complex, we can never do that. Might be true, but not be true, but what’s important is to start with what you have, because we almost always have more data than we think and we can use each bit of data. If it’s like LinkedIn data of our tenant market, if that’s our time to hire our funnel data, our hiring sources, that’s all data that we can really use to make better, better decisions and also to really convince also our stakeholders to do the right thing.

Thomas Kohler:
In case you have any feedback or anything you want to share with me, please send me an email on thomasseoplewise.com or hit me up on LinkedIn. And in case you really enjoy the show, please subscribe. I would really appreciate it. Definitely. And that’s something also, I think what regularly needs to be trained and challenged, because I think you really just learn on what to use in terms of framework to approach certain stakeholders when you have been in a situation already, right? And I think also the more training you get upfront maybe, or the more shadowing you get, or the more just you can perceive, the easier it is get for you to spot these things up front or you have a strong leader that is supporting and enabling you, you in that situation instead of doing it for you. Because that’s also often the case, right. That then somebody else is jumping in, you don’t know what’s going on, suddenly the problem is solved, but you never learn how.

Sergej Zimpel:
Yeah, I think what happens still way too often is if recruiting teams don’t do that, and if recruiting teams don’t do not identify, think that they can improve proactively, then other people will do that for them, right? Then that’s like the annoyed hiring manager or the management who then jumps in with some otherworldly ideas. What can be improved? Some of these might be good ideas, some maybe not so good. And then you have to deliver on these ideas. And I think it’s much better if we are way more active and proactive with that and really come up with these ideas ourselves.

Thomas Kohler:
Cool. Yeah, definitely. And do you have any tips or best practices, what you can share to get your team ready for being data driven? Because also that’s often a learning process, right?

Sergej Zimpel:
Yeah, I think maybe. Let me start with an example for if you add people to your team, if you maybe have an open position, something that I did in the past and that I still do when I hire for my teams, is to also have a small and simple case as a part of the hiring process. I’m not a big fan of cases where candidates need to invest, like a full week or whatever. Doesn’t make sense. But basically it’s just, it’s hiring funnel, a messy hiring funnel, and they basically have to look at it and have to solve it, have to figure out what’s the problem here, what could be an approach to get better? And there’s almost always more than one approach to solve it. Some people also come up with better ideas than I have. I think that’s important. And then in your team, I think it’s important, again, to be as transparent as possible with the data.

You have to show the data to also share your own experiences as a recruiter and your own mistakes. Not blaming people, but also to encourage them to share their learnings. Maybe also fuck ups, but that also needs a bit time, and that needs a lot of trust with the team. So what I also often did, for instance, brownback lunches or lunch learning sessions where we would talk about our own experiences with data or maybe also share some nice article that we found online. There are so many sources that you can use for that. But really, to establish this culture of trust and openness and transparency is at least, if you ask me, almost the most important part of creating a data driven team. Because if you have that, then you can have open conversations about the data. Then you, of course, also need, as a manager, as a leader, provide this data, and then really, people will start using that data proactively.

Thomas Kohler:
So do you also, at your current company, are you already onboarded that? Well, did you say, hey, yeah, state of the art in terms of data driven approaches there, or do you need to build something out or what’s the plan for you at the moment?

Sergej Zimpel:
Yeah, I mean, data is always a journey. I think it never done with data. I just started right today. We have, I don’t know, we have the 22 march. I started my new job on the 1 March. So as you can imagine, I’m still trying to understand what data we have. And we also talked about data quality. We have to figure out, can we trust this data and then also start using that data in our daily business? What I already did, and I talked about that a couple of times already today, but it’s really so important to really also already sit down with the team and tell them, look, I will start measuring stuff, but not to micromanage you, but because I want to do a better job as your leader, as your manager. And I think that’s an important first step. And then you need to say that again and again and again and again. But again, understanding what data we have, understanding the quality of the data, and start talking to your team about it as early as possible are the most important steps.

Thomas Kohler:
Thank you, Sergej, so much. I think that’s a very nice end to our episode, and I really enjoyed it.

Sergej Zimpel:
Thank you. I also enjoyed it a lot. Thank you so much.