Summary
HR was traditionally considered a soft area with little quantitative data. However, with the advent of people analytics, HR has gained a stronger foothold at the decision-making table by providing data-driven insights.
Consultants like Japneet Sachdeva are critical in empowering HR with people analytics and developing new possibilities, such as generative AI.
In the People Analytics Talk episode, Japneet and Abhinav discuss the about:
– Why People Analytics is important in Consulting?
– How Consultants Use People Analytics?
– People Analytics Maturity in India
Key Takeaways
- People analytics is a core part of consulting work and has become a product in itself.
- Data is essential for people analytics, and organizations can start with imperfect data and iterate over time.
- People analytics provides quantification and ROI for HR functions, helps consultants stay ahead of the curve, and allows them to pioneer new practices.
- The maturity of people analytics involves a complete ecosystem, including data culture, structured insights, and automated actions. Data-driven decision-making is crucial in HR and can be achieved through people analytics.
- Attrition prediction models can help identify systemic issues and individual-level insights to improve talent retention.
- AI is transforming people analytics by providing actionable and hyper-personalized insights.
- Data democratization and conversational interfaces can bring people analytics to every employee and manager.
Full Transcript
Abhinav (00:00)
When large enterprises face critical business and people challenges, they reach out to big consulting companies like Deloitte, Accenture, and McKinsey. But did you ever wonder how do these consultants, who have very little idea about the company, get up to speed and end up consulting about what the company should do? The answer is they rely on data. And often in many business problems and all human capital problems, they rely on people data, in short, people analytics.
People Analytics is a core part of almost every consulting company, be it McKinsey, Bain, or Deloitte, and lately it has moved from becoming an enabler to a product in itself. Hi, everyone. I’m Abhinav and welcome to the Peoplebox Analytics Talk, where we invite incredible leaders to go deep into the fascinating intersection of data, AI, and people. Today, I’m delighted to have Japneet Sachdeva on the show. Japneet, a partner at Deloitte, is one of the most experienced people in People Analytics in Asia.
She’s also the co-author of People Analytics Maturity in India Report, which we spoke about in our very first episode. In her previous role, she led a large people analytics practice for global clients at Accenture. Welcome to the show, Japneet.
Japneet Sachdeva (01:13)
Thank you so much, Abhinav. Glad to be here.
Abhinav (01:15)
Japneet, talk to us about your journey in the consulting world and how you get into people analytics.
Japneet Sachdeva (01:21)
Okay, so I’m an MBA from IIM Kozhikode, and then Accenture was, you know, I was a campus hire for
them. They were also looking to build their HR consulting practice. So I joined what they used to call Talent and Organization Consulting, joined fresh from Campus and somewhere along the line
Accenture wanted to invest in this whole talent and organization analytics capability. And we set up a team. We started from basic surveys that were very in, you know, like 10, 15 years down the line. But from there, we evolved into, you know, more of analytics around, you know, machine learning, using more of algorithms, Python and statistics and stuff.
And then I think India got me to Deloitte closer to home, closer to my heart. So the last two years, almost two years now, I worked on setting up the people analytics practice in India.
And that’s how the journey has been Abhinav.
Abhinav (02:14)
That’s fascinating, Japneet. A lot of our audience here are founders, are CEOs.
HR heads and people analysts understand that for a company it’s a very critical piece.
So Japneet helps us understand why people analytics is such a critical piece in consulting work.
Japneet Sachdeva (02:34)
So I’ll tell you, I’m going to have two, three things, you know, why it makes a lot of sense. Okay. Number one, A lot of other functions, if you see a finance, a supply chain, a consumer, for example, they were pretty solid on analytics as a part of their business, right? HR, I mean, fortunately, or unfortunately, we were always the softer, you know, practice.
So for me what analytics brought in was that opportunity to say we mean business as well. Right. So whenever I have discussions with CHROs, it was my biggest, you know, sort of pitch to them is that, see, this is what gives you the seat on the table because you can go back and say that, you know, there is a concrete investment that the quantification that HR function lack. I think that’s to me, the number one, you know, what I see enabling all organizations to do as a consultant for me, that’s the biggest shift I can bring to my CHRO. Right. At the end of the day, you know, what a consultant should be able to do is
as an HR consultant, I should be able to help a CHRO and a CEO to make a better case for their people. Okay, so that’s number one, right? To me, it gives you the power of quantification. It gives you the power of sitting on the table and saying, this is the ROI I’m going to get to you for any kind of investment in people. Okay. Number two,
As a consultant, just staying ahead of the curve. That’s what everybody else expects from us. If we don’t do that, we’re pretty much not sustainable as a business. If I can’t stay three steps ahead of everybody else, that’s the core part of being a consultant. We are ahead of the curve. To me, that’s another opportunity analytics is giving. Gradually we saw, that while we started with basic analysis and analytics and service stuff, we have moved to using Generative AI for a lot of stuff in the HR space.
We’ve recently done something on individual performance paths, and learning development paths using generative AI. But I think a little bit of it’s also me as an Indian feels good about it, Abhinav. So while, I mean, as India, we are pushing boundaries, right? We are really going ahead and doing stuff that nobody believed we could do.
So to me, that’s also like a close-to-heart right part of it being as an in-depth consulting practice to say I am pioneering stuff that you know nobody else has done. So maybe those are the top three Abhinav that as a consultant I feel very proud of.
Abhinav (04:51)
That is so fantastic. I love the phrase that you said that everybody expects us to be ahead of the curve. And just going back to the thing that I spoke about initially when you go into a company, you know, there are experts there. They are, you know, people who have so much experience. How do you get yourself up to speed, understand the nuances as an outsider and are able to bring some, you know, out-of-work insights to help them achieve the objective?
Japneet Sachdeva (05:07)
Thanks for watching!
As a consultant, I feel we succeed only when we have a strong partnership, right? So nobody knows the organization better than the organization themselves, right? So I think
how this whole partnership works right between the client and the consultant is one.
we are able to provide a little bit of outside perspective to say that sometimes it’s okay to just take a step back and say that I know my stuff very well, but what is it that I can learn from the outside world? I think that is what sort of a consultant helps in whether we like it or not, but we would have seen similar stuff happening across organizations. So that’s number one. Number two, I think…
as consultants, we also spend a lot of effort and time to understand the organization, right? How it works so that it’s both ways, right? I mean, they open up to say, okay, let’s see what outside can bring to us. But we also open up to say, let us learn on how your business is structured, right? What you are doing so we can also best tell you what will work versus what will not work for you. So stronger that partnership is right, stronger that to say that we are working on this together rather than a consultant has come in to help us do something or whatever.
The stronger the partnership is more successful the outcome is going to be.
Abhinav (06:35)
Fantastic.
So, 16 years you have been in the consulting world, almost a decade in people analytics. A decade ago, people analytics would most likely be an enabler for other human or business consulting projects. Today, it’s becoming a product in itself. Today, companies are asking you to solve their people analytics problems. How did this transition happen? And who do you contribute towards it?
Japneet Sachdeva (07:01)
So I think…
two- three things that have changed, which has contributed to generally the growth of analytics and AI. One is the whole tech transformation, the whole digital transformation that has given us so much data to talk about. And in the people space, I mean, you see the biggest business that’s happening in human capital right now is the HR tech stuff, all the HR tech implementation, whatever the HR MS may be. That’s the big one. Now, what it’s doing is it’s giving you a lot of data, which we didn’t capture before.
I think second, is the whole treatment of data itself, right? Moving away from simple, you know, just having a survey and asking people questions. There’s a lot of possibility now to use unstructured data, right? What I generally call proxy metrics, right? I mean, I’ll tell you culture is an example, right? So typically, how would you assess the culture of an organization? Go float a survey, okay? How are you feeling? You know, is it good? Is it not good? I’ll tell you the past two years, I have done 20 culture assessments, and one survey.
Only one survey. 19 of them have been non-survey based. You use proxy metrics. You use more unstructured data. So if you want to see a safety culture in a mining organization, don’t go ask people, do you feel safe? Look at the safety incidents. So look at just the treatment opportunities we have. How much we can do with unstructured data these days? Today, I can look at a worker working in a factory and see if his helmet drops. There’s an issue.
safety issues. So, I mean, there’s just a lot of possibilities right now on what I can do with data. So I think one, just the data that you start capturing, two, the possibilities and I think third, just the general business data progress that’s happening, right, has helped people data as well. Now, if you’re investing in a data lake, right, I’ve seen, while we did this people analytics maturity that you mentioned, two things we saw. We saw that
the organizations either were heavy in consumer analytics, or actually picked up people analytics sooner because they already have a data lake set up. They already have capability to do predictive modelling, which they’re doing for consumers. Easy to replicate to people. Secondly, so all of these big tech companies, right? Very good at it because the capability is there. The appreciation of using data and analytics is there. So I think the overall progress of data and analytics in the business also helped people analytics.
Abhinav (09:08)
Good. Yeah.
Japneet Sachdeva (09:29)
So to me, I think, these are some of the things that have helped us progress as well.
Abhinav (09:33)
I think this is so interesting that you talk about data and, you know, especially the use case of not using the surveys And I mean, being an employee engagement platform ourselves, we see that there are so many of these non-employee driven data that you have, whether in their one-on-one notes, whether in their goals, it’s hidden in their, you know, reviews where you can go and identify whether the
you know, employees engaged or whether he or she is more likely to leave. But coming back now to the use cases Japneet, you know,
Having said that almost every people analytics expert that I’ve spoken with has mentioned that, you know, your output, uh, and, and then, the objective that you want to achieve out of the people analytics exercise, all depend on the quality of data at the end of the day. You know, it’s the garbage in garbage out. So when you reach out to these large companies,
Japneet Sachdeva (10:06)
doubt.
Abhinav (10:28)
I mean, it’s hard for me to imagine that they have everything in place, all data, all historical data, all in one place. And just ready for you to just crunch numbers and get inside.
Japneet Sachdeva (10:39)
Spot on Abhinav and I have yet to come across an organization that has all the data in place. So it’s, I don’t think so it’s happening in this world and that’s the reality of it. Right. I mean, data can fall apart anywhere. Data can fall apart when you enter into data, data can fall apart when somebody is updating the data. I mean,
Even today, we find those simple errors. What is a date format? Is it MM-DD- YY, or is it? So generally, my recommendation for Abhinav to most clients is that now there are two ways to do it. One is you actually invest a lot of time in getting that data right, having a data governance structure in place. And to be honest, some organizations are doing that. I mean, the Data Strategy Project that I was talking about, we’re doing a six-month just Data Strategy Project, just getting their data right.
It is not even getting to the analytics part, just getting the data right. So that’s one way to do it, right? That you actually, you know, I mean, the criticality of data is so important to you that you spend time and money and effort to get it right. Okay. But…
a lot of organizations, I mean, you’re also going towards that whole, you know, agile mode and I want to see something in six weeks. So if I tell somebody, you know, I’m going to take, and it takes time, right? Getting data right does take time. Okay. So if I tell somebody, you know, I’m going to take five months and tell you, just get your data right okay, you know, what are you even doing? So
Abhinav (11:49)
Easy.
Yeah.
Japneet Sachdeva (12:04)
Then there are types of organizations where we’re not willing to invest and not, I mean, it’s not judgment, right? We want to be quick, right? I don’t want to wait for eight months to see a result, right? So there generally I’m saying is give it a first pass, right? Even if you’re 70% on your data, what the analysis will at least help you do and see is that, okay, you know, unless I see a dashboard in front of me, I’m not even going
Abhinav (12:19)
Yeah.
Japneet Sachdeva (12:30)
to be able to tell you also whether this data is right or data is wrong. Right. So you need to start somewhere is what I say. So, you know, a lot of, a lot of my recommendation is let’s go parallel. Let’s, let’s do something. I mean, 70% accuracy is better than 0%. Right. So don’t compare 70% to a hundred percent compare 70% to 0%. So start somewhere and then it’s an iterative process. Right. You will get to 70, you will get to 80. And I’ll be honest, I’m working with my own CHRO, right? Deepti Sagar Chief People and Experience Officer.
Abhinav (12:42)
Yeah.
Japneet Sachdeva (13:00)
We’re doing a people analytics project with ourselves as well. Okay. And exactly the same approach, right? Let’s get the MVP rolling out. So we’ve just rolled out our first dashboard to our business leads. Let’s get the MVP rollout, right? Let them also come back, and say this is not right. This is not right. And we saw some major glitches in data, right? Our iteration has significantly gone down from last year. So, but
Abhinav (13:13)
Yeah.
Japneet Sachdeva (13:23)
At least start somewhere, right? You will not get to the moon and the star the first time, but at least start, you know, so that’s that sort of my recommendation on data. Data is not going to be perfect, but that’s our life, right? Nothing is perfect. So let’s start somewhere.
Abhinav (13:37)
That is so fascinating. You can see the smile on my face because listening to these startup analogies of MVP that we use and practice in everyday lives to bring the, you know, V1 or V0.1 in the hands of the user as soon as possible, even though it’s breaking. And hearing that, you know, large, these gigantic consulting firms also follow the same. It’s so nice. It’s so incredible.
Japneet Sachdeva (14:02)
That’s the way to go, right? We are in an agile world, right? Let’s try fail, go quickly and you know get on to the next step.
Abhinav (14:05)
Absolutely.
100% I couldn’t agree more on that and on the maturity bit, you know, Japneet you author this Fantastic report. I think everybody in my company every one of our clients that are shared with this absolutely loves that So fascinating work there, you know along with Nitin The report, you know, you spoke about you know, maturity of people analytics in India Could you explain what you meant by it? and you know, what are the different stages of maturity that organisations
typically go through like very sort of powerful insights that you got out of that.
Japneet Sachdeva (14:44)
So I think first and foremost thing Abhinav that I touched a little bit while I was talking about it is just in the thinking of it, just in the approach and framework of it, we try to go away from just the analytics maturity of it. So analytics maturity is your typical descriptive and then exploratory analysis, then neuroprotective and neuroprescriptive. So one thing we focused on is just one part of your people analytics maturity.
the ecosystem around it that will define maturity. We discuss a lot about data. So that’s an important part of maturity. Some of the softer elements that we generally don’t discuss when we discuss an analytics maturity, are equally critical, right? What I said earlier, is user engagement, and business alignment. I can create hundreds of fancy Chat GPTs and dashboards and everything. But if my user doesn’t like it, if my business feels this is not useful, so.
Abhinav (15:33)
Thank you.
Japneet Sachdeva (15:42)
I mean, one of the dimensions we also looked at while we were discussing with maturity is how business and the HR analytics or whoever is managing HR analytics are actually working together. And I keep giving the Google example, right? Google at the time of Laszlo Bock they were so successful because the business and the team, the team of his managing HR analytics was so closely entrenched, it was almost like a, you know, this is my problem, solve it for me using data, right? So to me, that was
Abhinav (15:57)
Yeah.
Yeah.
Japneet Sachdeva (16:11)
One big thing about the maturity, right? We moved away from saying that, you know, any analytics maturity is not just about analytics, right? It’s about a complete ecosystem culture we spoke about, right? Do we have the data culture in the organization? So that’s one part of the maturity. Second, in what you asked how are we seeing India, right? So, I mean, if I broadly look at it, we define four levels of maturity, right?
Abhinav (16:30)
Yeah.
Japneet Sachdeva (16:35)
pretty much where you’re all over the place, you don’t know where data is. Two is at least where you’ve done some sort of sorting, right, there’s some basic dashboards that come through, whether it’s your HRMS who’s pulling it out, or at least you’ve done something, you know, hands-on. Three is where you’re more structured and focused in terms of saying, I know what my business problems are, I have some sort of data lake where the data is getting pulled up from, and I’m deriving insights that I find useful.
Abhinav (16:45)
Hmm
Power BI.
Japneet Sachdeva (17:05)
And four is where we say that the whole loop sort of closes. So there is almost like an automated machinery, where I say, let me take the example of attrition where I say.
so and so is going to leave because his skill group is getting paid higher in the market. That red flag comes in either it action-orients to the manager to go have a conversation or it automatically goes to the compensation team to say you need to give the person a hike in the next year. And it’s almost like a self-running machine of sorts. Whatever the underlying, I mean, in certain use cases, even analytics is enough, basic statistics is enough as well.
Abhinav (17:44)
Yeah.
Japneet Sachdeva (17:47)
in certain cases, you might need AI and machine learning and complete the whole loop together. So irrespective of analytics, to me, the final state of maturity is to say that
preemptive issues have been identified, action has been determined, the system or person who is supposed to take the action has been generated and the loop goes back to say we took the action this was the ROI. So to me, that’s sort of the state for I mean we’ve seen I mean we were also a little liberal in our rating because otherwise nobody would have leaned in it so we have given people who at least achieved the complete loop in one or two of their use cases.
Abhinav (18:10)
Hmm.
Bye.
Japneet Sachdeva (18:25)
was most critical for them. At least they were there.
Abhinav (18:29)
Fantastic. I truly hope that with experts like you and companies like Deloitte, who’s actually investing so much in this, we see more and more companies moving into the fourth category.
One of the most critical parts of business metrics, like you also mentioned is about, you know, retaining your talent. It’s about the turnover or the attrition. Our audience would be super curious to know what happens, what does Japneet do? What does Deloitte do when a large company invites them and partner with them to just solve one problem, which is talent retention? What are the, what are the steps you take? What all do you
go into? Would love to hear that.
Japneet Sachdeva (19:08)
So two, three things, okay, Abhinav. So typically, one is a little bit of the education process also, right? One is the analytics part of it, right? Okay, so we typically build an attrition prediction model for them. We’ll use some of their internal data to see, what are we seeing in terms of trends of people left in the past. Why have they typically left? We’ll add it with some external data. So for example, I’ll give an example of external data. So skill group level compensation. A lot of compensation benchmark sometimes happens only at a level, for example.
And let’s say if I give my own example as a consulting firm, right, we would look at an analyst versus an analyst, how is an analyst paid in Big Four or other consulting firms? But within an analyst, you might have somebody who’s doing org design versus somebody who’s doing generative AI. OK, now they are not being paid the same in the market, right? So for us, both of them are analysts. And hence, we are putting them in an analyst company. So some of that external data, we try to sort of marry with the internal data
Abhinav (19:37)
Mm-hmm.
Hmm.
Japneet Sachdeva (20:07)
two, or three things, right? One, we’ll tell them what are the systemic issues we are seeing in the organization. So systemic issues could be one I just mentioned, right? There’s compensation for a particular skill group. Systemic issues could be certain managers or departments as well, right? You will see, and we see both kinds of things, right? We see attrition because staying with the same manager for a long time, attrition because the managers move too frequently, right? So either of them could come up as a cause, but that’s another.
Abhinav (20:30)
Thank you.
Japneet Sachdeva (20:33)
kind of systemic issue, right? To say that, you know, either people are not getting enough stability or people are feeling too stuck with the manager. So some examples, that’s one kind of insight we’ll give them. The other we also most often go into an individual level detail to say, based on the model, what the model is saying, what is the likelihood of an Abhinav or Japneet leaving tomorrow? Right? I mean, and again, you know, you look at all the factors you look at, you look at the performance, you look at the demographics, you look at
Abhinav (20:55)
Yeah.
.
Japneet Sachdeva (21:04)
If somebody has a skill that’s going to get a 50% hike in the market, then all your party culture and experience is not going to hold the person back. So there are multiple things that you will look at, rules, and flexibility.
And this is that we’ll tell them at an individual level, what it looks like. So that’s the analytics part of it, right? Which says systemic issues in the organization, individual level. Then there is, now what do we do? How do we action it around? So that’s one part. We also educate clients as, okay, so how do you intervene? Certain interventions are easy, right? Maybe you are feeling like the onboarding training is very simple, right? You just go and bandage it and you’ll see the impact. Certain are more complicated, right? So every time, I mean,
Abhinav (21:43)
Yeah.
Japneet Sachdeva (21:49)
clients to say that just because somebody is coming at an 85% likelihood of leaving, just don’t pick up the phone and call the person, you know, are you going to leave? So there’s a little bit of education, right? What you want to do. Another thing I felt is that I’ve actually lost a couple of attrition prediction projects up enough because of that because the expectation is that if we get the model, the attrition will reduce. That’s not what’s going to happen, right? So you will have to do something, right? There’s an investment that’s going to go. I mean, what’s going to happen?
Abhinav (21:57)
Yeah.
Yes.
something.
Japneet Sachdeva (22:18)
model is going to help you is to say that what needs to be done, right? What do you need to do? So don’t just, you know, randomly just that’s, that’s one.
Abhinav (22:22)
Yeah.
Japneet Sachdeva (22:29)
we’ve seen is also start measuring the cost of attrition as well. Now there are people who are going to leave, which is okay, right? I mean,
Abhinav (22:34)
Oh god.
Japneet Sachdeva (22:38)
But like, what is it, what is the cost of, let’s say, replacing in Abhinav versus replacing in Japneet, right? If Japneet is an easy skill in the market, it’s okay, right? I mean, we’ll find another, find 10 other people to whatever does this work that she’s doing, then my energy doesn’t need to be focused on that. So there’s a little bit of cost and the criticality of skill as well, right? If there is nobody else who can do people analytics, then you know, you want to keep, I mean, one, the skill itself, second, the demand of the skill you are seeing as well, right? If it’s a growing business, right?
I’m growing at a 50% rate and I will struggle if somebody in the team leaves. So the criticality of the scale also is sort of a parameter. It’s not just about, you know, personally leaving or not leaving. So these are typically the things Abhinav that we sort of do in an attrition prediction project.
Abhinav (23:24)
Amazing. That’s this space is so close to our hearts and I’ll talk to you about that as well But just want to follow that when you go into a company to take an attrition project A lot part of your diagnostic must depend on data So do you only rely on the data that is available to them and of course the external factors? Or do you also create data through either surveys focus groups or doing things that you know give you some external insights as well
Japneet Sachdeva (23:52)
So generally flexible client by client. So certain organizations do a very, I would say very uncomfortable doing surveys and focus group discussions. And we’ve also, I mean, there is a mix. We’ve also moved to some digital platforms. So we’ve tried new things.
As far as possible, I mean, the objective is, as unbiased and honest feedback as we can get, right? The whole point of data is also that, right? That it’s telling me something that people are not saying, you know, if I may put it that way, is right.
There are organizations who still want to do a survey, but I’m seeing more and more hesitation towards, let’s not do a survey. And even if we are using, I mean, we still use survey data in attrition prediction, but typically what we do is we use whatever standard survey they had done, right? Something they would have definitely done, some engagement survey they would have done. So we try to use that, then again, specifically doing another round of, some, and again, case to case basis.
Abhinav (24:38)
Mm-hmm. Yeah.
Japneet Sachdeva (24:51)
help right I mean sometimes I feel they’re all over the place right people start cribbing about the food in the cafeteria and you know I mean so that’s not the kind of insight you’re looking right the food is I mean yeah maybe we can change the food vendor but I mean food doesn’t keep you in the organization or makes you leave the organization
How many sources of data input do we use by client? It’s also sometimes the availability of data itself, right? If there is a lot of unstructured data or a lot of non-intrusive data, let me put it that way, is available, then I’ll not do a survey. But at times there’s nothing like that. I mean, there is nothing. And what do you do, right? Then you just go and ask people. So.
Abhinav (25:12)
Yeah.
This whole word of turnover prediction is really, really amazing. And I have two follow-up questions. One is, how long does typically this whole engagement take, especially the prediction one? And I know there is a lot of things to be done after you give a diagnostic report. And second is, what has been the most accurate prediction that you have seen in all of these projects?
Japneet Sachdeva (26:00)
So I think it’s a one, it’s an iterative process, right? I mean, the first ML model, predictive model, yeah, you will get in seven, eight weeks, right? That’s not too much of an effort, you can do it faster depending on how many functions you are considering and so on and so forth. So you will get to something in six to eight weeks, right? As I said, right, it’s an iterative process. Maybe the first time around, you’ll only get to a 60% accuracy of the model, right?
The more you keep feeding in the data, the more you keep using, okay, this person left, this person did not leave or etc. The more data you keep in mind, I have seen it go up to 92% in one case, which I think is alright. Yeah, I mean, I’m not even expecting more than that. So I think it, it.
happens with time, but we just have to be patient with it to say that. And again, as I said, we’re doing it for ourselves, right? And the way we’ve thought about this, it’s 60% versus zero, right? Today, I absolutely have no clue who’s going to leave, right? I mean, of course, there is some, there’s something from the engagement survey you will get, but there’s something from what managers are feeling about it. But there is no systemic way to say that, you know when I look at my, you know, CHRO and when I look at my CEO,
Abhinav (26:57)
Go.
Japneet Sachdeva (27:15)
from their perspective, there is no systemic way to go and tell X business leader versus Y business leader, watch out, you know, watch out for these people. For them, 60% is also a great start. So I mean, you can start with a 7 to 8 week, you know, is a 6 to 8 week, you know, you will get a good model running like a 67%. But yeah, the more you invest in it, the better it will keep getting.
Abhinav (27:38)
it’s hard for me to imagine how incredible a return on investment would be for a company that is absolutely at zero like they have no idea who is about to leave till of course the letter comes to even achieving a 60% or 92% I mean it’s if they would save so much of money they would have so much of control they can have contingency they can have backup plans now they can take actions which is, of course, all business dollars.
Now coming to the most interesting topic, which it was so hard for me to control myself to come to, which is technology and AI. We are a Gen AI company. We are heavily investing in this. You talked about AI in the previous answers. I’m very curious about how do you see the whole advent of Gen AI into the people analytics world and some of the trends that you’ve already seen.
Japneet Sachdeva (28:35)
Great question Abhinav. And the reason I say, you know, the people analytics was the best place to, you know, kind of capture the AI and GenAI use cases. They quickly came to, you know, because ultimately people analytics was anyways doing a lot of AI work, right? Now I see, I mean, I see so many use cases out there, Abhinav, and the discussion that I’m having or the ones that I’m running.
Because the inherent part of it is again the same, right? It’s data and it’s data’s, I mean, just the form of data has changed, the ability that we have to process the data. LLM has given us, you know, a new sort of life, I would say, for data analysts to, you know, the kind of processing of data that we do. So I see it impacting almost everywhere, right? Culture assessments, for example, right? All this recruitment, CV screening, I mean, recruitment end-to-end, right? I was talking about candidate matching, CV screening. Earlier, I have…
I mean, we can still use the LLM and customize a little bit basis, my organization, my stuff, et cetera. But just the potential it gives me like, I mean, it’s almost reducing a lot of my effort. You know, from starting from a zero code, I’m starting from probably an eight and a half or a nine in terms of code if I were to reach 10. So to me, I’ve seen I mean, it almost has given people analytics and accelerator, right? I mean, a lot of stuff that I was doing manually, I’m quickly, you know, using GenAI to do it. I mean, a lot of skills work.
We used to, I mean, we’re still doing a lot of work because it’s very custom and stuff. But at least that validation, you know, Gen AI quickly helps you, right? Even role skill matching, for example, used to be a big exercise for organizations, right? Can I get the skill ontology for whatever 100 roles that I had? Now you just go to Gen AI, I ask you to get a basic list. So to me, I see it as a big source of data, a big source of processing the data. It’s opening up a lot of opportunities to actually do more than, you know, what
Abhinav (30:04)
Peace out.
Japneet Sachdeva (30:27)
I was able to do. So to me, actually, this opened up a lot of more opportunities to do things faster to penetrate into more use cases as well. Right. Because you need speed, right? You need speed at scale to impact multiple areas. That’s what I think Gen AI is giving me.
Abhinav (30:42)
It sounds so sweet music to my ears. One of the very interesting things that we are doing is to solve this problem of data democratization using a Gen AI with a very strong access control. We are bringing the conversational interface into the hands of every employee, every manager, every HR VP so that they can also ask any people analytics questions about engagement, about who is working on which projects, what are the different projects being run. I’m curious to know that.
How do you see the impact of AI on the, you know, bringing people analytics in the hands of rank and file and how do you see this will impact the businesses?
Japneet Sachdeva (31:22)
So I’ll give you a…
this first use case that you know, just people that we worked on. We called it a fact or insight generator. That’s what we called it. It was simple, right? You upload your Excel sheet of data. And that was the first one, you know, we worked in as you upload the Excel sheet with whatever data fields you have. And you ask them, okay, what was the attrition in the past X rate? Even the descriptive analysis. And I know, I mean, as we keep sort of customizing the LLMs and using more and more of them, we’ll go to predictive as well. But even basic descriptive.
analytics that is not in the hands of a lot of CXOs today, Gen AI provides that. I’ll give you another example. A lot of times the question we hear from CHROs is that
See, I’m not going to go look into a dashboard and do the cuts and slices and see what data is happening. Okay. And primarily for two, or three reasons, right? One is how the dashboard is designed. So I mean, as people, analytics professionals also, and that’s why, I mean, people keep asking me this question, right? Why do you sit in human capital and you sit in analytics? And the reason I sit in human capital is because I want to design a dashboard that the CHRO understands So, okay. So a lot of part was the user interface, but a lot of parts is also that who is
designed for right? A recruitment dashboard will be looked at by the recruitment lead. Okay. He wants to go into the cuts and slices. Maybe the CHRO does not want to. So to me, the Gen AI is also giving that one layer in between to say on top of the dashboard or whatever, on top of that data.
you can still have the dashboard because it helps. It helps with certain things. But for somebody who wants to look at 50,000 feet, or for somebody who has a very specific question on something, can I just ask that question and not go into the whole dashboard and cut slices lightly? Sometimes if I want to see.
you know, what was my manager attrition from this particular department last year, I might have to go to three different drop downs on a power BI on a Tableau to get to that. So those simple questions and even more complex questions that you know, you want insights into. I feel it’s kind of providing that user experience.
ease of me getting insights, hyper-personalized insights. That’s another thing. I feel it’s very powerful with Gen-AI It’s not standard. Before this, personalization was happening only at the level of personas. Now we are saying it’s you. It’s up to your level that we can personalize. So I see a lot of power. I’m just from the insights generation perspective. I think there’s a lot it can do.
Abhinav (33:45)
Yeah.
We have them.
Yeah.
You put it so rightly, Japneet, it’s not just data, it’s insights, it’s actionable insights, and it’s hyper-personal insights. I think that’s the thing AI can do really, really well. I’m really hoping that there are more companies who are bringing all that into the hands of the users and employees. Probably my last question for today, Japneet, coming out of this people analytics world and just sitting in the shoes of a CHRO who has so many other things to do.
There’s one thing that you believe that AI is going to make the biggest impact, you know, for a CHRO. What do you think it would be?
Japneet Sachdeva (34:45)
I think I’ll go back to what I said up Abhinav Their ability to be more efficient, more effective, which gives them a leadership seat at the table, which to me, you know, sort of has been an aspiration that all CHROs should have, right? Why don’t a lot of CHROs become CEOs? You keep asking that question, right? To me, this is the game changer that it’s gonna be, right? We’ve never had that.
where they actually can talk the business language with ROI, with concrete results. So to me, this sort of is the biggest change that CHROs can expect out of AI.
What we’ve seen Abhinav is it depends a lot on the organization’s priority. For somebody who’s hiring thousand people a month, let’s say, and we have organizations, they are going to look at how it’s making their recruitment more efficient and effective. For somebody who is, let’s say, more into innovation, that’s sort of their theme, they are going to look at it, how does this provide more research, more content in the hands of my people.
Abhinav (35:28)
Mm-hmm.
Japneet Sachdeva (35:50)
30% sooner. For somebody who’s into coding, are my engineers going to be able to generate a code, I don’t know, 30% time faster than normal? So again, business context, and organization context, but there is an ROI around efficiency or effectiveness associated with each one of them.
Abhinav (35:52)
Hmm.
Again, that’s so true. You said like, you know, one of the very important aspects of Gen AI is hyper-personalization And that’s so true that the impact of AI would also be, it depends on the maturity and, you know, the business of each organization. And I’m very hopeful that, you know, the speed by which we are seeing the change in the whole people analytics in Gen AI world, it will be a very different conversation, you know, if you have it in a few months.
But, Japneet, I really want to thank you for first taking the time to speak with me and give some incredible insights to the audience. But also I want to mention, thank you so much for changing this career path from being insight provided to the global companies to bringing your focus to India.
Indian organizations because this is so important. You know, as India grows so fast, as we bring Indian organizations to the world map, Indian leaders, both business and HR need to be more data-driven, they need to make more people decisions, which are, you know, came out of analytics and insight, and you are at the forefront of providing. And I hope as a technology and Gen AI company, we are able to provide that through the new technologies and all that. Fascinating to have you, Japneet Thank you so much for your time.
Japneet Sachdeva (37:26)
Thank you so much, Abhinav. Thank you so much for having me. As you know, this topic is very close to my heart, right? I mean, I believe in the potential of people analytics. So, and as you said, right, being able to do it for Indian organizations is sort of at the heart of it, right? So thank you so much Abhinav for having me.
Abhinav (37:42)
You’re most welcome.
