As the latest research we have conducted at Insight222 finds, People Analytics functions continue to grow, with over 50% of leaders we surveyed saying that their team has grown in the last 12 months.

My guest on this episode of the podcast is Daniel West. His 20 years of experience in HR Leadership roles at data-driven companies like Uber and Apple have taught him how analytics can drive business performance, improve customer outcomes and enrich employee experience and culture. Daniel is now the founder and CEO of Panalyt and the technology his firm provides is helping organisations bring together and democratise their people and business data and enrich it further with external and social capital data to drive decision-making. You can listen to this week’s episode below or by visiting the podcast website here.

In our conversation, Daniel and I take a deep dive into organisational network analysis and how social capital data can be linked to performance, engagement, sales and innovation. We also discuss:

• The level of analytical skills required by HR professionals and what we can learn from our counterparts in Marketing

• The impact of democratising data across business stakeholders and how this drives more agile decision making and improved business outcomes

• The critical role of ethics and trust in People Analytics

• A great example of how a Japanese company had to virtually onboard over 200 new starters due to the pandemic and used People Analytics to help that effort

• What HR can do to prepare their organisation for an increase in remote and hybrid working

This episode is a must listen for anyone interested or involved in People Analytics, Employee Experience and Social Capital.

Support for this podcast is brought to you by Panalyt. To learn more, visit https://www.panalyt.com.

Interview Transcript

David Green: Today I am delighted to welcome Daniel West, Founder and CEO at Panalyt to The Digital HR Leaders podcast. Daniel, welcome to the show, it is great to have you on board. Can you provide listeners with a brief introduction to your background and your role in setting up Panalyt?

Daniel West: Yes, thanks very much, David.

My name is Daniel West. I am the Founder and CEO of Panalyt. I have been in in-house HR for over 20 years. Early in my career, I was at Morgan Stanley as the Head of HR and Systems, Data, Processes for APAC based in Tokyo. I then joined Apple as the Head of HR for Australia and Japan through the high growth, iPod, iPhone boom, and then moved with Apple to the US where I ran US Sales HR and the Global Online Store expansion.

I then came back to Asia as the Head of HR for a global commodities firm, a company of about 17,000 people based out of Hong Kong. I did that for about four years and then joined Uber as the Head of HR international. Basically we had about 300 people between APAC and Europe but we grew that to about 4500 people in about 18 months, which no one should ever do again, it was a tremendous learning experience.

And then like a lot of ex Uber Executives, I ended up working for a couple of VC’s. I supported, super high growth companies that were getting to their series A or Series B that were looking at huge expansion in their headcount. I basically helped them to set up their HR for scale, buy in their HR technologies. And, I did that for about two years before founding my own startup, Panalyt, we did that about three years ago.

David Green: You have seen it on both sides of the fence and we will talk about how your experience plays into what you are doing now. You have been in HR in all different companies at different stages which I think is really interesting. And now you started Panalyt, I think back in 2017, and you are focused on what you call “practical people analytics,” I would love for you to be able to describe that for our listeners and what this means? And why it is important?

Daniel West: It really does come out of, that kind of hands-on experience as a practical HR practitioner at very data driven companies. So you have fantastic access to your financial data, your sales data, customer data and at Uber we had a system sitting on the desktops of Managers where they could in real time, see the driver and rider behaviours. They knew when to put in a discount and when to put in driver incentives.

And this is an incredibly complex dataset, sitting behind the tool, but because the people involved in designing the tool understood the source of that data and understood the challenge that they are trying to achieve through the data, they were able to simplify the interface so that Managers could grasp that data and take action on it, again, in real time. So, where that kind of falls down with people data is typically HR people just like myself, they are not particularly technical. I have got no better an understanding of analytics than my colleagues. HR people that are not particularly technical and not particularly data oriented . But then you have the BI folks who don't understand the source of the people data, and don't necessarily understand the challenges that you are focused on. I mean, there are some great BI professionals out there, but they tend to be very trained in understanding the front end of the business and solving those challenges and then just tend to not be that interested in the people data.

So you end up with analytics that are either, too simple to take action on because it is HR trying to drive it themselves off insufficient data and with insufficient skills to do that kind of analytics. Or it is over complicated because you have got the BI teams that can source every piece of data and throw it all in and come up with a fantastically brilliant visualisation, that doesn't actually solve the challenge you are looking for.

So with practical people analytics it is trying to find that balance and that is why I think at Panalyt, what we are doing by having our own data science team, building the tools ourselves, we build the dashboards ourselves and with myself and a couple of colleagues like me with the HR background, we are trying to make sure that we understand the source data. We understand the challenge that is trying to get solved. We understand there is tremendous complexity in the data, but we are building interfaces that is simplifying that and making it accessible and real to Managers. The other thing about practical people analytics is it needs to be repeatable. So there is nothing more frustrating than applying a whole bunch of BI resources and getting a Data Scientist on board, doing a one-time piece of analysis and then realising that, okay, hey, we have now learnt something tremendous about attrition or engagement or performance, but then if we want to take action and then measure it again, we are going to have to apply all these same resources again. If it is not repeatable, you can't get into that virtuous cycle of seeing the data, taking action, seeing the data again and seeing that you are doing the right thing.

So if it is not repeatable, then it is not practical.

David Green: So the principles are the same as, you gave the example of your time at Uber when you have got that sales and customer data at the fingertips of the Managers so they can make real time decisions. You are really looking for the same thing applying to people data so that they can make decisions about their teams and everything else in real time.

But it is simplified in a way for them that they don't need to be experts in analytics.

Daniel West: And I think when you look at the historical change in Marketing, so how Marketing has gone from being not particularly data-driven at all, where now half the headcount of any Growth Marketing Team is Data Engineers and Data Scientists. They went through that evolution because it wasn't just that now we have a lot more data, now we really need to understand how to use it and how to make it actionable. So it wasn't the Marketing Team trying to leverage some Data Scientists time, it was Marketers becoming Data Scientists or Data Scientists getting involved in Marketing directly.

Whereas HR, I think, is still very much in the position of having to beg, borrow and steal Data Scientist time to get anything done on a one-off project and that doesn't lead to scalable practical results.

David Green: You are talking about scale and obviously one of the areas of scale, as you said, is about making it repeatable. I think another area of scale that we hear a lot from organisations that we work with and others around the world is we need to increase the capability across the board of our HR professionals around using data. They need to be more confident of it, they need to be more capable and they need to be able to ask the right questions. In your opinion, given that you have been a HR Professional for most of your career, what level of analytical skill does the HR Practitioner themselves need?

Daniel West: I think there has got to be a clear line between the folks that are already in HR like myself, who have had 10, 20 years doing it and folks maybe just coming into it now. I think there definitely should be a push and there should definitely be a good career path designed for grabbing budding Data Engineers and pulling them into HR and showing them why it is interesting and relevant to do.

I think for folks that are already in their HR career, I think it is madness to think that we are just going to reinvent ourselves. But what I really do think that any HR person can do, in terms of making themselves more relevant or making their organisations more relevant to that people data discussion and making what they do, more relevant to creating actionable people analytics, is, understanding the activities that you are designing. So you are designing a new on-boarding process, you are designing a new engagement survey, the decisions you are making are either going to make that data more useful or less useful.

And I really worry about the number of HR professionals that are still doing engagement surveys once a year. And I meet a lot of them. And I think there is something fundamentally missing in, how you are understanding how that data is going to be usable, because they are the same folks that are doing onboarding surveys at one point in the year, not when people are finished onboarding, just everyone who onboarded that year, we are going to survey them at this point. The data is so far removed from when the activity was happening that it is essentially useless. And so I think that HR gets criticised for running surveys that aren’t very useful and I think that criticism isn't often that valid, but in this case, it is valid because that survey is fundamentally useless because it is so far removed from the source of the activity. So with a very small change in the design of your process, you can create data that will be actionable. Even if it is actionable a year later when you finally get some Data Science time, but it will be sitting there and it will be usable because you did it at the right time. Let's say you asked the right questions about the right activity and that all got stored somewhere and then that will be massively useful. But if you are not doing it in an intelligent way, then it is not going to be practically useful.

That comes up a lot when we go to clients and we explain the difference between, terminations and resignations, not being that actionable, but regret, non regret, attrition being tremendously actionable, but now so much of their attrition is in the history and they don't remember if it was regretted or not.

David Green: I think, like you said, it is partly about HR professionals understanding the importance of data without actually becoming the data experts themselves. But understanding the importance of it and why it needs to be collected, when it needs to be collected, how and everything else.

Daniel West: Yes and recognising that that is going to require some process change, but probably not that much of a process change, considering the positive impact of what you are doing.

David Green:  I know we are going to talk a little bit around ethics and stuff like that later and obviously making sure there is a benefit to the employee of collecting this data in the first place.

But I guess you drew that really great analogy with Marketing, ultimately a lot of the data that the Marketing Teams collect is about customers, customer preferences, customer loyalty, voice of the customer etc etc, to help shape products and services. I guess it is the same with HR, the more you take the voice of the employee, the more you understand their frustrations, things they like or things they don't like and their ideas especially if they are dealing directly with customers, that you can use to actually help improve the business. So I think there is definitely that. And I think the other thing that is worth talking about from an HR practitioner, I think as a function we have too often looked internally and not often enough looked externally about what are the key challenges facing the business, with a business strategy and where the business is going.

And it is that ability of the HR professional to have those conversations with key stakeholders in the business, and then identify what is some of the people analytics that we could do to help drive those business outcomes?

Daniel West: I think this is something that obviously HR is criticised for, for not being on the front lines and not supporting the business directly. I think it goes back to that question I was saying before about, you have got to understand the business challenge in order to design the analytics that are going to support that business challenge. So we often talk to clients that always have attrition as a challenge. So attrition is a challenge, we are losing internal knowledge, we are losing sales contacts. How do we address that? And so we built a number of tools that help you to understand, for example, the relationships between salespeople and their clients. So that then if you do lose sales people, you are able to then sort of pick up those relationships using the network communication data that we can pull in and align with people data. Or we could take a step back to identifying what the drivers of attrition are. So we can do predictive analytics on that attrition and then targeting that to, who are the people that are critical to the business, and that is where we should focus our energy.

However we then also, having encountered a number of clients particularly because we do a lot of work in Japan, where attrition is just not a problem. They are just not losing people at all. The challenge that the company has is that the people don't leave the organisation, They don't have a performance culture that at least pushes people out and so their real challenge is effectiveness or productivity and engagement links directly to that. So then how can we start to adapt to all those tools that we have been using to identify attrition risk and leverage that to identify engagement risks, and also, how does engagement link to productivity?

David Green: It is interesting, all the best HR people that I have met, whether they are CHROs, Heads of People Analytics, Heads of Learning, are also good business people and I think sometimes we need to remember that we are HR people but we are Business people as well. I think that is important.

What is the impact that you have seen of democratising the data across stakeholders and speeding that process from insight to outcome?

Daniel West: We talk a lot about data democratisation, it is one of the fundamental things that I want Panalyt to achieve. I fundamentally feel that one of the sources of breakdown between HR and the business is that you are kind of talking a different language, at least in the terms of data and understanding how the people data is illustrating certain behaviours or hinting at certain activities in the people.

Managers who are very familiar with certain financial metrics, even if they are not in finance, they understand key financial metrics because it is put in front of them every week, every month it is made to be important to them. Whether they started their career thinking, oh yeah, I really need to understand a P&L but they will understand a P&L by the end of running their business for a couple of years. Because people data isn't in front of Line Managers in the same kind of automatic way. It is not, again, made useful and relevant to them and structured into that day to day understanding of their business. But then when you go to a Manager with an attrition number, with a gender pay gap analysis that they don't understand the underlying data, the underlying data isn't familiar to them.

So, I think an average HR professional can see an attrition percentage and know instinctively based on your industry and region, but still you know instinctively whether that is a reasonably good number or a reasonably bad number. And the difference between the first six months attrition and the long-term attrition, you understand the difference in that.

And again, what those numbers should look like. If I think about the best commercial Managers I have known, I used to work for Tim Cook and Tim Cook would visit the Japan business at Apple and he could look at a massive spreadsheet of our entire quarter's business and he would pick out the one field, which is the one question we didn't want him to ask about, because he just recognised where the numbers were just slightly off, even though he couldn't say why. And no commercial Manager is able to do that looking at attrition data, looking at pay disparity data, looking at gender imbalanced data because it just isn't familiar to them. And I think if we were more routinely putting that data out to the Managers, trusting Managers to take this data on responsibly, they would become more familiar with it.

Therefore, every discussion between HR, their HR partner and the Manager is going to get easier because there is a familiarity. That is fundamentally what I think is important about democratisation of data.

I think the other aspect is getting employee's own data in front of them and I suppose this is really part of what GDPR is pushing companies to do. To make data transparent to the employee. But I think there are very few systems that are doing this well, but I do think if employees understand the data that is being held, they have the ability to say that is not accurate, this is what I think it should be or I don't think you should hold that data, tell me why you are holding that piece of data? They have a sense of trust that they are seeing everything. Then I think that that level of trust between the individual and the organisation is going to be so much better. And then when something does go wrong and we have seen stuff in the press recently where data was being held without the company's permission by Managers, then the company can stand up and say to the employees, this wasn't sanctioned. You know what data was sanctioned because you can see it there in the systems. This wasn't a sanctioned data collection and, those people are out of the business.

And I think those situations where you have built up trust allows you to weather the storms when illicitly kept data emerges.

David Green: I mean, at the end of the day, as an employee, what use do I get from this data being collected? Is it going to help my career within this organisation? Is it going to help the way I am treated? Is it going to help my performance? All those types of things I think are good for employees at the end of the day and good for the organisation. So I think it is about getting that balance.

We talked about democratising data, we have talked about HR having a responsibility to really understand what are the business challenges and linking that to some of the work that we can do with people analytics. But who is responsible for ensuring that they know, i.e. the stakeholders, how to use the data and the insights effectively?

So if we are democratising the data, that is one thing, but how can we ensure that they know how to use it?

Daniel West: You can't get away from the fundamental point that we are talking about people analytics, it is people tech and the people team is going to have to show the leadership of saying, this is the data that we are showing to you, this is why we got it and this is how you can use it.

I think you have got to be the subject matter expert. And I think this does go back to your earlier question about what are the modern skills of the data driven HR professional? I don't think it is becoming a Data Scientist, it is about knowing and understanding the principles of the data, how you are collecting it, how you are using it, it has got to be the practical application of the data.

And so, I think the educational aspect of that does lie with the people team.

Two trends that kind of pushed me towards starting Panalyt. One was obviously seeing the massive explosion in SaaS computing, the specialist SaaS tools that do a lot of very very specific things in the employee life cycle really well, producing huge amounts of data that HR was completely incapable of using.

And so that always frustrated me and I always felt like we have got to solve that problem. We have got to collect that data together and make it actionable.

The other trend, which I still think is continuing, is HR as a percentage of total headcount just keeps on getting smaller. I think HR headcount is always under pressure. There is clearly a perception that you have now got all this technology, you don't need so many people and in a way I sympathise with that. The technology should be making us more efficient, but as I said, each of these pieces of technology is allowing us to understand an aspect of the employee experience and the different levers you can pull and I think we are pulling resources away from HR, just when we have got so many more levers to pull where we can possibly influence the organisation. I think what you are referring to about the educational aspect is part of it. We have got an opportunity to democratise the data, put more data in front of Line Managers. But we don't necessarily have the resources to apply to do that training and be the kind of real hands-on business partner that you want to be, to educate all your Line Managers to use this data effectively.

Years ago, back in the days of things like PeopleSoft, where you are putting in the pre-online pre-SaaS, HRIS in, the selling point of putting PeopleSoft in and you can get rid of half of your HR admin team.

But unfortunately what that always skipped was the reality that you then needed to hire almost as many people to manage PeopleSoft and to manage getting the data into it and getting the data out of it and making sure that you are utilising the system. You were supposed to be getting more out of this investment, not just getting rid of people and staying flat, because technology doesn't work that way.

So I think unfortunately we are buying in systems, but not supporting that with the resources of people.

David Green: It is kind of the evolution of HR, isn't it. Yes we are taking out quite a lot of the transactional stuff, which is good, the more admin related stuff. As you said it, so now with these additional levers we have got, we can create more value. So it is not just about shedding numbers from HR. It is actually changing the mix of the HR population and being more focused around driving business outcomes and personalising the employee experience or making it better because if we do that, that adds value to the business and it is the right thing to do. And it is what employees expect because employees are consumers at the end of the day and the technology they use as consumers is all personalised for them and it is relevant. Whereas the technology we have historically used in HR, is one size fits all. So there is a huge opportunity for HR, which we are seeing and people analytics is kind of at the forefront of that and underpinning a lot of that in many respects.

Daniel West: I do worry about the number of investors that we have pitched to over the years that have leaned in and said “that is really interesting, so basically you can get rid of HR by using this tool companies just won't need HR people anymore.” And I'm thinking, what am I doing wrong in my pitch, that is giving you that impression? I am starting to feel genuinely guilty now, but it does mean that there is a number of investors that think that those are the right goals for organisations or they are hearing from businesses where their goal is to find out how do we just get rid of the HR function completely?

And I know that is obviously not our goal. Our goal is to empower a better relationship between Managers and HR and to let HR sort of move on to the value add part rather than the crunching data part. But yeah, it is a worry that that concept is out there.

David Green: Yes. It is to create more value rather than saving money. I mean, it does save money as well, but it is the creation of value I think that is important.

I know from conversations we have had previously Daniel, we are both pretty passionate about relational analytics, ONA, social capital or whatever you want to call it, there is all different names for it.

So we are going to dive in a little bit deeper on that and I know it is a topic that our listeners are very interested in as well. Can you give some examples either from Panalyt or from your career previous to that of where, we will call it ONA because it is the shortest word, has been used most effectively and why that was?

Daniel West:  It is a fascinating dataset to get your hands around and to get into a format where it is kind of usable and you can start playing. We have spent a huge amount of effort making sure that our interface satisfies our goal of making it usable by Line Managers. You don't need to understand the underlying principles in order to start playing with it. The really early use cases have come out of this whole working from home experience that we have been going through this year. We have got a large Japanese company that onboard 200 to 300 new grads every year. This is their first year where those two to three hundred new grads are not coming into a ballroom and having in-person lectures by the CEO and then spending the next six months in little teams moving around the organisation, learning everything in a really hands-on way with daily check-ins by their manager. Japanese companies are an absolute machine about how they do new hire onboarding and it creates such a deep loyalty that will last the rest of those people's careers and that has gone, that just doesn't exist. So there has been a tremendous angst around how are we going to onboard these people? It was even a consideration of just saying, don't even bother, start next year. Just write off this first year of employment, just go home. And obviously because the HR group was already using our people analytics and then we were putting together this relational ONA data together with the people data, they saw the opportunity where they could, on a real-time basis, see how those new grads were forming relationships within the organisation. Comparing that against, if you spin back in time to last year's new grads, look at their relationships that they built in their first six months, try to map that against how this new set of new grads out there are forming good relationships.

So we are tracking the metadata out of email chat and calendar. So we are seeing the metadata, not the content, but the metadata gives us who they are connecting with and because that is all aligned with people data, we can see are they making relationships at different grade levels? They are not just all making buddies with each other. That they are making relationships across departments and even across locations.

So once the team saw the potential of this, they are actually trying to now make this a positive rather than a stop gap measure. It is like, okay, well now we can just do global onboarding. So now spend this time, do these projects with these international teams because now we can actually track whether that is actually working.

So they are able to drill down to the individual to see, is this person advanced enough along? Are they making the connections that are expected? Then do a direct intervention to coach them, to make those relationships richer.

And so that has been a really positive use case during this work from home crisis, that I think, we have instantly positively impacted those 300 new grads lives in a way that we feel really good about.

So something from Uber. We were kind of looking at it the other way round. The analysis we did at Uber that really got me into understanding the potential of the ONA data from the very beginning, was us looking at the last hundred or so exits that we had in Head Office Technology Group. And we were looking at those 100 exits, looking at the slack data, at Uber we lived on Slack internally, we didn’t use any other tools. By looking at those hundred people's activities, there was always an expectation that we would see some change in their communication activities prior to them leaving. I was working with the Data Team and we kept on pushing the timeline back and back and back. And you realise that it is three to four months before almost everyone left and it didn't matter whether they were terminated or resigned. You saw the number of meaningful relationships that they had within the organisation started to drop dramatically and it hit a floor. It never hits a zero. There were always a handful of people who are your actual friends and actually the average relationship score, the average value of those relationships, went up. So if you are looking at this data the wrong way, if you were just looking at the relationship score, you would never see this.

But if you looked at the number of people where they had a high number of relationships, it literally fell off a cliff. Obviously for some people it falls off and then recovers, their Manager has changed or their job role has changed, or they are just having a bad week, but where it falls off that cliff and then sustains for three to four months, 99% of those people left the organisation. Then we interviewed them and out of those hundred people, the vast majority, 80 to 90%, reported that they agreed this was happening but they saw it happening three to four weeks before they left. So that means there is a three to four month gap in the data where the employee themselves isn’t perceiving this happening, where they are not perceiving that relationship drop, but then suddenly they are aware of either being rejected or that they are rejecting the company and then they leave or they realise that they are out. And so this becomes an incredibly powerful and very simple and direct way that we are now using the ONA data to predict loss of engagement and in the worst case attrition. It is something that is very, very, very visible and it is intuitive, it kind of makes sense that you lose that connection to the company.

The other thing that I also noticed that was interesting, it doesn't make any difference whether people are exchanging cat photos or whether they are actually doing work with friends, your relationships inside the organisation, the things that bind you into the culture, we never found any direct correlation between people talking about the nature of their relationship.

David Green: Interesting. Obviously bringing all those data sets together and I know in Panalyt you bring together classic HR data from the variety of systems that organisations use, this relational data from collaboration tools and passive data basically, and business data. So you bring all that together. What are some of the challenges in doing that?

Daniel West: The fundamental challenge, we had to address, kind of in our early stage, we spent a good year and a half building our data warehouse from scratch. It is an entirely custom build for the architecture to deliver people data to our front-end very, very fast. And so there is a real setup of how we ingest the data, how it is structured, the calculations that we are doing on a daily basis and how it is delivered to the front end, that is pretty unique to people data. And so we kind of got past that challenge of structuring ourselves. So people data in general, in the scale of the big data world, people data isn't that big. There aren't that many transactions that happen to an employee on a day-to-day basis. So any Data Scientist that has worked with us has just said, this is the predictably least clean data set that they have ever had to work with.

There's so much customisation that happens and I know this from being a buyer of HR systems, one of the first questions you ask is can I customise the fields. You expect to have that and that now being on this side of the fence, I realised what an absolute pain we built for ourselves in this. Because the HR teams basically customise everything because we just say yes to the business when they say can we track it this way? Can we call it that? And particularly recruitment systems are incredibly customised. So the HR data is incredibly hard to structure and so we kind of solved that problem.

Then we started looking at the communications data, which is entirely the opposite. It is absolutely enormous complex datasets, but very very clean because obviously no one customises their email systems.

So I think some companies that do ONA approached the data structure differently than we do, but obviously we are tying all of this to individual employees. So the way we are doing the ONA analysis or structuring the data as we put it in, is essentially tying it to the employee record and the relationships between two employees.

So we essentially treat the relationship between two employees as another data point that has attributes hanging on it and so we still structure it the same way as we do the people data. So we spent a long time solving one challenge that actually let us solve the second one and was actually not that hard. And then the knitting that data together, starting with the people data is actually quite a nice place to be because it's very rare to not have the company email within the people data, it is almost always there. And then that gives you the key to every other system. So the employee ID very rarely sits anywhere else except in the HR data. But if you can trust the email connector, then you can connect to any other system and comms or CRM data it is all driven off the same keys.

So I think our only real challenge is looking at datasets that aren't necessarily structured based on the individual employee. So there are obviously CRM attributes that are linked to Sales teams and then we do really have to look at, okay, what is the client expectation on how do they want to see this data?

Then we have to structure separate data tables for those objects that are not employees, it is not that difficult and there is always an employee related to the data in some way.

We have a certain starting point that gives us certain challenges, but nothing has really been massively insurmountable.

David Green: Yeah. Yeah. It is interesting, obviously bringing in that email data or data from other collaboration tools, there is a lot of chat around that both from practitioners, partly because of GDPR and other legislation. A lot of People Analytics professionals will tell me we want to do network analytics, but we have a challenge with our Privacy Teams, with our Employee Representative Groups. I say to them, well, don't start with the “I want to do network analytics” start with the business problem you are trying to solve. But even so, the technology has had some bad press fairly or unfairly, I don't know. What is your take and how do you advise companies should approach using this type of data?

Daniel West: Yeah, absolutely. I mean, we do have some what of an advantage that we are based in Asia. Most of our clients are Asia based. And so there is not quite as much focus on that, but also I think we have made some good, well I believe, we made some good fundamental decisions. One of those being that we are never touching content, we are absolutely focused on getting the most out of the metadata as possible.

So we are not taking subject lines, meeting headings, headers or the content of any communication. Even in some tools you are able to get not the text content, but you can get the emojis and I went, no, no, no we are not taking the emojis either. We are not taking any of this, we are not looking into what the employee is doing, we are looking into how they are relating to people and how they are building networks and who they connect to.

So that is a good starting point.

I think the other one is we absolutely, I can't say we require, but we very strongly recommend that every client that has done it, needs to tell their employees what they are doing. The number of clients have started off saying, “okay, we will do it, we will see what we get and then we will tell the people” and I have to say “No, no, no, tell them right up front, this is what you are doing and this is why it is going to be useful to them.” Which I think it goes back to what we were saying about the data democratisation, having an interface that is open to the employees. And this is on our current product roadmap to open up a lot more of this data and finding to an employee view. I think this is actually necessary under GDPR, but I think just a necessary thing to build employee trust. Start with the business challenge, start with what it is that we are going to discover. So we are going to discover this thing about how you relate to people and how that is helping you to do a better job. And we are going to show you so that you will understand it and you will know the same as what your Manager knows about you and I think that can only be a useful thing.

I think the fact that we are also starting a large part of our client use cases are around sales and sales effectiveness. So the responsiveness of a client to a sales person's communications, so how quickly the client replies to your email. Most sales people know instinctively that this is a key measure yet they have got no way of actually measuring it. So we are actually showing them that and so this is actually being shared directly with sales teams so they can see their own performance with client A versus client B versus client C. And they can kind of make sense of it, they can get coaching from their Manager, but because we are starting with that use case with a sales person's own data, there isn't any hurdle to get over. They want to see this, this is an active request from the employee base. So if you can find scenarios like that, where it is really employees that want to see this information, then yeah, I agree with you, you are not going to get that kind of pushback.

So you have got to start from that point of view of how is this valuable to us and to the employee. And again, total transparency.

David Green: Completely agree. And I think this leads nicely onto the question that we are asking everyone on the show in this series, I think there is a real link with some of what we just talked about with network analytics. What can HR leaders do to help their organisations for the future, as uncertain as it is, where there will likely be an increase in remote and hybrid working?

Daniel West: We have had tremendous interest through this kind of working from home period because it has made looking for any other ways to touch base with our employees and also doing it very, very quickly that has been important to companies. They feel that you can't wait to design a new survey and set up a new survey tool and get all that in place, in four weeks time, we can roll that out. Where obviously you can put Panalyt in place to do this basic level of analytics. We can get going in one to two weeks and then you are seeing the data in real time. The most important thing I think that the companies can do from this perspective is to get that data into the hands of Managers and make sure they do understand how to use it. And you are getting that real time feedback from Managers is if I could just see it like this, if I could just see it like that, this is more useful to me. I think empowering Managers with more information about their employees, particularly what has come up on a very practical basis is time shifting. So the aspect of where we had a lot of companies where their Managers were feeling, and it is definitely a feeling, that their employees were not working as hard. They were not as available online, weren't coming to meetings as much, or were kind of arriving late, leaving early to online meetings. And by looking at the ONA data, you are able to see that the actual total amount of activity, digital activity, was completely the same. People were shifting that from the 9:00 AM to 4/5:00 PM period activity was really reducing because a lot of people are looking after their kids, and then getting back to work after five, six o'clock. And so that there was a real time shift that on a practical basis, once Managers understood this about their teams, they could start to shift when one-on-ones were happening. Start to shift their stand-ups, their team meetings and when they expected to hear back from their people. For good Managers, it just kind of reset their expectations and the Manager adjusted and carried on supporting their teams. So I think having that kind of really simple practical data in front of Managers, I think is really valuable.

The other thing that we are hearing is companies trying to run a lot more pulse surveys and I suppose this isn't anything to do with Panalyt, but I have run so many of these things in the past. I end up consulting with most of our clients around how to set up a pulse survey, I think Uber was always one of the best companies I have ever seen running a really frequent really actionable pulse survey.

But then I think what the power of it is going to be is that having the pulse survey results in real time, or they are getting the survey results and getting it analysed instantly and running that beside the ONA data, because we can measure engagement or activity and network behaviour as a proxy of engagement together with the survey.

And what I think is incredibly interesting is when the survey says that someone or a team is super engaged, but the ONA data is showing that they are not connecting, they are not talking as much as they used to. So is the survey wrong? Is the ONA data giving you a misleading answer? It is actually that the two things are measuring two different points and the fact that they are disagreeing is telling you something really, really interesting but that depends on what the questions are on the survey. But I think there is something that you learn from having both the subjective and the objective data together.

David Green: I think we could probably talk all day about this, but we are not allowed to because otherwise it will be the longest episode of the podcast ever.

So I have got to say thank you very much for being a guest on the show. Can you let listeners know how they can stay in touch with you, follow you on social media and find out more about Panalyt?

Daniel West:  I am djw@panalyt.com. The advantage of having an odd name, Panalyt, is that we are very easy to find on Google and social media. So you will find us on medium and same on Facebook and on LinkedIn. You can reach out to me directly on LinkedIn or through my emails, you can tell I am always happy to talk about this topic with anyone who is interested. And David, I really appreciate everything you are doing to make People Analytics better understood, more professional and a more enjoyable place to work in. My whole team all super admire the work you are doing and we are super excited about being on here. We really appreciate everything you are doing. So thank you and thanks for this opportunity.

David Green: That is very kind of you to say that Daniel, thank you very much.