Using People Analytics Beyond Pie Charts and Whizzy Graphs – Tom Ricks, Qlik



I'm assuming everyone can hear okay I have been slightly taken aback by just how much we have covered already that I had planned to go through so I'm actually going to try and whiz through beyond the pie charts and bar charts to some of the actual use cases that we've got a click so that you can see some of the kind of inspiration stuff behind this so as you know I've been introduced I am the senior director for HR systems and people analytic to click like Steve I have a very long job title they like to call my team spa s.p.a so the common perception around clique is that I spend all day by the pool it's not true not in the slightest my objective today is to showcase to you how we are leveraging people analytics to democratize as a word that we heard earlier and serve information to consumers speeding up our business and giving a better user experience this is all stuff that we've already heard about to some degree this morning even in the time between 8 and 11 and what I'd love to leave you with is is to think about who can use your data how it can be delivered and consumed to support people centric data-driven decisions while adding stakeholder value throughout your organization again these are all topics that we've already talked about this morning I'd like to show you a few of the ways that we are doing this click so I think very straightforward agenda I don't have long so I'm going to whiz through the first part who consumes HR data what does that look like and I really want to focus on four areas that click me and my team delivering people analytics is kind of trying to think of it differently to push people analytics in a different way people may not think they are consuming people analyst it's data but it is being served up to them in a way that they can consume it from our people analytic structure I will just say that these are all live apps these are all live and there's a lot of live data in there but I have jumbled everything up so what you're seeing is pretty mixed up please do not draw any conclusions from what you're seeing it's not real life so who are your consumers people analytics is central core to every business or at least the employee data is so we have a set of data that pretty much the whole business uses from time to time be that from the executive team sitting at the top of the tree through facilities events talent development the list goes on and on and on pretty much everybody in my business will come to me knocking on my door saying can I have please could you provide could you send me this report and so this is where I start who are I'm Who am I trying to democratize my data to in order to help me understand what that data needs to look like as we serve it back the common denominator here is to try and serve as many people as possible with a single piece of data if you can do that it reduces the risk of giving two different people two different sets of data so what we do is we have a core people analytics it's called PA v4 and clique and what we do when we get used case requests is we take from that and we create something new but it's the same data so it doesn't matter whether your deliv being delivered something very small in the you know example of a starter or a lever report or app or you're looking at it as part of the overall people analytics platform that we've built the data will always be the same between the two they used the same set of data they treat it in the same way one of the really big things here and kpi's benchmarks and measures we've already been here this is not what I want to talk about one of the really big things is being able to layer in the data we talked about it earlier in the panel and this view alone within our people analytics tool click uses all of these data sources to derive the data that's shown in that dashboard and one of the key things as you are building people analytics solutions is to think about your audience and to think about how quickly they're going to get lost if you present it in any other way than the very simple or the very basic what I've been trying to do is to present it in a way that people can then go further they can create a new view or they can self consume a different view but the idea is very straightforward you have to present it in a way that makes sense to the maximum audience you should not need to be a specialist to understand what you're looking at at this sort of level given that there is so much data being fed into this one dashboard I think we did quite well the questions always come later when they move to the next sheets so this is a really great use case for people analytics it always starts with a question and it's usually a very simple one how many who or where are our people which entity do they work for how many do we have in this country are they contractors or are they employees what kind of contractors are they do they have systems access you know in in the traditional sense HR has been very used to I go to my HCM I pull out an Excel report I look at it and I give an answer and within five minutes you get the follow up question so you go back and you pull out another Excel report and you look at that and you surf back the answer and then ten minutes late you get another question and by this time it's lunch it is really important that as your business is consuming data and I refer back to all of those stakeholders that sat around the people analytics platform it is really important that we give consistency because if those questions are being asked on Monday and then same questions being asked are Thursday chances are it's going to be different we work with highly dynamic moving organizations so what we did here is we pulled out of our people analytics platform a view that I provided to the executive team and their and the view is very straightforward you've got starters on the left you've got employed in the middle and you've got levers on the right and underneath there is a graphic to say that's the wizzy bar chart piece just to say how many of each you're looking at because people analytics has a calendar I can now provide this to basically anybody who want who wants or needs to know starters and levers information and the list is longer than you think sales enablement want to know how many queue seek quota carriers start so that they can organize all these sales enablement training talent development want to know how many employees we should what we have because they're looking for data around maybe the talent view process facilities need to know who's who started who's left so that we can provide workspaces for them in certain locations and at an executive level we just simply want to know how many people do we have how many people started and how many people left in a period so we can do all of that from within the app by simply selecting so I'm going to go in and just choose here in January 2019 so during January 2019 you can see in the app is telling you everything you need to know we had 92 starters we had 2000 136 employees and we had 36 who left those are names you can then drill into that and add more data as you need to to get to are they contractors or kind of contractors where were they okay so I provided that simple screen with pretty much an open distribution list this is the democratization of the data so if anybody comes to me now and says I need to know starter lever information I don't provide them with a custom report from workday I just give them access to this app and that means now that pretty much the whole of my organization is using the same data source through the same app I govern what's in there so that we know that it's accurate here's my organization it's relatively small and pretty much most of it is is centered around work their administration that is our system of record for our employee population this is provided via a tool we have a piece of software called org chart now some of you may have seen it heard of it but we we use that to visualize org charts this only goes partway though I have a blank spot there what am I doing with it it's already filled because this is from some time ago but what happens when we use people analytics to visualize an org chart well what we've done here is we've actually built an old chart over a point layer map sounds very exciting and it is quite fun and this is click this is what we looked like looks like a giant airline map right but every one of these lines has an arrow so you can see which points are reporting to which people so you can see here this top one has an arrow pointing from the UK into Canada and what I can now do is begin to visualize and understand the gap analysis so if I have a vacancy in my team where is that going to be it's not quite as important for my organization it's relatively small I know what's going to go on but when you look at somebody like Chris Jones and Chris is our Sales Leader when you're looking at an organization the size of his it becomes increasingly important to be able to place your next hire near to the customers that you want to be targeting or selling to or key customers so from this perspective again it's probably a to higher level he'll want to come in and look at more detail around that but if this is beginning to inform as to the network of people that you've gotten the geography that it's covering and how you want to cover it so the for example I've not got the selections available here but I could just look at a particular kind of sales person whether we're looking at key account managers or whether we're looking at territory sales managers and you could really hone in and understand what geographies you're covering from that perspective so I'm just going to take Chris away here the other thing that we have found this useful for is for those of you at clique you're gonna giggle with this one James Fisher always pick on James Fisher his afternoons must be a nightmare dealing with his direct reports in America because he's based in the UK so I'm guessing he works fairly much quite a lot of late nights the other thing that strikes me is so he's the head of product marketing he's our senior vice president product marketing where's his APEC coverage we're a global organization how is he serving product marketing to our APEC customers so this is a question that when I saw this that I started asking and I think you know despite all the panel discussion that we've had this morning one of the key attributes to anybody and a people analytics function is to be able to ask the next question to understand what you're looking at and to be able to say does that look right where are the gaps what's going to come next and for me APEC when I saw this was the absolute missing link so if somebody resigns from james's team he now has a decision to make our guest as he put somebody into a pack to cover that region or is he he's doing it himself because in a lot of our business areas we would cover a pack from maybe Europe but he doesn't have anyone there either he got one person in Stockholm perhaps but where is it going to come from so people analytics and the specifically here the Geo analytics piece of our software is is helping us identify and understand how our employee distribution is being laid out which in turn asks a number of other questions and by the way just in case you were interested I'm not sure you were but we'll we'll try it anyway and that works a little slow in here too here we go that same old chart that you were looking at for my organization that's what systems and people analytics that's where the spire team resides there's a hope there's a hotel swimming pool in every location not okay so that's its second use case employee networks being able to see over a geography your employee population to see the relationships between those points on the on the on the map we have also talked a lot today about reporting and the panel just now we were we heard that there were a repository of 250 reports that people could go in and self consume I see that slightly differently I don't want to deliver reports I just want to deliver an app I can control an app and I can control what's in there and if everybody uses the same data source there is likely to be multiple consistency whereas if you produce a report that ages immediately something will change within the 24 hours between the day that you send it to your recipient and that recipient reads it something will have changed if you can both just log in to the same system and consume it that way that's a much more robust way of consuming your data so on that note we've are these kind of some top-level thoughts around why I am using people analytics to deliver what I would term is reporting but it's just it's serving up data to the users so it produces reliable results we can use data preparation so everything that gets loaded gets treated in the same way every time you load that data which is hourly in the case of people analytics speed of delivery it's actually always there you don't need to stop and ask somebody for it no one in the business needs to stop and ask who were our starters and levers in January of this year they log into the app they make a selection and the data's there it becomes a trusted source if you guys over here log in and find one answer and you guys over here log in and find the same answer there is a much higher chance that you're going to trust what you're looking at you can get into the data quickly you can layer in the additional sources so we saw before we've got global force we've got work day we've got rest connectors to external sources as well a whole bunch of stuff and you can very quickly layer over any data with some new source so that you can begin to derive more value and more answers to the questions that your business are asking self-service once you've got access you've got access you can access it any time as I say we're on an hourly refresh so it's pretty much live data anytime you need it whether you're based in Singapore APEC whether you're based in California USA same system same data anytime you need it it's right there you don't need to stop and ask the team who might have gone to bed for a report from workday you can just go self-consumed it's always on bit at the bottom another real-life case this was my first project as I took over people analytics or kind of constructed people analytics we were producing a board of directors report every month and it was taking us we worked in it to be about two hundred and eighty eight hours a year that's quite a significant amount of time that is because there were seven people involved and it took them on average three hours each to pull the reports a work in Excel filter through the data work out what was wrong corrected aggregated back in produced the visualizations before sending it up to the board that now takes me an hour and a half a month so I pulled three reports put them into our app and away you go the visualizations all update we just take a screenshot pop it in a PowerPoint and send it off to finance it's really as simple as that so I can show you a quick view here this is the kind of stuff that we're delivering into our board so we've got the bar chart on the far left the one with the orange in the blue this is the number of vacancies that we've got in the average time that those vacancies have been open so we can very quickly identify there's a lot in sales but actually this bot here is super high so what's happening there are we got some wrecks on hold are we doing something different is the spec changing are we trying to refined Yeti fur coats I don't know but there is something that it's really driving that average duration much much higher on the other side we've got these thoughts right at the bottom so these guys are super proactive I know that because having come from recruitment I know that those departments are super proactive and they keep their average duration really really low we've got a view of the candidate pipeline we've got wreck aging data we know where the wrecks are open by region and by type whether they're incremental or replacements and then the kind of hiring statistics to date the really nice thing about using this for people analytics is that it doesn't stop there we can now use that same data to really explore what's going on so maybe this grouping by executive leader is is a bit granular actually I just want to see it by a different view so this is a higher level view General Administration sales cost the services which is our Consulting Group R&D and marketing and now that now your dots have moved around and you you're beginning to see different trending from the view before or if you really wanted to and I used to do this a lot in my previous role is that we can look at actually individual department level recruiting statistics so how many wrecks are open and how long is that department taking to fill these wrecks if we just choose some of those of course all that all the other visualizations change so that we can start exploring so now I've chosen these specific areas the pipeline has adjusted so that we can see how many candidates are in here we can understand the aging at the bottom and where those roles are in their location and that app takes three minutes to update once you put the reports in and reload the data through so this is where the time saved really plays a part now who wouldn't want to reduce their time spent in reporting especially to the board you know it's always one of those things that gets in the way conscious of time I'm going to swipe through to that one this one emergency response so this this is actually one of my most fun apps and it always we were talking earlier and I thought came through my head about the need for information and the desire for information and the way that just I wasn't on the panel so can answer the question to that one but if you get asked and the question leads with it would be interesting to leave it on the desk and because it would be interesting is one of those leading questions that says yeah I kind of think I know what I want and I just want to check it out first so that is the sure way path to spending a lot of time trying to understand what the what the consumer what your stake holder is looking for in the first place but they come to you and say I need to have that is a different request you can be pretty certain that that's going to be a much more robust query the emergency response app was born out of a desire from the North American business partner team to be able to contact our employees at the time we were going through a lot of severe weather stuff tornadoes hurricanes freezing winds whatever else and we just needed to be able to check that our employees were ok without having to go into work day and look up every single record so what they basically wanted was an employee directory so that they could reach out to and be in touch with our employees this was the solution that we came up with but it's had a couple of interesting side effects or on downstream advantages as well so what you see here is a map of the world this is now mapping through G analytics our employee population at zip code postcode level and what we're able to do now is to understand who's an employee who's a contractor but one of the side effects of this was being able to look at the data quality so one of the most frustrating things that we were finding is we've run a report and a workday and say here is a report of all your employees that you want to contact but there's a bunch of missing data and they would come back and say but this date is in workday and this is because in workday there are about 12 different fields that you might populate telephone number in and and that report that I was providing didn't capture all of them so what we've done here is we've pulled out and aggregated all of our data now I know exactly how many pieces of data are missing what does that do it means that if I go through and look up in Belgium say I need to contact everybody who is in Belgium I can now with some certainty say these are the items of information that I believe are missing from word day this is where we need to clean up our data now I'm not spending my time trawling through work they trying to find something that actually I know isn't there so it's an interesting side effect from just needing to understand where people are – being able to actually increase and measure the increase in our data quality so coming back to the emergency response and this is where the multiple data connectors or the multiple data sources come in this is what we wanted to get to here is a map of the world and at the time that I refreshed the data in this app I know that there were some severe weather events occurring in America and what the app does is it now plots where the information is coming from so this is coming from the North American severe weather warning service nice catchy short name and plotting it directly into our app here and it's now telling us who is affected one employee potentially affected by this one if I though go to if I just choose all of them probably be besting it's now going to show us all the difference weather warnings when the internet speed catches up and tell us who or which employees are affected here we go and it we'll serve up for us as well their contact details so their phone number their managers phone number their spouses phone or the emergency contact phone number their address I can't show it because it's a live app when there's a tornado warning or a hurricane warning it actually plots the path over the map so now we can go to our employees and say and she we see that this is coming away please could you check in with us every 3-4 hours make sure you're okay so another another great use case that's not bar charts and pie charts a final thing I'm going to talk about really quickly because I'm conscious of time is predictive analytics we talked again about it in the panel this is one of the predictive analytics projects that we have been working on and it is very much driven from the fact that every time we get a new starter IT complaint the manager doesn't submit the ticket in time for us to order the equipment and then we need to ship it out and the worker starts and mave there arrived and they've got nothing they can't login strikes me that we've been recruiting that role for six weeks how is it that we don't know that that person is arriving so what we've done is we've we've built a black magic box that processes a bunch of data that comes out of workday from the position that we open job category job banding job department and job location each of those gives us a set amount of parameters now with an about an 80% accuracy I can now predict what distribution lists and security groups that person's going to need when they arrive what hardware and accessories they're going to need day one what software licenses what system accessing configuration they're going to need when they arrive and that even goes into the country location so if we're recruiting in Sweden that person's going to want to have a Swedish keyboard as opposed to somebody in the US who's going to want to have a u.s. keyboard so even even to that level we can be predictive with these are the people that we're recruiting today and this is the equipment that they're going to need so that the only thing that needs to happen when we hire that candidate is we plug in the name and we create the user account we install it on the equipment and you are ready with your box the minute that that manager come turns up and says right I've got a new high starting is all the equipment is all done we can even email the manager and say we know you're hiring here is everything that we're going to prepare please let us know if you think any differently so that is that is the predictive analytics piece that that we've been working on and that in itself we reckon is going to save weeks and weeks of IT time every year thank you very much any questions if we have time

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *