May 31st 2019
CRFCast – HR Insights from the Corporate Research Forum: Agile HR: Sky UK Case Study with Tracey Waters
Tracey Waters, Head of People Engagement and Development at Sky UK, discusses how her team have deployed agile development techniques to transform their working methods and increase the business impact of learning and development. Tracey describes four behaviours that underpin Sky’s approach: being collaborative, iterative, data-driven and person-centric. We discuss some practical lessons learned from implementing agile.
Gillian Pillans [00:00:03] You're listening to the CRF cast where we explore research we've been working on here at the Corporate Research Forum and we discuss the latest thinking and strategic HR topics with academics, practitioners and leading experts in the field. I'm Gillian Pillans, research director at CRF. You can explore the full podcast archive and subscribe for updates by searching CRF cast on iTunes or Spotify. You can also find more about CRF and access all our podcasts and other research materials at www.crforum.co.uk.
[00:00:48] In today's podcast, I'm talking to Tracey Waters, head of People Engagement and development at Sky UK, about how she and her team have used agile development to radically change the way they work and really shift the impact of learning and development within the business.
[00:01:06] Hi, Tracey, and thanks very much for talking to us today.
Tracey Waters [00:01:08] Thanks, Gillian. Great to be here.
Gillian Pillans [00:01:09] I thought it'd be good to start by just getting some definitions out of the way. So, you know, I think when we talk about Agile, that can mean lots of different things to lots of different people.
Tracey Waters [00:01:19] What's critical to us is the principles of agile. And they if I summarise them for us, they are being highly collaborative. That's actually getting a team of people to work together on a single problem and doing that. And quite an intensive way, actually. Secondly, it's about being iterative. So actually testing and learning along the way, trying things early and quickly finding out what works and then doing more of what works and less of what doesn't. Thirdly, being more data driven. So how can we get data that tells us how people are actually using what we're providing? And that's quite different and much more powerful than opinion or speculation. And the last one is being quite person centric or human centric or in the case of Agile, our employees centric, which means really understanding your kind of end user, the person that you're actually in service of and those four things. So being collaborative, iterative, data driven and person centric are really central to the way that we approach agile in our team.
Gillian Pillans [00:02:27] We'll go down into some of those a bit more detail in a minute, but I'd like to write books and start and the business case for this, the business drivers for what was in effect a complete change of approach to the way you and you tackle your to do list. Can you share a little bit about. What were they part of the business drivers behind this and how did you get to that point where taking an agile approach to your work was the right thing to do?
Tracey Waters [00:02:52] So we'd find the fact then to the beginning of 2016 and said this wasn't something that we were asked to do. It was it was that we were facing against some problems that weren't going to go away by doing the same thing over and over can or even slightly differently over and over again. We really had to think quite radically. And that was because I think like many businesses and many, many HR teams, their request to do more with less comes from the business all the time. But it's gathering momentum. It's definitely more and more with less and less so that that was ever present. And Sky's very good at operating efficiency. So so that wasn't going away. We also had a clear line of sight in terms of the strategy and there was a big need to build capability and actually make sure that our people were able to deliver their best work. So when you kind of looked at just even those things and then you look at the way that in that case LRAD works. But I would say HR generally the way that HR in itself is set up is very waterfall. So if you do anything in software development and look up software development waterfall, it basically talks about these handoffs from one team to another. And it's about in H.R. it's defined and which is usually business, HR design, which is usually the centres of expertise, and then it's delivery, which is usually operation. That's kind of how HR is structured, but it's actually the opposite of Agile. And and what we found is by being designed in that way, defined design deliver, we were quite slow. We were quite slow. And we were also getting requests for, if you like, kind of orders of what the business needed that was being defined from an HR business partner. We would then have to interpret that and design what we thought it meant and then we would ask someone else to deliver it. And often you would find you were too slow. Sometimes you didn't meet the need at all because it got lost in translation. We also had problems with opinions, people going, this is what I think the problem is. And there was a fair and real lack of data and lots of waste. And what I mean by waste is we had a really amazing actually three day leadership program, which everyone laughed. But six months later, I would go back and ask those leaders, what was the thing that made the impact? And sometimes I was lucky if they could remember one morning session out of three days. So that's why that's two and half days of waste and we couldn't sustain that anymore.
Gillian Pillans [00:05:38] So when you started implementing agile, you kind of did it in an experimental way. Is that right?
[00:05:46] So, yeah, I think you decided to learn, test and learn underpins the whole the whole approach.
[00:05:52] So you just kind of decided to run an experiment, is that right? Yeah. Yeah.
Tracey Waters [00:05:57] It's we actually going from the kind of the problems that I was just mentioning to we think agile is the answer actually took us about six months and we only really came across agile when we were really then also trying to understand digital because as soon as you start to understand digital, you hit software development, as soon as you hit software development, agile and then you're in that world. And so once we kind of thought there's something in this, there's this is a different way of working. And if you can also step back and realise that this isn't about software development, this is about how groups of people work more effectively and efficiently together to produce outcomes that are of high value. You can kind of get beyond the software. There's there's a lot within it.
[00:06:46] But it's a massive change, so to just kind of announce to my team or to the business or to H.R. that this is what we going to do. I knew that wouldn't work. So I took an approach that that required more. I could change management approach. Obviously, I sought out a kind of senior sponsor as as you would do as any responsible kind of change manager. And that was my H.R. director. He's really open minded. And she was just like, yes, we need to Daphne, try something different, go for it. Then I spoke with my team. I spoke to them individually. I said, I think we've got some problems we need to solve. And they we all agreed on what the problems were. Which is a really great place to start. So the why was really clear. I then want the team up. Right. We had a couple of hackathon which even using the language of kind of software development and that was we were trying to crack a problem in a day. I just got people thinking really differently. And then I made them a promise, which was let's try working in a different way for 90 days. And if it doesn't work and if you all hate it, if it's unsuccessful, I promise we will go back to the way that we were.
Gillian Pillans [00:08:00] So at the start of those 90 days, how did you determine that you were going to be able to make a decision at the end of that period as to whether we rate it saying this is worth what?
Tracey Waters [00:08:13] What I was confident of is that, you know, three months actually seems like a short amount of time. But it's a long amount of time when you are really experimenting and you're doing these experiments. It's clear hypotheses and you're paying attention to the kind of inputs and outputs. So we actually knew after two weeks that this really had potential. And then the data came out quickly. Okay. So help tell us about the data.
Gillian Pillans [00:08:42] What what sort of data came quickly and what data backed up your hunch?
Tracey Waters [00:08:48] So there's the data came from a couple of sources. One is we had also made considered choices about which digital solutions we were going to trial. And we made sure the the absolute minimum requirement was that it had a good user experience, which we actually tested in our own. We have a UX lab which is normally used for our customers, but we decided to use it for our internal customers. And they also had self serve real time analytics, which meant that the team could actually be the masters of their own destiny and see what was working and what wasn't working. So the data came in very quick. Once you've got those kind of tools, so not an elements. Once you've got a a true digital platform, you can really engage with people very differently and you can run experiments and AP test and you can do all sorts of things. The other thing was so we do our engagement survey twice a year. The scores in terms of empowerment, feedback, recognition, role, purpose, some of them went up 20 points. Now statistically significant shift is 3. So you've got 20 meant that we had really done something quite different.
[00:10:02] So I had a team that was now highly engaged and motivated and we were now able to understand what people were using, not using, who was engaging, not engaging. And then we were able to understand our kind of end user groups, whether they're colleagues or managers or leaders in a much more segmented way, instead of just thinking about all employees or all managers, which means that you're never going to be able to do anything that feels personal or relevant.
Gillian Pillans [00:10:34] OK, so can you walk us through an example of a learning product you launched using agile methods? How did you determine the business needs and how did you use data to determine the end product was what the business actually needed?
Tracey Waters [00:10:46] What I often recommend to people is if you're going to try something, if you're going to experiment, you don't have to do it like we did it, which was whole scale 90 days. We're doing everything differently. You can just pick a kind of very targeted experiment and just try. That way we had kind of just made the decision that we were we were going to try this agile way and the very real business problem came along. And that's that's like what you want, right? It's happening now. The business needs help now. And we'll be able to sort of see if it's made a difference very quickly. And that was very simply that was Sky. It needed to actually, unfortunately, let a number of people off in a very short space of time, which is never great. And it's hard for the business to the best of times. And they kind of said, is there anything you can do with what some managers that are going to find this really stressful? Some of them never done it before. So what we do now where we're asking. Do something about two weeks later, and under most circumstances you'd either have to. I'd say no. No, we can't do something as quickly as that. You need to give us more notice or you might wit and leave something e-learning off the shelf. Or you might try and do some kind of webinars. You know, you try and do something. But we were like, we think we can do this differently. And we actually said so out of the kind of one hundred and fifty managers that are affected that are going to need to have these difficult conversations. Are there any that you think, a particularly high risk block that you're really worried about? So already we were going we're not going to go after all of them. What's what? Which are the ones that are kind of we can add the most value to? And they weren't actually if these 50 that we think in these particular parts of the business. Great. Okay. So then we went, what is our kind of minimum valuable product or MVP? Again, a software language. But it's what is it that we could design in this space of time that is shipping all we can basically bits done and live. And it ended up being again, it's really simple. We had we went, let's go with five resources, digital resources that people can access on their mobile. Because if you're about to go into a difficult chat, you don't want to log into the elements and let's run some like really short, small group sessions. And we weren't sure at the time length was and we actually experimented with 60 minutes, 90 minutes, two hours and 90 minutes was the winner in the end. And that was what we went after. Let's see if we can reduce the risk here. If we had, I forget, nine days, nine working days, we're going to produce these resources, give people access to the app. We're going to schedule these sessions one before and one during. And that's all we get. That's all we're gonna do. And that's what we did. So the challenge there is you've got to be able to write resources that are helpful to people. So you have to test that before you go. Life is helpful. It's too long. It's full of jargon. And so we did that and it was from tip him that of the two week. Can you do something? We literally ran it for two weeks and then we were measuring how many people access the resources which we could do because it had self serve analytics. We had people attend the 90 minute workshops, which we had about a 93 percent attendance rate. And then we looked. We surveyed afterwards. How useful was it? Did you feel better equipped? How did the conversation go? And those scores were really high as well. And then no E.R. cases, which that which was what the business was worried about. So it all went as smoothly as it could have gone. And we feel we had a small part can play now and then that gave us more confidence to then go, how can we help more managers, more points of need, bigger, more often.
Gillian Pillans [00:14:46] Coming up in the second half of the CRF cast, I'll be discussing with Tracey how the Sky Learning Team uses data to understand business needs and to work out how to prioritise resources. We'll also talk about how they use data to evaluate the impact on business outcomes. And they will finish off with some lessons learned about how to rule out agile ways of working untraceable share some of her top tips for success.
[00:15:16] If you're enjoying the CRF cast, please subscribe to our channel by searching for the Corporate Research Forum on iTunes or Spotify. You can sign up to get the latest editions automatically and to share with your colleagues and leave us a review.
[00:15:33] How about that sort of day to day? So how do you on an ongoing basis understand what's really going to make biggest impact to the population that we are supporting and then use that as the driver for designing the products and tools that they need?
Tracey Waters [00:15:52] So as a couple of depending on the kind of, I guess, end user or customer you're talking about. So if it's leaders, then you really are taking this deal from the strategy and you're going back to the senior team and going based on what you've got here. These are kind of the five things we could go after. We recommend going after no more than two. Can you prioritize the two you want to go after and then going after those things when you get to kind of managers, which are often the kind of forgotten middle of the organization because they're quite up there. So many of them at Sky, we have close to 3000. You roll out anything that traditional, it is going to take years, which is the thing I keep saying to people that we have built manager development iteratively over the last 18 months, which may sound like took 18 months. But just think how long it will take him to roll one thing out to all of those managers. It would have taken longer than 18 months. And what we've got now is less of a Titanic and more of a speedboat that we can deploy with managers. What we stuck to was. So managers are really busy, right. They get loads of demands and messages and comms and things that they have to do. As we've shifted to it's the manager's responsibility as a manager. And so we really then thought about what are the points of need that managers have? What's going on in their day? Where do they get stressed? Where are their pinch points? What causes them anxiety? And let's try and make those easier, which ultimately then makes employees happier, better, etc. So those are we. We haven't deviated much from what you might consider a typical life cycle. I have to recruit someone I have on onboard them all the way through to feedback, performance, development, exit. So we thought actually those are workflow events. And the thing that's different about that is what you would often get from HR is managers need to be better at difficult conversations. And then you realize that there are so many contexts that you have. So what do we actually rather than go after difficult conversations? Look at where in the lifecycle in the workflow, they might have to have a difficult conversation. And you find that there are a lot. And say what you're building in rather than difficult conversations being the primary driver. It's how do we support people within those points of need? And if difficult conversations is one of them, then let's build it from there. If you don't have context, people are practicing something hypothetical and that's not helpful.
Gillian Pillans [00:18:29] So how do you work copes where managers are in that life cycle and who to prioritize? As you say, you've got 3000 people in that population and you can't do everything for all of them yet at the same time. So and yeah, I think you use data, is that right? Yeah.
Tracey Waters [00:18:48] And there are and we have a couple of sources of data because one is obviously our HR system.
[00:18:53] So we can obviously see when a manager has raised a red, you know, recruitment wreck. We can see when they've got a new person about to join their team. You know, a month out, two weeks out, we can see when they've got someone who's leaving.
[00:19:10] So that data is available. And the ultimate goal is to make it automated. So it is a lot of systems that will send automatic emails based on triggers in the system and we hijack those and change the wording and then point people to a resource or a mini workshop that's that's directly relevant to what they're what they're about to do or deal with. The other thing is we have our people survey data, which allows us also to segment. Now, we don't go after low scoring managers. I'll be honest, we tried and it was very unsuccessful. And when you look back, you go, well, of course, it was unsuccessful because people don't like being targeted to hang out with other losers. And what they want to do is they most of them do want to get better, but they'd rather learn from people who are already quite good. So that was a good learning between those two data sources. We can segment managers and we can also automate things. And the other thing we go after is we look for senior leaders who, based on their people, survey scores are going on, going after growth, and that's all based on data.
[00:20:22] And that has been a really key shift over the last two years. And it means we can prioritise and also work with the people who are most interested.
Gillian Pillans [00:20:33] So how about evaluation the historic team? And that's always very difficult in the end, a development that basically means, as you know.
Tracey Waters [00:20:41] Yeah.
Gillian Pillans [00:20:43] So, I guess for me, what's really important is understanding the business impact of the interventions that you've invested in. So how are you tackling that?
Tracey Waters [00:20:55] Where we started was happy sheets and not your impact. Assess? Absolutely no. But yet we still do them over and over again. What you're able to do, certainly at a minimum when you use digital platforms with self-service analytics, is that you can definitely look at sort of three things reach, engagement and consumption. And that works like as where TV companies. So actually those metrics worked well for our leaders and it also skills are received as the court. So it's all with perfect. So this is meant to be what reach means to us. And this doesn't mean it's right. But what it means to us is we have two thousand eight hundred managers. How many of them have logged into our app? So how many if we kind of got to voluntarily kind of put themselves into this into this platform? And how do we keep growing that month after month? Who haven't we got? Who have we got? So, again, you can start to segment even your kind of marketing engagement is. And how many of those have we seen in the last month? Now, that's a really important metric, because if they come in and see something once and never come back, then you didn't give them what they were looking for. They weren't impressed. And so we need to we need to understand that more. And we want to make sure that we've got 25 percent of managers. Is our target returning each month to learn more about being a better manager? And the last one is consumption, which in some platforms is amount of time that they've been watching or kind of consuming videos, for example, or how many resources they actually kind of binged on while they were while they were in the app. And that's our starting point, because then you can start to get a sense of is what we're designing valuable? To triangulate more of our data. So we now have enough learning data, that means that we can go let's have a look at people.
[00:22:51] So our kind of people, HRIS, data, location, tenure, where that how many direct reports they have. Even moves, etc.. Let's look at engagement survey data so we get a better sense of which managers seem to be creating cultures or kind of environments that people enjoy working in and feel like they're excelling. Let's have a look at whether the learning that they're engaging with or the type of learner that they are is a predictor of anything. And as you start to look at that, I think we can get to an even better place of what a great manager looks like and we can start to even tailor it even more to be about Sky. And it's the context, the richness of the context that actually we find the most valuable.
Gillian Pillans [00:23:39] One of the first things you said at the start of this conversation was that this this is a change management process. What did it take to get the business to buy in to this? Were there any barriers that you encountered along the way? And then how did you overcome those?
Tracey Waters [00:23:55] We weren't just changing our kind of way of working. So in some respects, if we changed it but kept all the rules the same, then they probably wouldn't have noticed. But what they noticed is that we were also breaking all the rules. So we really moved away from programs. We were moving away really from classroom training. The idea of a kind of day to days, the idea of a 12 month program. The idea of six modules, etc.. We were breaking a lot of those rules cohorts. We were kind of just if you list all the things in Allende and just kind of break them. And so they were noticing that. And digital we had we were introducing digital solutions not only, but they were having to now engage with content in a different way. I noticed that. And the thing that was helpful is that we were also on a digital transformation drive in the companies, so they were receptive to that idea. They also operating efficiency is very alive and well in skies. So there was an appreciation that we can't just keep doing what we've always done. That has helped. But I would say what what was unexpected. If I had my time again is that I did underestimate. I put a lot of effort into the team and less effort on kind of stakeholder management. I did not put effort, but I would have got a lot more. And I would have, for example, done a lot more road shows where I would have gone and spoken to business town halls and kind of gone to leadership meetings and that we've taken up a lot of my time, but it would have been worth it explaining to people what we were like, what we were daring to do and why and how they could get on board and how we could help them.
Gillian Pillans [00:25:42] If you were advising other people who are perhaps thinking about following a similar journey to you, are there particular pitfalls to avoid?
Tracey Waters [00:25:53] I think people try and agile because it seems to be the latest fad or someone said you should really go agile like me. You really need to kind of have a good reason for going agile.
[00:26:06] But if you scrum and camp like software development and ask Agile, you need to be really clear on what problems it's going to solve. Ideally, you need to really be on a digital journey as well, because again, if you don't start getting different data other than tracking completions and sign ups, then you're not going to be able to make smart decisions that allow you to pivot and kind of test and learn.
[00:26:30] I think another pitfall to avoid is people thinking too big. I need to do my whole team or my work all at once now.
[00:26:40] And the I guess the smarter way to do it is to run some experiments, to try some different things. It's like when I speak about the kind of manager example, you know, it is an incredibly small example, but it's an example of how you can be responsive and quick and solve a real problem. if you are approaching it in a different way.
Gillian Pillans [00:26:59] One of the things that perhaps helped you along the way was learning from teams within the business who were already further ahead than you.
Tracey Waters [00:27:10] Yeah. What particular software development?
Gillian Pillans [00:27:11] Yes, exactly.
Tracey Waters [00:27:13] So we we approached them and they were they were surprised and delighted to find HR coming to talk to them about how they worked and wanting to learn from them. They were incredibly welcoming. And we had some of the agile coaches come and spend time with us. And we've watched a number of their kind of ceremonies, as they call them, the stand ups and planning and retrofits.
[00:27:34] And that was an enormously useful because we we did self teach, so we didn't go on any training courses. We kind of learn as we went and they really helped us. And then since then, there's a number of teams like our internal comms team who have come to us and then said, what are you what are you doing?
[00:27:53] And so we feel like we're kind of organically growing a bit more agile in Sky. That's beyond software development.
Gillian Pillans [00:28:00] Okay. Looking back, how would you estimate the impacts on the business of this shift in how you approach the whole development space?
Tracey Waters [00:28:13] So I'd say this shift is more evident now because where we start, we started with L &D, as you can sort of tell from my title, it's kind of engagement and development. And that's because I am now responsible for a number of different, if you like, employee products ranging from kind of recognition, career performance development. And we because of the way we work and where we don't have individual people responsible for individual things. And I think when you start to see for me, when I start to see someone in people's squad that's working on recognition, talking to someone in manager squad who's then thinking about how it links into manager development, who's doing something, a manager development, and then going to people's squad and asking, have you got anything that can support this part of the lifecycle?
[00:28:57] But they're doing that in a very fluid way because actually can the way it all connects for the around an experience is is where I think we're starting to get some real impact. Employees are not experiencing all these separate things that are coming. Separate points in time because it's convenient for the person to launch it. Then they're really starting to think holistically about the user experience. So any any final recommendations that you would make for others who are listening into this conversation, starting to think about sort of test and learn mentality and the thing that my team had the hardest letting go of experience was they were used to producing beautiful.
[00:29:42] Polished, complete products that took months and months and months to design. And now they produce something a bit less polished and a bit less complete. And they test and learn and the number of times we have been wrong. But we thought this was what people would love in their life now. I mean, we've been wrong more times in the last two years than we have in the last, you know, 20. But of course, we've been wrong the whole time. It's just that we didn't know it. So there's really something about testing and learning and embracing experimentation and and doing that in small ways that I think will give people more confidence to understand what this is before they do something radical.
Gillian Pillans [00:30:26] Tracey, thank you very much for taking the time to talk us through your story today.
[00:30:34] Before we finish, let's take a moment to recap the key messages from my conversation with Tracey. First, use data to segment your customer base and really focus your resources where they're likely to have greatest business impact. In Tracy's example, where her team had to support managers through a redundancy program, they minimized business risk by targeting those managers who most needed support or who were least experienced in handling these types of situations. Second, treat the transition to agile ways of working in the same way you would handle any other change program. Pay attention not only to supporting the team through the transition, but also to getting and keeping stakeholders onside. Put more effort than you think you're going to need and to stakeholder management. Third, be clear about the business problem you're trying to solve. By adopting Agile, otherwise you're just in danger of following yet another management fad. Finally, if you'd like to hear more about Tracey's story, you can read her blog, which is full of warts and all tales of her experience and what she and the team learned on the way.
[00:31:40] Find it by googling: Medium, Agile in Learning.
[00:31:48] You've been listening to the series cast with me, Gillian Pillans, research director at CRF. You can find out more on our website at www.crforum.co.uk. Follow us on Twitter @C_R_Forum or join the CRF group on LinkedIn. Bye for now and thanks for listening.
You May Also be Interested in
July 24th 2020
Post Meeting Notes - The Agile Dashboard
On 24th July CRF and PARC members gathered for The Agile Dashboard: Designing Agility Into Your Company. “If the keyword for the 20th century was ‘speed’, the keyword for the 21st century is ‘agility’.” So said Professor Joe Perfetti, our speaker for this session. That was before the coronavirus outbreak re-wrote the rules and re-ordered the business world. His statement is now more apposite than ever. Entire sectors have been devastated. Business leaders are now surveying the world after the lockdown with no idea of what business-as-usual will look like or even whether the very idea has now become outdated. Of one thing we can be sure, though, businesses will have to adapt and change fast.Read Post Meeting Review
July 22nd 2020
The Agile Dashboard - Joseph Perfetti Slides
Joseph Perfetti discuss the forward-looking metrics - speed, interaction time, and pivot - that are key to agile companies' success.These are his slides from the Zoom Interactive Event: The Agile Dashboard which took place on 22nd July 2020View Presentation
October 11th 2019
IMD Business School - Unlocking Agility through Purpose and Impact
Agile organisations are both stable and dynamic. Purpose brings stability, a critical attribute to flexibility. Purpose enables company-wide capability to pivot away from static siloed business models into new sources of value. Currently, transformation has become the status quo. From digital, to customer-centric or sustainable business models, change is the order of the day. Agile businesses react quickly when challenged and can transform and adapt. Purpose instils leadership clarity and material strategic priorities that drive business to act.View Presentation
October 11th 2019
RHR International- The Greatest Leadership challenge
The Greatest Leadership Challenge - Impacts on how we assess and develop talent Finding those critical few leaders who can set the course for the future and galvanise an organisation to deliver is the single greatest leadership challenge facing businesses today. There is not one organisation that isn’t feeling the pressure from growing market complexity, technological disruption, and an evolving workforce. It is obvious that having the right leadership is critical if organisations are to successfully weather the storm, but what is the real problem that we are trying to solve? Are we defining the problem correctly and then using the right mechanisms to find the answer?View Presentation
October 11th 2019
Mercer - Agility in Action
At Mercer’s masterclass, delegates learned about how organisations have increased their agility to improve speed, nimbleness and adaptability in order to create a competitive advantage. Delegates were encouraged to answer questions such as how agile is your organisation? How agile do you need to be? And, how can you increase your own personal agility?Read Article
June 13th 2019
Digital Culture: How Technology is Changing the Way we Communicate, Work & Behave Presentation
In this presentation, Rahaf Harfoush explores the impacts of the digital age, taken from CRF's 2019 Summer Lecture. What are the five forces which influence our behaviour in relation to digital culture? Are we holding ourselves accountable for the impacts of our technology?Read Event Document