 Rwy'n gweithio, wrth gwrs. I'm Tim Deeson, and I'm here today to talk about creating a collaborative agency culture that scales. I wrote that title about 10 minutes before the submission deadline, and I didn't really know what I was going to say at the time. As I wrote it, I realised that what I was writing was the story of our groan pains as we've grown and developed as an agency and increased in size. I think what I found was writing the story of what's worked, what hasn't worked, and what we still don't understand about what we're going to do next. I've deliberately left some time for questions at the end, because some of this stuff to me is most interesting in discussion and in the detail rather than the big theory, because I definitely don't have the magic answer of how you do this, but I can say what's worked for us and what hasn't worked for us. In that same time of writing the story, I realised a lot of that was also my own personal development, learning how to get out of the way, learning how to work in a very small team to a much bigger team. I didn't really find there was a right way or a wrong way to do that, or a way that I would definitely recommend that everyone should do it, but I thought there was definitely some lessons and some kind of findings that I thought were worth sharing. It's a bit of background about who we are, because I think it's important to understand the size that we are now, the size of where we came from to understand the context of what I'm saying. So today we're just over 30 people. I think we've probably added a couple of people in the last month or so, and we're about doing about £2 million worth of Drupal build services, so we only have full-time permanent employees. We don't use contractors and freelancers that much, and we tend to do full service. We do a discovery and a production phase, and that means we have the UX, the design, the project management and the development all in-house. Over the past probably two or three years, we're growing about 30 or 40% year on year, but in the past our growth was much slower than that, which I'll talk about in a sec. We started back in 2001, which I realised made me about 20, which I think was probably responsible for some of our growing pains. I had a development background. I was working doing project management, the new business stuff, and initially just working with a designer. The two of us were all together in one room. We didn't really need lots of documentation and process. It was quite easy to collaborate. We didn't have a Google calendar where we'd put meetings down and things, because that would have been ridiculous. It was really, really efficient, so we spent lots of the time on the work. We didn't really have to spend lots of time planning how we were going to schedule things or how we were going to manage things. We won some small accounts and we started to grow and add a couple more people. We won things like the British Alpaca Society and helped them manage their herds as one of our early projects. We carried on growing. We picked up people like Robbie Williams about six years ago, Johnson & Johnson, some B2B enterprise work, people like ITV, who are one of the big media companies in the UK. By about 2012, we were about 16 people. We were doing about £1 million, probably a little bit over in turnover. On paper, things felt good. We had a very slow growth in some ways, but I think that was partly because we'd started a little bit accidentally. I don't think we meant to start an agency. I think we just started doing some work, so there wasn't really a big plan. We were still really, really efficient. There was minimal meeting still with those 16, 18 people, minimal unbillable times. We were spending all of our time on client work rather than on internal process stuff. Most of us probably only had one or two, maybe one even, unbillable roles. Pretty much everyone was focused on delivering projects every day. I think it was around that sort of time it started to be probably a bit painful in some ways. I think where it felt easy and fun, easy and fun, it didn't really seem to be true anymore, or quite as true. I was involved in most decisions, and I think that was through that organic growth that we were still very flat, or flat plus me in a way. Obviously, if you're involved in running an agency or working an agency, you know some of the sorts of tasks and issues that you're dealing with, and often some of them are very short-term focused. We need to do a pitch on Thursday. We need to do six days of pitch work in the next two days. We've also had a deadline for three months. We need to hear we're running our recruitment pipeline, we're running our sales pipeline. There's lots of conflicting pressures that I think you can be in an agency environment, and some of them require that long-term view, and some of them require a much shorter-term view. To me, scheduling is a... I've always hated scheduling, I think. It's a really complicated, annoying thing because you've got so many different dependencies between clients signing things off, if they're late signing something off, then by the time you push back your front-end build, for example, then that team's waiting for work and knocking into another piece of work. There's just that. There's those really difficult to get around issues that also have a commercial impact, I guess. I think when you do it badly, it starts to get slow and difficult, and we were starting to get to the size where there wasn't a simple way to schedule those teams anymore. Also, holding the major client relationships, which when we were dealing with someone like the British Alpaca Society, they had a relatively unsophisticated digital requirements, but as we started to pick up those bigger clients, what they wanted from the relationship with us was much more complex. I think they wanted to be able to be looking at their kind of three-year plans with us. They wanted to be really spending a lot more time and kind of engaging with us, and we hadn't really built a way for the team to be working on projects that weren't defined quite well. I think that was starting to leave me with a feeling that my head was starting to hurt, that it wasn't... it didn't really fit in there any more, I guess. That was probably an expression by Friday afternoon that was fairly common. Aside from me, I became more conscious that I think we were definitely wasting people in the team's time and probably their motivation. I see those as two different things. You can waste people's time in the sense that they don't have something useful to do, but I think when you waste their motivation, when you waste their time a lot, you start to waste their motivation, and that kind of teaches them that it's not useful to be looking for the right things to do because there's always going to be an inefficiency or a blocker in stopping them doing that. I think that can start to kill that kind of spark and enthusiasm when you're always kind of thinking, well, should I bother going to find out what's going on with this? Maybe not, actually, because I don't think I'll be able to resolve it. I kind of realised, I think, that it was effectively I was micromanaging people, and I don't think it was an explicit, deliberate kind of thing that I did. I think it was kind of an accidental organic growth, and these were smart, talented people that we'd worked really hard to recruit, that we respected their opinion, we would put them in these roles for a reason, but I think we were struggling to then actually make sure we used the capacity they had and the skills and the ability. I think it was that point we kind of felt really stuck in a way because we'd had this kind of success and growth working our way through things, but we didn't really feel kind of, it didn't feel like more of this was possible because if we kept growing it felt like these kind of problems would get kind of worse, I guess, not better, and I'd always really been, I think, excited about our future. I obviously started the company and kind of driven it along, but I thought, well, actually, I'm not really sure what we'd have more of if we carried on doing this, and growth, you know, there wasn't really possible to plan for growth, I think, at that point because it wasn't clear how would we do that. And so we kind of started to, I mean, I started thinking there's obviously something kind of we're going to need to change to have, you know, a bigger kind of brighter future and trying to kind of understand the background of what was there. I think there was obviously the fact I was a bottleneck, we kind of, what had been really efficient in a small team was starting to kind of cause a problem. And looking at the root causes of that, I think we didn't have the right data. So we didn't have systems that were giving us good information about what was going on at the time. And that's both, I think, retrospectively and kind of forecasting that. We didn't have the right structure of people, so we didn't have the kind of the roles that where people could make the right decisions. And we didn't have the kind of process point, so we didn't have a place where there was a natural place for that person to make the decision at the time. And up until then, kind of everything had been in my head, which had been efficient in the sense that I could kind of map it out, but it wasn't in my head anymore because we were too big. And it was really clear that we had these people with skills and knowledge and expertise that was greater than mine in probably most areas, if not all. And we were kind of wasting that opportunity because they could have made these better decisions more quickly than I could, but we weren't kind of giving them that opportunity. And that was really because of that kind of slow, centralised model that had worked at a small size, but now really wasn't working at a larger size. And that was obviously frustrating. I didn't really feel that I would be doing, I was doing my job as well anymore. It certainly wasn't as much fun. And I think it was probably frustrating for everyone else, particularly the people who really felt that we could do something differently or move more quickly. And I think that was kind of, it became clear there was kind of fundamental changes we needed to make. The engagement, restoring that kind of energy and drive that we'd had to make sure that we could keep pushing on into the future and finding a new operational model, finding a new way that we could keep working that was going to have to probably be a complete replacement for the old one. And I think often the challenge with kind of organisational change is you're still delivering projects. So you still have these commitments, you still have these deadlines, you still have recruitment, you still have sales, you have all these things going on. It's kind of changing the, you know, the tyre of the car going down the motorway type problem where you're trying to kind of both, you both need to think the distance to be able to have the insight to see what you need to change, but you also need the actual time to implement those changes as well. And I think that can be, that can be something that can be quite difficult to find. And so we started to kind of think about some kind of quick wind solutions, you know, what could we do? And one of them was, well, you know, management has been around a long time. It's not kind of a very new concept. So, you know, could we hire managers? Is that what we were kind of lacking? And on paper, I think obviously, yes, you can have a management hierarchy, you can have delegation. But I kind of felt that people knew what the right thing to do was. I didn't really feel that we needed to add people to tell these people what to do. I think I felt the problem was that people weren't really empowered. They didn't have the right information to kind of hiring people in to tell people to do the thing that they already wanted to do in the first place. Seemed a bit kind of expensive and weird. And I think, you know, there's kind of, it felt like there was a software option, I guess, as well. So there's kind of those agency management tools that do everything from your sales pipeline to your bug tracking, to your resourcing, to your forecasting. And I didn't really know anyone who'd had a good experience with them. I didn't really believe in them, actually, to be honest. But also I think they would have just given us more information and we didn't really have a model to make decisions with that information in, so it wouldn't have actually really fundamentally addressed the problem. And I think, you know, what we really needed was both kind of better tools. We needed a way to have this information, but we also needed a way to have a new model to actually make use of that information. And then I thought, well, again, you know, I don't think this is the first time, that agencies aren't new, digital agencies aren't new. This isn't the kind of the first time that anyone else has been here before. People have grown and outgrown their kind of model before. And of course, you know, yeah, they have. This question is much older than digital agencies, much older than agencies. There's a guy called Ricardo Semla in the 70s in Brazil. And he, I think he had about, I think it was tens of millions or low hundreds of millions of turnover. And he was a, he owned an industrial manufacturing company. So they produced, you know, industrial dishwashers, industrial washing machines, et cetera. And he's kind of saw huge inefficiencies in bureaucracy in the initial organisation. And one day, he just decided to sack about 60% of the management. And he thought, well, we're going to have some problems if we do that. But actually, I'd rather have the problems that come with this than I would have about having one of these managers who kind of get in the way. So he created this system where the teams could hire each other, or within the team, they could hire each other, fire each other, set their own pay, choose their own suppliers, choose their own product lines and their own prices. So he really had decided that those managers weren't kind of adding value anymore. And that was something that would be far better done by the people who were delivering the work day to day. And that felt, I think, really intriguing. He'd written about it in this book called Maverick in the 80s. And he'd gone through a kind of a through iterations with it and had some success. Although he'd changed some bits of it quite fundamentally later on. And it felt, I think, like the right direction, but it also felt like a huge leap. It didn't feel that it was something that we kind of had the confidence to say, well, if this worked for a fridge manufacturer in Brazil in the 80s, it would definitely work for a group agency in the UK. I mean, one's a product business, one's a service business. There's quite a few things there that I didn't really feel that this was kind of an obvious thing to just jump in and do. There's some kind of closer-to-home kind of interesting kind of writing and thoughts about this as base camp. They've written a lot about the kind of motivation, company culture, scaling beyond a founder. They've got a couple of books, Signal vs. Noise and Rework that are worth reading. They're quite good on the principles, I think. They're also a product business, so they're slightly harder to apply again directly to an agency, but they don't kind of waffle on too much either. They're quite kind of short and succinct, so they're worth a read if you just want to kind of get a conversation going at the office or try and explore some of those topics there. They're worth a look. There's also a start-up in the US called Buffer, and they've been documenting their journey on their blog about self-managing teams, so I think they've been doing it for a couple of years, and they've been quite honest. They've given some quite good write-ups about kind of what's worked and not worked. So they do things like, the team set their own salaries, they got rid of management appraisals, I think, so that you can each, the individuals kind of score each other and set salaries and things, and they've actually put back in place some of the things they took away, because they found that those, for example, mentoring really did have value, so not giving feedback to people and experiencing them, tell them how they were doing, they actually realised it was something that they did want to keep. And there's also a really interesting talk I saw yesterday, which I just slipped into the slides this morning. I think it's pronounced LEAP. They're a Swiss company. They're a Drupal agency, actually, and they're quite a lot further down the road with this than we are, and it was a really good talk called Teal is the New Orange, which you can find the recordings that hold as it's very at-structure. I think there are about 100 people and doing about eight or nine million Swiss francs worth of turnover, so they've managed to kind of scale quite a long way through that, and I definitely recommend having a look at that and talk if that's something you're interested in. And so kind of from that background reading, and unfortunately I didn't see the talk yesterday before, so I was working on bad information at the time, but on that background reading, on the themes that I thought we really should kind of need to apply, I think one of those was intrinsic motivation, so if someone's intrinsically motivated to do something, if it's something that they just enjoy doing and would kind of do if they weren't getting paid for it or if no one was standing over them making it do it, I think you can kind of start to understand your organisational culture based on if there's things that people just really hate doing and you've created an environment they hate doing in it, then management starts to become a cost of how do you monitor and force people to do? If you can find the right people and the right tasks and create the right environment, then you're probably not going to have that resistance of how many ways can we force people and how many ways can they try and get away with not doing something. I think you can kind of start to break away a lot from that is what I've found. And also I think as a company, it's something we've failed to do for a long time, I think, was probably setting clear goals, so really high level, probably quarterly annual and three year goals, and you can also do a 10 year goal if you really want to make sure you kind of have a really clear arc of direction that you're taking. If you don't do that, then you're not really enabling people to act independently or to kind of collaborate without your direct instruction, because without any clear kind of purpose, how can people make those kind of right, smart decisions on your behalf? And I think that was something we kind of got stuck in a loop in almost of having to be two directional because we didn't really have this big picture clear idea of where we're going. We'd grown luckily and successfully through a kind of opportunistic an opportunistic approach, but I think as we got bigger there wasn't that, that wasn't really going to kind of cut the, cut it anymore. And I think finally this is definitely really relevant to founders or management team. If you're going to do these things or try and roll them out, you need to look really hard at yourself I think to make sure that you can actually see through your part of the bargain. So if there's no point kind of saying these things if your actions immediately discount them. And I think that kind of personal reflection and the personal growth you might need to do to make sure that you can actually try and kind of support these kind of initiatives. If it is something you choose to do, it's something that's well worth kind of focusing on. So kind of into the action I guess in January we started kind of rolling this out. So we spent kind of a little bit of time in the painful stage and a little bit of time working out what we were going to do. And we started kind of creating a new decent handbook. So there's another director, Simon. We put kind of together in a Google doc how we thought the kind of the highest level kind of company operation stuff should work and probably spent about a week or two on it. I think we kind of built up in our heads kind of an idea of how we thought it should go. And then we shared it for kind of comment changes in debate. So we deliberately put it in an open document format that everyone could have input into and we probably spent, I think about a month working on it with everyone and it did change a lot. There was things that we thought we wanted that people really resisted and I think they were really right to resist actually. I think there were some bad ideas that got cleaned out in that process. There were some kind of questions about things where people just didn't really feel confident or that it was clear how things would work. I think it was a kind of edge case, well what if this and what if that kind of stuff that we we tried to satisfy with a kind of a different approach rather than creating a kind of 400 page long document that had every possible outcome in. And to kind of to drive that document I think we thought it would be we needed to kind of create some principles that we all could agree on and I think a key part of that document was the kind of debate and discussion process was what made it something everyone bought into rather than it just being something that was kind of pushed on people because I don't think it would have really worked if we'd taken that approach. So we created these five principles and we also kind of made some fundamental modifications to our working practices because we thought we need to get this different kind of mindset happening if we're going to continue to kind of grow in the future. And so those new principles the first one was our team members are intrinsically motivated to do their own job best so not spending lots of time on monitoring people not spending lots of processes on working out what people are doing or if they're doing what they should be doing we're going to make the assumption in the company culture we build and the systems we build that people actually do want to get their job done and the company's job is as much to get out of the way of that as it is to try and enforce it because do we want to put that much energy into the kind of enforcement aspect of things. Principle two was everyone must have a meaningful input and a culture of continuous improvement so by meaningful input we meant that it's no point saying well anyone can change anything or we always want to hear what you think if you don't act on that and if you don't provide a process that actually happens within then people are going to learn really quickly that you actually don't care or haven't found a way of caring yet so we wanted to make it sure that it was really clear that if someone could make a change that was within their kind of remit I guess that they should just make that change themselves because we did need discussion than the smallest group of people possible should make that decision as quickly as possible rather than it being something that we were kind of waiting on a hierarchy to do. Principle three was focus on results not process so we need as a company or a project on whatever level we're working on to determine what the results were looking for we need to launch this thing by this date the process of how we get there we should have as little touch on as possible but enough to make sure that we're not kind of reinventing wheels Principle four was around knowledge work so we do knowledge work in agencies we're creative whether that's through coding or design we're solving problems and we're using bits of our brain that we can't just turn up at work for six, seven, eight hours a day and the work happens it's not work that we can do by being present we actually have to bring that kind of time and energy to it or that focus to it and we all have good days we all have bad days we all have times of the day where we're not that productive and we don't have lifestyle choices or commitments that mean our lives are different shapes and what we wanted to do was actually say we really will care about the results and we don't really care about measuring who's in the office at 9am isn't really any useful measurement of productivity it just shows who's physically in the office at 9am and principle five was that big principles are more useful than small rules so we really wanted to keep this kind of really simple we didn't want to end up with a kind of contract of employment style intranet or kind of system that governed the kind of working practices we wanted to just make sure we had some bigger tests to kind of ask around those things and we felt that we were trying to give people kind of a map and assume that people we recruited and had within the team would have great judgement that was why we recruited in the first place so let's not spend loads of time then assuming that they don't want to do the job or they're not happy or they're not going to be engaged and I think some of those things they're kind of not I don't think many people would necessarily disagree with them personally I think they kind of seem obvious but I think it's a lot harder in practice or in company cultures where you evolve what you would like to kind of say on your poster if this is how things are but it's actually a lot harder to make sure those things happen every day and it was something we were really kind of conscious of is that this didn't really feel kind of like rocket science but it also felt like something that we certainly weren't in the place when we introduced them that certainly wasn't true we also had these the kind of the work in practice stuff that we changed so in the past we've been quite traditional everyone was largely based in one office working kind of a nine to five and we started to roll out these changes so things like a personal tool budget so every member of staff I think it's £500 a year they can spend on whatever they think helps them do their job better so the company still supplies kind of laptops, software the kind of core stuff and we deliberately didn't put this up for approval so you didn't need to get approval before you bought something you could use it for a standing desk, a wireless mouse a window shade, whatever it is that kind of helps we just put it in a central Google Doc so that there's that transparency and particularly for tool sharing if someone bought a Wacom tablet or something so make sure that we didn't end up with eight and actually we only ever needed kind of three at once we also introduced kind of flexible hours so effectively you could decide when you wanted to work there's some kind of we introduced the only limitation to that was really around core hours so if it was a day that you were going to work then you needed to be available I think it's between ten and two but other than that you can hit your kind of hours target for the week at any time that you want to and we also introduced distributed working at that point so you can decide where the most appropriate place to work could be a home could be a different city could be a cafe in town whatever their kind of choices when we first introduced it and up until now we've had a guide of that three of those days should be in the office but we're kind of trying to work out whether that actually makes any difference or not and whether it's actually quite different for different roles because if you're working in reception for example greeting clients that's quite difficult to do from home because those are a lot more straightforward and we also we kind of had this but didn't really say it very clearly I don't think we introduced an unlimited training budget so that people just needed to demonstrate the business value of what they wanted to do but really selling people's kind of expertise and if we felt that if someone we wanted to have no barrier there really if someone felt they needed those extra skills then they should kind of go and get them if you were pitching kind of kite surfing or something I think you'd have to work a bit hard that there was a business value case but otherwise you could kind of go for it and to wrap those up to kind of put some structure around how that worked we had a target number of billing hours per week so as long as you could kind of do all those things in any way you wanted the thing we were kind of really caring about the thing we knew that mattered to us as a company as a company that sells time was the weekly billing target so that was the kind of structure around it it was up to you how you got there and there's lots of reasons you could miss your weekly target so it wasn't that there was this kind of a punitive issue if you didn't hit it you might have been doing R&D you might have been training there's all sorts of reasons why you wouldn't hit it but it was just there as a kind of an awareness thing really that if you weren't going to hit it that you were kind of thinking about what else you were doing that was useful with your time to kind of to help people kind of work out what the right thing to do was we added these self-check so kind of questions that you could ask yourself or someone else if they were saying I'm going to go and work in Belize for the next three months we wanted to make sure there was a kind of a litmus test to hold these things up to and to see if they were if that kind of action would be compatible with what we thought was really important and these were the three things that we thought were really important is the way I'm working making our clients happy and effectively collaborating with my team and I'm happy with the quality of my work so really those were the things we thought were most important and actually if you were able to say yes to all of those then it was probably going to be okay what you were doing so that kind of gave us the working principles and the practices but it didn't really fundamentally change how we were going to work as an agency that didn't give us our kind of operational changes and back to those points we kind of needed that better data for people and we needed a new way of making decisions so at the same time we kind of changed how we were using some of the SaaS platforms we had introduced some new ones and also changed some company structure so we've been using Harvest for a while we used that for time tracking across the whole company for project budgets and we find that to be pretty good to be honest we've more recently started using Forecast which is kind of a sister product to Harvest that is for team resource planning base camp pretty standard for kind of baseline kind of team collaboration or client collaboration and comms document storage we used JIRA for more complex projects or bug tracking or task tracking about the same time we rolled out Slack I think previously we've been using something horrible I think we've been using Jabba actually using iChats with Mac based and there was no group chat I don't think it file transfers didn't work I'm surprised actually we stuck with it so long because looking back it was a complete nightmare that Slack has been really awesome actually I don't think a lot of what we've done would have been possible if we hadn't shifted to Slack and also we've been using Zero which is a cloud based accounting package and that meant we could give everyone access to the financial information to credit scores to invoising information to the right people this kind of collection of tools meant that everyone could pretty much see everything that was going on all the time the problem that we kind of found was that is that we had endless different systems kind of spitting out endless bits of data and the kind of the short term there's a longer term solution I'll come on to later but the short term solution is we started kind of mashing that into Google Sheets and that kind of works so we kind of have a dashboard of project statuses and project health that we can see as an overview of all projects that we're working on and some different kind of other company metrics but it's definitely not ideal but it was a lot better than what we had and to kind of to use this information that we had we were kind of adding these new meetings so we had a scheduling meeting that had kind of the key some of the key roles were involved in that and a delivery management meeting so they would happen every week there'd be a group of people they included me initially but they don't anymore and those people would get together to kind of resolve delivery management is project problems and scheduling is just the team scheduling so scheduling was now made by kind of a that kind of key project team people and they would discuss kind of schedule conflicts or blockers issues that needed resolving and that probably was probably about February I guess that rolled out and there were some negatives you know going back to that last slide with a billion dials we did have endless amounts of stats but I don't know that we were in some cases kind of much wiser for them so we were generating huge amounts of performance information but I'm not sure who was actually going to look at it or what they were going to do about it we'd also gone from kind of not much internal meetings in process and now we had lots of people in lots of meetings and we hadn't really in hindsight we hadn't really worked out a good meeting or maths we didn't have good meeting discipline so I think things were a lot better but they were the meetings were too long and unfocused I think and we were kind of sharing back to that dial slide in a way we were sharing everything with everyone but it wasn't really clear who was supposed to do something about it necessarily so kind of everyone being able to see everything doesn't necessarily mean that someone's going to do something it's it can get a bit messy but actually my feelings from what I know of the teams it was overwhelmingly positive I certainly felt re-energised I think there was a new purpose and a new insight into what we could be doing in the future people felt I think a lot more enabled to be able to make changes to use new tooling I was talking to someone from the team last night and he was saying that the amount of development process changed that we underwent at the same time and I was saying would you think this was related to that and we were kind of uming and aring about whether it would have happened anyway but it certainly felt like there was a big shift in our ability to kind of solve problems and do things better and I think that probably came in some ways from that energy and autonomy and certainly I think we were having conversations about the important things again so rather than lots of kind of detailed nitty gritty week to week problem stuff that doesn't really change the world we were able to start talking about were we going to start hitting these bigger goals what was going on and people were talking about should we add more team members for example so there was definitely a shift into the quality of conversation and also I was really surprised I genuinely didn't know what was going to happen around revenue and profitability I didn't know if we were going to see a slump through change or if we were going to have a rough period where we had to do a lot of reorganisation whether this was actually going to work or not but what we found was that revenue which had been growing actually grew a little bit steeper and profitability actually increased so I think what we were seeing was some of that the wasted time and energy that the previous model that we had kind of outgrown had introduced we were actually able to be able to be delivering in a much more kind of focused way you know I had some kind of well I think probably everyone did involve we had some kind of realisations I guess as we kind of went through that and some of them were kind of fairly quickly apparent you know I think one thing we found which didn't seem obvious was that lots of the team loved their actual job so they loved being a designer and the idea of sitting in a kind of scheduling meeting for two hours every week is pretty toxic to them and I think we kind of got along with the idea of oh everyone can control the destiny everyone can kind of change whatever they want isn't that a great thing and I don't think it is actually I think you need to be really selective about who's going to do what and why but surprisingly not all designers want to run an agency in hindsight I don't blame them actually I think realisation too that it was probably a personal growth one I think was that leadership isn't control and control isn't leadership and I think being able to let go of control where it's the only way you know how because you've grown the business that way in a way that you've been used to having that control I think understanding that I had a different purpose that wasn't about kind of keeping all those plates spinning that there was people who were going to be able to keep better plates spinning faster wherever I'm going with that analogy but they had a kind of it was much clearer that what the purpose of how those things could work and I think also there wasn't the perfect time for the perfect action it will never be the right time to make fundamental changes and you'll never get it right the very first time and I think I'm glad really that we did just get on with it we accepted that there was going to be some bumpiness and we did have to change things as we went but we knew we could adapt we knew we could adapt it and we knew that we wanted to set an aspiration we really wanted to raise the bar and kind of assume that there wasn't going to be people abusing the system assume that there wasn't going to be those problems and actually set a really positive bar out to go this is how we want to work and if there really was problems deal with it but not kind of build into the system that there was people who really didn't want to do it that way so that kind of took us through to probably around August kind of time and that's where we've kind of moving into the next round we've been kind of changing some of those things as we've gone but now we're kind of looking at another round I think of bigger changes and that's particularly around growth where we've been able to kind of keep that growth going but also around distributed recruitment what's going to be possible in terms of UK and Europe of how we recruit because now the kind of the historical team has that flexibility and it's increasingly distributed we've had a couple of distributed hires I think in the last month it then means we've kind of our London and South East recruitment pool is much wider now and I think we've failed in our kind of knee jerk swing to the opposite direction but accountability the first time round I don't think it was clear to anyone including me who was definitely accountable for something to be happening and in the past it kind of been in a broken way I think ultimately me but we kind of just introduced a different type of broken and I think that can still waste time and energy and it kind of gets frustrating and I think we've been lucky that there's a really kind of good team ethic they've worked together a long time we don't really hit big problems with it I think there still can be that frustration of who's going to make the final call on a scheduling decision for example if the six people who all have conflict in different projects or three different project managers with ten projects someone has to win or lose each time and I think there's not always a clear way that that can be a kind of democratic decision around things like resourcing so there's a couple of points I'll make later about how we've kind of changed that but I think we tried to introduce kind of democracy right through the kind of the base but we've done it slightly differently I think we're going to move we're kind of discussing it at the moment smaller fixed teams so rather than trying to schedule across 30-35 people as a resource pool which is great in theory because you can bring lots of people to a project quickly in reality that's not I don't think how projects work and how people run projects and clients so by having smaller fixed kind of pods of teams who always consistently work together I think we can kill most of our scheduling complexity and accountability problem because that team will always work together and is responsible for delivering those projects I think that kind of once your resourcing gets to beyond a certain point I think it just becomes really complex and I'm not sure it's a problem that's particularly worth solving because it won't be a free solution and those pod teams will be multi-disciplinary so that's a designer, UX, PM solution architect a couple of developers would be kind of a typical base team so they can completely control all the things they need to do to deliver a project I think one thing also was around kind of company scorecard so we hadn't really found a way to set out we'd set kind of some annual goals but I don't think we've really found a way to make what were actually connected day-to-day actions in with those so we're looking at the moment of how do we build a weekly company scorecard which has the kind of stats that change on every week that we want to have a conversation about if there's something that's not kind of looking right and I think a big part of that is something I was reading about recently was the kind of idea of trailing versus leading indicators and I think new business is a good example to demonstrate the difference between the two so revenue for example is a good trailing indicator it's the outcome of lots of tasks that you do and eventually you actually sell something the leading indicators to that are really going to a pitch doing a chemistry meeting how many people have you met at a marketing event for example so there's lots of things you might do ten marketing events to do five pitches to send three proposals to win one piece of work and with leading indicators what you're really trying to do is measure the lead through to that final piece of revenue because the leading indicators are the things you can influence if you know you're not going to that many marketing events for example or anything you can change quite quickly next week to actually go and influence your revenue next week is really difficult because you've probably got a two month leading kind of trail on that and those high level company targets I think also were something that we wanted to to work on on a quarterly an annual and a three year basis so that we had a kind of ramp of how if our three year goals are this how do they connect back to much more short term tactical goals so within that scorecard we'd have for each of those roles in the company whether it's marketing the delivery of projects recruitment, finance kind of three or four key stats that we think give a really good idea of how that area of the business is working and if there's a problem with the numbers that's a good thing really because that gives us a chance to talk about it at the right time so it's not kind of a performance management tool it's much more about understanding where do we need to spend kind of time and energy rather than someone's necessarily doing anything kind of right or wrong and so looking at the kind of what we were trying to do is also build this kind of model that would allow us to kind of recruit around UK and Europe so finding a replicable simplified model that we could then kind of clone out so that we didn't have to work out how to keep going through different growth barriers and I think there were definitely we're going to hit different challenges and kind of rolling this side of things out but I think it feels like there's a kind of a simple way of doing things that will actually be more straightforward and I think kind of in conclusion my kind of takeaways from what I've learnt so far I think we're definitely around in the early days I didn't really, I couldn't understand the difference between the signal versus the noise of what we were doing so where we were having problems and growing pains and things I think we got stuck in a loop of doing lots of short term fixes and actually understanding the root cause and recognising that we needed to do kind of a model change much more quickly so I think that's definitely something that I would want to make sure I pay more attention to myself in future I think for both, for culture you have an operational model and you have a company culture and I think it's definitely true that you have one of each whether you like it or not you will have a company culture whether you choose to engage with it or understand it or influence it is a different question but there will be a culture it's not something that you stick up on a poster or if you don't have a poster then it won't happen and I think both are really hard to see clearly particularly when you're really close to them and it was useful for us having kind of external debate and advisers to be able to just gain that much more insight and see what we were doing really well at and it was worth celebrating and see things that we weren't doing well at and we needed to change and I think for both of them again what we found is the process and the ongoing process really debating and defining both of them involving everyone that's really important it's not that you can you can't just kind of apply it to people and it will happen, you need to really engage people and work out what fits you and what will work for you and that's I think what we found looking at a fair bit of research and reading in some ways there isn't, what I realised I guess is there isn't a perfect model it's not that the model on the shelf is kind of good or bad it's your challenge in how you're going to apply it how are you actually, how relevant is it to you and what's your route to actually making it your own and I think that's the bit that it wasn't kind of clear at the time it's easy to get excited about let's do helocracy, let's do this, let's do that all of them are going to be a really long journey from where you are today probably depends where you are but you need to understand what kind of fits with you but then start making those small changes all the way rather than spend, kind of get stuck in analysis, paralysis of we could spend three years understanding different models in hindsight I wish we'd spent probably a little bit less time looking at options and more time looking at what immediate incremental changes could we start kind of rolling through and for the future I think self set pay is something that companies that go in this direction is something that people choose to do more and more I don't know if we'll go that way I think it's some of the most acutely personal and sensitive stuff that can go on and I don't really understand how that would work for us yet but it's something that is a natural conclusion in lots of the models that peer reviewed pay is something that people kind of work towards I think also self managing business units that's where going back to Ricardo Semla Brazil he ended up by kind of creating business units that could act independently because the overhead of managing these really large units was just really inefficient and bureaucratic so trying to create small business units that could self kind of manage is something that is interesting I don't really know I think we're too small for quite a while for that to be relevant to us but I think the idea of having those kind of inputs and outputs connected really closely and directly is an interesting one and something we'll definitely kind of keep an eye on in the future so yeah thank you very much to any questions so the question is how big is the company today it's about 33 people and about 2.5 million, 2 million worth of turnover so we're mostly based in so historically we've been based in a town called Canterbury in the UK which is just near London we're now over the past kind of six months or so we've become more and more distributed so we're probably only about 10% of the people I think by the end of next month we'll be not based in that area but we're deliberately kind of putting in the infrastructure that means it's much less relevant whether you're working at home 10 miles from kind of our historical location or if you're working 100 miles away the kind of the systems and practices we're working on now means that should be less relevant I think for us to grow the south east of the UK is a really for recruitment is a really difficult market so that's something that we're looking at of how would we continue to grow with the kind of people that we think are a great fit that we need to look a bit more imaginatively than where we have done in the past so the question was what's our largest project I think in terms of initial spend I think we'd probably be looking something like 200,000 UK so that would be a full service build with design UX and then some clients would spend that on a new basis on the same platform for example I don't think we've delivered as a one-time project anything larger than that and I think for us to do those kind of projects I think they change nature quite quickly I'm not sure if we were going to do kind of a half a million pound project I think that would require too much of a total percentage of our people that I'm not sure actually it would be healthy or a good idea necessarily Yeah, it's a shame I didn't put a kind of a slide in What I'd kept coming back to was really that the journey in the kind of how we do things is much more important than our actual size if that makes sense so we needed to keep a culture that people looked out for each other it was fun to do people enjoyed their work that we felt like we were delivering really good work that we were doing the right things for the clients we always have kind of really long term relationships with clients so we're often saying don't build this 50,000 pound feature it's just a bad idea I think those are the things that we felt or some examples of the things we felt were really important to keep rather than moving to kind of a very sales driven culture where the target was we need to hit this revenue this quarter no matter what and we'll compromise whatever it takes to kind of go that and I think during this process it was interesting actually to kind of and we just did some kind of seconds and three year and ten year goals actually and they weren't really around financial targets they were actually around much more about the size of what we do but making sure that we kept those other things in balance at the same time and I think it is important to make sure is your purpose to be a certain size or to hurt a certain size of revenue, size of people type of profitability if you can't answer that actually a level where you're setting the strategy I think you're going to have endless confusion in actually how can other people go and operate independently if they don't really know what's the best thing that could happen if they were all this happiness or this, this hello the question was around results versus process how do you not compromise that everyone's doing things in wildly different ways within the company I came I came from a development background myself so I've always I think been slightly too interested in the methodologies that we use or the IDEs or those sorts of things but I definitely don't know what I'm talking about anymore probably only borderline did at the time but I think what we found is that the cost of standardisation, I think there's a sweet spot really, the cost of standardisation can potentially be too high so where you think that standardisation is really important it isn't free, you have to train people, convince people possibly force people push those kind of tools or those practices onto people and what we wanted to say was actually if we've all followed the right methodology but the projects three months late and over budget well actually that's not success the success that our clients are buying from us and that the company can take to the bank is the project on time and on budget and those were the kind of if we had to choose something we thought well actually if developer A wants to use that idea and developer B wants to use a different one that's probably not the most important thing that we're going to hold up at the end of the quarter and say was that the kind of important decision we made so I don't think there's kind of a black and white answer to it I think it's just that you have to be careful not to get too bog down in things that actually you're kind of smart, talented, motivated team can probably work out five different ways and how much time within the company do you want to spend necessarily I mean it could be and I think there's also a balance between something like knowledge sharing versus mandating so it may be that you spend lots of time encouraging people to do things but you don't want to lose their motivation by forcing them so we have some things like security policies there's some things we just say this has to work this way and then there's another layer of things so if they would risk the company integrity client integrity, contractual issues there's some things that we say you just absolutely must do and then there's other things that we say well as long as you hit the big things does that answer your question yeah yeah so we use and that's a good point so we do use scrum actually and it hasn't come up as a a debate about whether we shouldn't yet and I think that would be an interesting one actually if someone said oh you know we fix team we want to do this in a waterfall way I think that would be an interesting kind of debate actually because we said we don't mandate those things but I think we probably might at that point so yeah yeah so in the past we did them as appraisals mostly me doing them and mostly the time I didn't get round to them often enough which I think was a really bad thing actually it's not good for anyone we then moved to peer reviews so we tried to keep it really simple we asked I think three peers two questions and I think the questions around can you tell me three things that I'm doing really well and three things I could do differently so that's completely self driven you just fill it out in a google doc it's up to you to go and choose the people that you want to ask those questions of and you just have to record it in a place that I think you and your manager can see it but you can choose to share it with everyone if you wish and you also set I think a goal at that point for the next quarter but again that goal could be I'm going to come to work and no one's going to challenge that it's up to you because in the past we've kind of been pushy I think about setting people's goals and I realise that if it's not your goal you're not going to do it it's just no point they have to come from the individual it's too much process around that appraisal stuff I think is just it's a false economy it doesn't really get what you think it is from it you just make people just keep jumping through the hoops but nothing's actually the real important stuff isn't changing had any problems I wouldn't have always had if that makes sense so rather than forcing most people to do things in a way they don't want just to make sure that one person isn't going to do the bad thing if that makes sense we kind of said well if someone has an issue at work that things aren't working that well they're probably going to have that no matter what so let's not make everyone work in a model that forces them not to be to do things in a certain way so the question was around how do we allocate projects to teams and capacity that's changing at the moment so in the past we could we would put a project team we would win a project we always try and bring the project team to pitches and to work and do the estimating and quoting on projects so we try and we don't have a separate pitch team or a sales team who are doing that we bring the people who are going to do the work into that process right from the start so they're the ones estimating specifying etc but occasionally that isn't true because either the right people aren't available or we need someone senior who isn't going to be the person doing all of the work if when we do win that project then we put together the right team to deliver it which should have been the same as the pitch team and that then but that's where we end up in that really complex resourcing environment because anyone could work across all of those different projects that's what we were finding wasn't efficient and we try and keep people on the same project for years if we can because you get that continuity but what I think we're doing in the future is we'll win a project different project teams different pods will pitch will be working on the pitch of the project and then that pod will deliver the project so rather than the resourcing being across 30 plus people it will just be it goes to this pod it goes to that pod the talk I mentioned yesterday Orange is the new teal or teal is the new orange or something they talk in a bit more detail about how they do things like business development at scale that then matches back to those individual pods too they're a bit more advanced than we are in doing it so it's probably worth watching soon Anything else? Great, thank you very much