Emily Nicholls 0:00:06.0: 

I'm Emily. I look after supply chain here at Anaplan, part of our product organization, and I'm joined today by two of our customers who have invested in our supply chain planning applications, one from CVH Spirits, and one from Mitsubishi Electric. I will let you introduce yourselves. Maybe, Derek, you could go first. 

 

Derek McCombe 0:00:25.9: 

Yes, hi. I'm Derek McCombe from CVH Spirits. I'm a planning manager for the Spirits Group, and I've been with the business for 31 years in a variety of different roles. 

 

Emily Nicholls 0:00:38.6: 

Thank you. 

 

Geoff Gelbier 0:00:39.2: 

Hi, everyone. I'm Geoff Gelbier. I'm the head of supply chain at Mitsubishi Electric Living Environmental Systems. I've been with the company about two-and-a-half years now, and I look after planning, procurement, and the factory relationship areas. 

 

Emily Nicholls 0:00:54.8: 

Amazing. Thank you. So both of you took our applications, both covering forecasting and demand, branching into supply inventory management. I suppose the best place to start with this conversation is to really understand maybe your motivations for taking on this project in the first place. So Geoff, maybe for you, what triggered your journey with Anaplan, and then, subsequently, obviously, Mitsubishi's full journey? 

 

Geoff Gelbier 0:01:20.4: 

Yes, so when I joined the company, we'd had an Excel-based PSI process in place, and I was tasked with bringing in an automated forecasting tool. We looked at various options at the time, when I came in, because we've got many systems within the company, so we were looking at various different - all the companies that you put up on the chart before. Anaplan wasn't one of them, but then we did a project with Japan, which is a factory planning tool, which, basically, is Anaplan, that they've turned into a Mitsubishi system. From that point on, we thought that this is a good idea to actually make a system that we can connect into the factory planning tool, because it will make it a seamless transition between the two. I was already talking to Vuealta, actually, from another project, and I knew they did a bolt-on system, and I was looking to kind of bolt-on to the Japanese system. I wasn't sure it could connect at the time, but the more investigation we did, we found that we could do it, they were willing to do it, and then you brought Vuealta's software as well, so you've implemented that. Then it seemed like that was the perfect partnership, really, for us to choose for our planning tool. So yes, it's been a bit of a journey, but we got to sign on the dotted line, so that was good.  

 

Emily Nicholls 0:02:44.6: 

Yes, I think maybe just a bit of contextual information for everyone in the room, with reference to Vuealta. Partner of Anaplan's. Historically, it was sort of two organizations. We had the applications part of the business as well, which is where I used to live, actually, and that part was acquired into Anaplan, which is exactly what you're talking about in terms of the solution. Derek, how did you start your journey with Anaplan? 

 

Derek McCombe 0:03:06.0: 

Yes, our journey was brought about - we were originally part of the Distell Group, a South African wines and spirits company, and our business got sold to Heineken Beverages, so we were annexed off as a result of that. So we became a standalone business, and we had sort of two options. We could stay with the Distell, or the Heineken planning systems for a period of time, or we could go with our own solution. The solution that we'd previously worked with under Distell and Heineken wasn't really set up for our business. We were quite a small business in terms of the scale of Distell, and all the processes and functionality that the previous system had were aligned to the parent company. So we decided to go our own road and implement our own systems, but we were then up against it time-wise. I think this was probably about June, mid-year, and we had to make a decision by the end of the year, so we tendered the system around about July, and met the guys from Anaplan and several other companies, and decided to go with Anaplan. We then went through an implementation project that started in September, and we implemented in 12 weeks. So we had peace of mind by the end of the year that we had a system that was robust enough to take us forward, and improving what we had previously, so we were then able to sign our exit agreement from the Heineken/Distell planning systems and go our own way with Anaplan. We've never really looked back since. It was a milestone moment for us, and a moment in which we could actually champion and determine how we wanted to plan. Whereas, previously, we were under the Distell rules and framework that you had to work within. So it was a real game-changer for our business. 

 

Emily Nicolls 0:05:10.0: 

Yes, that's really exciting to hear, and I think probably a lot of ears pricked up at hearing the 12-week marker. You were, obviously, under time pressures, for the reasons that you've just stated, but what made that 12 weeks possible, and were there any trade-offs or compromises that you had to make along the way? 

 

Derek McCombe 0:05:26.7: 

I think in terms of the 12-week timescale, one of the things that really enthused us about Anaplan was their confidence and ability to hit that timescale. When we embarked upon the journey, there was a really, really rigorous project plan, with really stringent deadlines. Everyone was focused and moving away from the system that we had. They were really encouraged to see the shiny new system that they were going to get, so everyone was really focused within the team. Quite a small team; five or six people at that time actually involved in the main implementation between supply and demand planning. In terms of the actual implementation itself, we had come through maybe two or three implementations of planning system over the previous ten years. As the parent company had changed system, we had to change accordingly, but none had really fitted our business. Previous to that, though, we did have a working model, and what Anaplan actually allowed us to do was, rather than ourselves having to fit in with the way the system worked, Anaplan actually fitted in with the way that we worked, so we actually picked the best of the three incarnations of planning system we had, and we designed our implementation of Anaplan in and around these various models. So we weren't starting from scratch. We had pre-ideas and concepts of what fitted and what worked well for us.  

 

Derek McCombe 0:07:03.5: 

There was a really dedicated team. You'd get emails at 11 o'clock at night from people, and emails at 7 o'clock in the morning, so everyone really worked around the clock, but I think everyone could see, even a third of the way in, that they were really gaining momentum, pace, and we were really going somewhere. So just real dedication, and we didn't have to really compromise. We covered all the main processes that we had, and in fact improved upon them. So we didn't really have to give anything up, but we had, as I say, a set idea in mind of what we were looking for, and they adaptability and versatility of Anaplan really shone through and enabled us to complete the project within the timescale. 

 

Emily Nicolls 0:07:44.6: 

Yes, that's really good to hear. We often talk to people about, typically, we are not the critical path in these projects. There's the customer, your data, how clean it is, how easily we can get hold of it, the people that need to be involved. Geoff, your journey was, obviously, longer. Mitsubishi is also a lot larger, with lots of business units, etc. What were the key, maybe internal factors, that ended up shaping your timeline with Anaplan? 

 

Geoff Gelbier 0:08:11.3: 

Yes, I mean, I didn't want to get involved in an 18-month project, so I wanted to do this as quickly as we possibly could, just so we could get up and running and using it, because I felt that once we were using the system we would be able to understand it a little bit more, and I think going through an 18-month project to build something is just not very productive time-wise. Vuealta said that - we, basically, brought in demand planning and the supply planning tools, or cubes. What do you call them? 

 

Emily Nicolls 0:08:43.3: 

Application. 

 

Geoff Gelbier 0:08:43.7: 

Applications, that's it. So three months for each one, so it should have been a six-month project, but as we got into it, we realized that our data quality wasn't as good as it should be, our hierarchies weren't where they needed to be either. Across the company, there were so many different structures. We've invested in other companies and brought their processes in, and the way they work, so from a material master perspective, we were all over the place, but it's been really therapeutic going through the process of understanding our data and trying to bring it together to fit into Anaplan, so that's been good. Also, whilst we've gone on that journey, we're trying to look at how we can turn it into more of a European or a global project, so that we can take the system and drop it into another branch as well. We understand our PSI process very well. The team are mostly here today. They know their stuff. The thing they don't know is how other companies do things, and what good looks like for other companies, and what I was hired to do was to bring in some other experiences from other places. What we could do by investing in Anaplan, we can actually bring those experiences in via your software, tried and tested on thousands of companies, and we can then try and tailor our planning to the way that Anaplan works.  

 

Emily Nicholls 0:10:05.9: 

Yes, exactly. I like that. I think you've both spoken to the premise of having something repeatable that is tried and tested, but equally, still flexible around the edges to make sure that it reflects your specific business needs. It's super-common to embark on a project like this and find you have more data issues than you expected. You used the word therapeutic; I think that's a good word. It's an exercise that needs to happen. It’s not painless, but actually sorting out your hierarchy, sorting out your master data, something that's going to have to happen at some point anyway, and when you implement advanced planning software, it really surfaces the issues in your data. It highlights issues that are maybe hiding underneath. So not uncommon to face those data challenges, but knowing what you know now, is there anything you would do differently, or would you approach the project differently, bearing that in mind? 

 

Geoff Gelbier 0:11:00.3: 

No. It would be nice to have it all the data stuff sorted out from the start, but I think, actually, the project was fairly smooth getting it up and running, Vuealta guiding us where we needed to put things, what the structure should look like, and that just gave us the ideas of what we needed to put in place in order to bring in the data in the right order. I think just getting the data in there so we could use it and see what it's presenting, and quite quickly we can start seeing things like the ABC categorization, and all those kind of charts and statistical forecasts, and how that's playing out within it. We didn't know how to use it, but we could start to see that it was doing something, and that piqued our interest I think. So yes, I don't think I would have done it differently, necessarily. I think for our sized company, I think it's been perfect. 

 

Emily Nicholls 0:11:53.5: 

That's good to hear. Is there any advice you would give to somebody embarking on their journey for the first time? 

 

Derek McCombe 0:11:59.7: 

No, I would just say, try and make sure that your data is as aligned as possible, and also think about the longer-term picture of where you want to house your data. One of the issues that we had in previous systems is that, being a Scottish whisky company, there's lots of different aspects that you need to consider when planning production of your product, and we didn't always have, necessarily, the right place to store our data, and Anaplan have added a master data structure, in addition to we operate SAP, as well, as our ERP system, but we've built a table that gives us really enhanced capability reporting-wise within Anaplan. So I would say, obviously, the cleanliness and the accuracy of your data, but also thinking about future-proofing yourselves, as well, for the implementation. We thought we'd done enough of it. We probably could have done more of it, thinking now, retrospectively, but that supplementary data and Anaplan reporting-wise gives us dimensional reporting that we could only have dreamt of before. So that was a real game-changer. Going forward, we're probably going to have to modify the whole system for some of the data, just as we make changes, as we're on our journey. So as I say, I would just think outside the box. If there's things that you're going to implement in a phase two or a phase three, think where you want to house that data. But as I say, game-changing for us in terms of the data that Anaplan could hold. As regards to data cleansing, everyone you talk to has got the same issue as regards to data cleansing. It's never as good as what it could be. 

 

Emily Nicholls 0:13:49.6: 

If only some kind of AI tool would come out to identify anomalies in your data search. We talked a lot about data and data challenges. I think everybody's quite familiar with those, but what about more the human side of your transformation, often, actually, the most difficult bit, some people find? Technologies is kind of easy, data is what it is, but we will always talk to people about people and process as well. So in terms of change management, in terms of bringing in something quite new - you both started from different positions - maybe I'll start with you, Derek, because you're almost fresh after being spun off, but what did you learn along the way about bringing people on this journey and really getting them to adopt and get excited about the system? 

 

Derek McCombe 0:14:31.8: 

Well, we're quite a small team, as I say. The team is probably five people that were really involved in the project at quite a detailed level, and they were involved, really, from the initial demonstrations to concept design, solution testing, and then sometimes you'd have to rewind and cross-correct and then modify. So the fact that they were so heavily involved, actually made them buy-in more, and I think when they could see what they were going to be getting from the application in terms of output versus what they had at the moment, that drove them on as well, but for me, it was a bit… Being a stakeholder, I was previously involved in an IT project about 15, 20 years ago, and CVH put in an ERP system, so I've always had a really keen interest in both systems and planning, so for me it was a perfect project to get involved with, and I think that's spun off in other members of the team. Basically, because they all felt that they had a voice at the table, they had a say in what was going to improve their day-to-day ways of working, encouraged buy-in and support. So I think that accountability as well. As I say, the project routine that Anaplan put in place was very, very rigorous, and you didn't want to miss any of your deadlines, and everyone bought into that. Because it was a small and compact team, as I say, everyone really was supportive. 

 

Derek McCombe 0:16:11.1: 

Another thing is, we had several different parties actually working with Heineken, actually, from a data management perspective, because we were still on their SAP system, although we were putting in Anaplan, so they were involved. We had our support partner, we had Anaplan, we had our own IT team, so for such a mixed spectrum of people, sometimes there were challenges, but everyone was really determined to hit the deadline for the greater good. As I say, everyone's support was required, because it was such a strict timeline we had to work to.  

 

Emily Nicholls 0:16:46.7: 

Yes, and I like what you say about that almost internal ownership. I've seen that as well. You don't run into as many adoption issues or change management issues if people have had a say along the way, or at least been able to see regular updates, see what they're getting, rather than, 'Oh, you want a new demand planning tool. Cool. I'll be back in eight months and you'll have something. Don't worry about it,' kind of thing. Geoff, obviously, your team are all here, so you've got to be a little bit careful about what you say, but did you have any issues in terms of bringing people on the journey, even just, not through willingness, but sometimes through their day jobs? They're really busy. This is another thing to add to the list. Do you have any lessons learned? 

 

Geoff Gelbier 0:17:23.9: 

Yes, definitely. Obviously, it doubles their workload, because they've still got to produce the Excel sheets, and they've got to learn the new system as well. I think, fortunately, we'd gone through the Japanese system. We'd started using that last year, so we were kind of au fait with Anaplan, and how easy it can be once you understand it, but this was a whole different ballgame, because it's so much bigger. I think, Danny, from Unilver, was really explaining what you need to do, and how you get stuck in that mud. We did go through phases where we were stuck in that mud, and we couldn't change it, but we've got to do it this way, we've got to do it this way, and we're going, 'No, this is how we want to do it.' So it is a journey of how we want to change and make things better, simplify, make it easier for people to do the job, reduce the time it will take you to do all the Excel sheets that you do, explain what the difference is going to be. It's a frightening change.  

 

Emily Nicolls 0:18:27.4: 

Yes, but I guess selling the value to them, how this would improve your life, as well. 

 

Geoff Gelbier 0:18:32.8 

Yes, in the future your life is going to be so much different. 

 

Emily Nicolls 0:18:36.7: 

Yes, totally. So thinking about value - you're, obviously, on different journeys - you've been live for some time now, you're still implementing - but I would like to think about now those aspects, maybe starting with you, based on what you've just said. How did you go about communicating even the early stages of value, or the indications of value that you were going to get - that can be a really key element to get buy-in from your team, but also exec alignment - really celebrating the success of the project, essentially? 

 

Geoff Gelbier 0:19:10.7: 

We knew where we were from the start, so we looked forward from a planning perspective. So there were lots of elements that would easily bring in the forecasting accuracy piece that often is so important, but we don't really look backwards as a company. Anaplan will allow us to do that much quicker, and we just didn't have the time to build those reports, so that's one huge benefit that's going to happen and come out of it. The data structure thing, as well, I think is really helping, because we've not only done it within Anaplan, we've actually pushed it back to our core ERP system, so we're getting one source of the truth. That feeds Anaplan. That will feed our other systems, that will feed our CRP system. So cleaning up the data, and allowing that to come through, and making that easier and better for us, also working at how we can bring that data in from the factories much quicker and smoother. So there's lots of things that will save time, and as we've gone along in the project, you can start to see all those time savings adding up, and that's before we really get into the ROI and what really is going to make the difference. 

 

Emily Nicholls 0:20:16.4: 

Yes, and I guess the second part of my question about communicating that I suppose internally, have you found that easy? Do you have anything in place to do that? Is it something you're thinking about doing? 

 

Geoff Gelbier 0:20:26.8: 

It's been difficult to do, I must say, because we're on a rollercoaster with the project, so we're just going through and getting it installed. We're trying to evangelize and we're trying to tell people about Anaplan. We're trying to explain what difference it's going to make to other areas of the business, as well, but it often comes down to people feeling, oh, you're going to need more information from us. So it's trying to explain to them that, actually, this is going to make your life better, not harder. Initially, it will be harder, because you'll need to give us more, but then we give you more back, and ultimately, for the sales guys and the marketing guys, they'll have the right products when they need it.  

 

Emily Nicholls 0:21:03.3: 

Yes, people don't like change, even if that change is good. I see that all the time, but it's good that you've got mechanisms in place to get the team excited, etc., and sharing it more broadly can be more of a challenge, for sure. Derek, what value have you seen with CVH? You've, obviously, been live for some time now, so you have some more quantifiable business impacts I suppose from the project that you've done. Could you share some of those with us today? 

 

Derek McCombe 0:21:28.7: 

Yes, certainly. Well, we've improved our forecast accuracy by over 20 per cent, so we're upping that every year. I have to say, we did start from a low base, but it is improving. We find that the information, availability of information, and the fact that information is live, similar to what Geoff said, rather than going through Excel after Excel iteration, you can actually go and view it on a screen as live. So that's one notable improvement is the forecast accuracy. We've gone from 33 per cent to 56 per cent at Lag 3. We work to a 12-week planning horizon, so it's brought about greater stability within that short-term period. In terms of other notable changes since the inception of Anaplan, we've reduced our stockholding by 25 per cent, and that, again, we're on a journey, and due to the historical forecast volatility, we had built up a considerable amount of stock. We're still working our way through that, but the idea is to reduce that even further. Stock days cover was 109 days, and we got it down to 84 days at the end of this fiscal. We're targeting 55 days, so we're, again through Anaplan, looking at all our [?slower 0:22:53.8] in stocks and ways to run that out. So these I would say would be the main notable. What it does do, as well, is it enables you to line up all your targets together, so your bat sizes, and your bat size is going to take to your 55 days based on the governance rules that you've got in the system. So it allows you to make sure that you've got all of your key parameters working and synchronization with one another to achieve the key business aims. That was something we struggled with in previous systems. 

 

Emily Nicholls 0:23:30.2: 

Yes, and I think a lot of people will think about the main KPIs, to your point, but there are always loads of softer benefits, as well, which can ultimately lead to the more tangible stuff. The forecast accuracy piece, obviously, a huge jump, but like you say, it's something that we'll continue to improve as you continue to use the system. Although that's not the aim itself, it obviously has a lot of very significant downstream impacts. A similar question to Geoff. When you think about those successes, have you found it easy to communicate that to the team? Obviously, you're a smaller business, so maybe your team are living and breathing those successes on a more day-to-day basis. 

 

Derek McCombe 0:24:10.3: 

Yes, I think, as I say, the fact that the team can actually see the information on screen, so the KPI information is published monthly, but I think the fact that they can see the various elements of information on screen. One of the things we got embedded in was a production [?LA 0:24:28.9], and when we saw the capability, something previously we'd have done in Excel that would have been quite time consuming, so we rolled that out to make that, not only production LA but a spirit reconciliation as well, so as we could tie in with the bulk and blending team, because they previously had one set of numbers, and we had another, and we could never agree. So it's the one version of the truth, really, that everyone, even departments outwith, can buy-in to. We also piggy-backed onto the spirit reconciliation. We did warehouse capacity planning, as well, so that again is really useful information to see where you might have available capacity, in your warehouse, even, to store dry goods, or to store other things. So there's lots of spinoffs over having aligned KPI targets, having that visibility, making sure that's well published within the business. More important than anything, good management information that you can rely on and you can action on to improve business performance, ultimately. 

 

Emily Nicholls 0:25:38.1: 

Yes, totally. I think you started to touch on there, spinoffs, other use cases. We do tend to find the technology is fairly insidious. It kind of lands somewhere, someone sees it and are like, 'Ooh, I could use it for XYZ also,' and you naturally start to branch off. Are you finding something similar at Mitsubishi? Do you have longer-term plans, appreciating that you haven't finished your current rollout?  

 

Geoff Gelbier 0:25:59.6: 

Yes, I have a lot of plans, but I don't know if Matt sees the same plans as me. We're starting small, but as we go through it, we've already found that we've got S&OP reporting that we're trying to implement now on top of that. That will give better reporting to the leadership team. We know we've been trying to get the sales forecast element into our forecasts, but why shouldn't that be part of the whole planning tool as well? We can use the data we're doing to actually base that sales information from Anaplan itself, bringing more CRM data, so we can do more scenario planning. Yes, there's lots of things. Budgeting, we can go back to budgeting as well. That's before we even get into the pricing thing. There's a lot of people that are asking questions on pricing within the business that, like I said before, we don't have that connected system, so it would be nice to do it in one place, and then we would benefit from that to see all of it coming through our demand plans, and hitting our supply plans. 

 

Emily Nicholls 0:26:59.6: 

Yes, and I think it's about finding a technology that can go at the speed that you're ready for, as well. It doesn't have to be one big-bang approach. You can really take it piecemeal and start small, start in a specific market, rollout to other markets, start another use case, because as I say, the critical path really is you and your business and what you're ready for, and what the data set is ready for. What about you, Derek, what's next on your Anaplan roadmap? 

 

Derek McCombe 0:27:24.8: 

Oh, many things. I've got a lengthy list. One of the things that we talked about earlier on is product portfolio, so we're really looking to see about actually getting something that automates, and the viability of your SKUs and the performance of your SKUs, the market share. We have a very grand collection of SKUs, I have to say. We've been through a couple of name changes, we've been through Brexit, and every time an event like that happens we have to change SKU codes, so that with it brings about a scale of maintenance. So thinking about that, and also thinking about different views that will give commercial readily-accessible information on stock. Historically, again, we had downloaded stock information from SAP, circulated it on a weekly basis, but we're looking at some sort of live projection, to show here's the stock, here's the orders we've got loaded, so as they can see the net position, just to, again, give greater access of information to maybe people who wouldn't think of accessing a planning system previously. So these are two of the many, but there's different projects that we're looking at to further expand Anaplan in-house.  

 

Emily Nicholls 0:28:41.0: 

That's great to hear, because we do often find that the more you do in the platform, the more value you get, for two maybe obvious reasons. One being you can rationalize the number of systems that you're actually buying - you don't have to buy something bespoke for every single use case - but secondly, if you have all of your planning in a single environment with a single data source, then you're all talking about the same numbers all the time. To your point, you can be confident in that KPI matching, productivity increases, and blah, blah, blah. Thank you so much for talking to me today, for sharing your stories. I think we've heard kind of two different paths to success, across two very different customers. One of the reasons that I like my job, is I can learn about how whisky is made, and I can learn about how Mitsubishi move goods around. Completely different, but ultimately, using the same frameworks. 

 

[Applause]