George Milic on supply planning across industries and cultures

George Milic - Navigating supply planning across industries, systems, and cultures

supply planning George Milic – Navigating supply planning across industries, systems, and cultures

Episode description

In this episode, Ben and Wim welcome George Milic, a seasoned supply chain professional with over two decades of experience at renowned companies like IKEA, Apple, Zoetis, and currently Unilever.

George shares his journey across different supply planning roles, reflecting on key transitions across sectors such as retail, tech, pharmaceuticals, and consumer goods.

He discusses building S&OP processes from scratch in a startup environment, adapting to fast-paced supply chains at Apple, and aligning financial and demand forecasts. George also provides insights into training planning teams, leveraging systems like JDA, Anaplan, and SAP, and adapting to cultural differences between the U.S. and Europe.

Key topics covered include:

  • Building S&OP processes in startups with limited tools
  • Differences in supply planning between retail, tech, and pharma sectors
  • Comparing supply chain cultures between the U.S. and Europe
  • Leveraging AI and machine learning in modern planning systems
  • Best practices for training planners and facilitating system adoption
  • Aligning financial and demand forecasts in cross-functional S&OP
  • Key advice for young professionals entering supply chain

TIMESTAMPS:
  • (0:03) – Introduction & George Milic career path
  • (2:30) – Early experiences at Dollar General: operational supply planning and forecasting
  • (3:33) – Moving to IKEA: supply planning, customer service KPIs, and working in Europe
  • (4:38) – Shifting to process support roles: training planners and system enhancements at IKEA
  • (5:44) – Implementing S&OP processes from scratch at a startup: Excel-heavy planning challenges
  • (7:15) – Joining Apple: leading global supply planning for iPhone models
  • (8:31) – Comparing supply chain dynamics at IKEA and Apple: speed, value, and priorities
  • (10:45) – Role at Zoetis: bridging financial forecasting and demand planning
  • (12:04) – Aligning financial and volume forecasts: challenges and best practices
  • (15:02) – System usage in planning: SAP for demand planning, Anaplan for financial forecasting
  • (16:01) – Current role at Unilever: demand planning for the nutrition portfolio
  • (17:06) – Reflections on supply chain systems: similarities, differences, and automation progress
  • (19:00) – How AI and machine learning are shaping demand and supply planning today
  • (21:00) – Best practices for training planners: staying close, customizing training, and process feedback
  • (24:07) – Cultural differences in supply chain planning: U.S. results focus vs. European relational emphasis

  • (26:29) – Final advice: continuous learning, attending trainings, and networking on LinkedIn

TRANSCRIPT

(0:03) Welcome to the Supply Chain Planning Podcast, in which we talk with supply chain professionals (0:08) about their experience in planning. My name is Ben, co-founder of Horizon, and Wim is our co-host. (0:14) Hello, Wim.

Hello, I'm Wim Farrazin. I'm the Managing Director at Lanard. For today's episode, (0:21) we have George Milic on the podcast, who has been in supply planning roles for more than 20 years (0:26) at companies like IKEA, Apple, Soedis, and currently at Unilever.

First of all, (0:33) welcome, George, to the podcast. Hi, Ben. Hi, Wim.

Thank you for inviting me. Happy to be here. (0:39) So, as a kickoff, you studied economics before working at Dollar General Corporation, which was (0:46) a discount chain.

How did you find the transition into a supply chain role after those studies? (0:51) Did that transition go well? It was actually a good transition that I'm happy with. I need to (0:56) be very honest. While I did go to school to study economics at the university, I did not initially (1:03) envision working in supply chain postgraduate study.

I was initially expecting to find a job, (1:11) potential job opportunity in the finance, banking, or investment industry, but I did happen to come (1:17) across an attractive supply chain role. And then you enjoyed it quite a bit since you stayed for (1:23) two decades in the industry, right? Yes. Yes, of course.

(1:27) Looking at the roles and the different process areas that you worked on, (1:31) that first role at Dollar General was global sourcing coordinator. Is that a more commercial (1:37) role because of the word sourcing, or is it really operational, more material availability (1:43) kind of aspect? Yeah, I would say the job title was a little bit misleading. This was predominantly (1:47) an operational role.

It was entry level. And basically the key functions of the role were (1:53) to do demand forecasting for the portfolio, but also to manage the supply planning aspect in a (2:00) sense that I was responsible to maintain appropriate inventory replenishment for the (2:05) company's warehousing sites across the US. And I was also responsible to do allocation to ensure (2:13) that when POs were arriving from overseas, that they were properly distributed and allocated across (2:19) the company's DC site network.

Ben quickly named a couple of companies you worked at. (2:25) Perhaps for the audience, can you very briefly talk us through how your career evolved? So the (2:30) different companies and roles you worked at? I did start my career at Dollar General, which is (2:37) FMCPG company, fast moving consumer package company. And then in my next role, I transitioned (2:43) to work for IKEA in their North America regional office.

My first role at IKEA, it was, I would say, (2:52) also kind of leaned heavily on kind of more in the supply planning aspect. The key responsibility was (2:58) to monitor the warehouse in stock that's supporting the store to make sure that we're having appropriate (3:05) orders coming from the supplier and appropriate volumes to support the business. I would say (3:11) customer service availability was the most important KPI driving the work.

I didn't really do directly (3:18) demand forecasting in that role that was done at global corporate location in Sweden. I actually (3:26) had a next step with IKEA. I actually went and worked abroad in their regional support office (3:33) located in near Basel, Switzerland.

This was the first time in my career that I basically (3:40) pivoted to working more on a process support capacity role versus doing the day-to-day (3:47) operational work. I had the responsibility to train the planning team. There was a team of, (3:54) I think, approximately 50 to 60 planners at the time in the business.

And I had the responsibility (4:00) to train them how to utilize JDA fulfillments and distribution solution systems to go about their (4:08) jobs. How did that change from a purely operational role into this more central capability role? How (4:15) did that go for you? Did it give you new insights or a different way of looking at things? Yeah, (4:20) it was a really exciting opportunity because I gained a lot of just additional training and (4:25) knowledge on how the solutions are working, how to go about facilitating change requests, (4:31) working with IT to develop enhancements, to improve process ways of working in the planning (4:38) space, and also utilizing that capability to train planners how to utilize that system was (4:45) very interesting as part of my development. The things that I always found challenging in these (4:50) central capability roles is how do you stay in touch with what is really happening day-to-day (4:56) on the floor or behind the desk? How did you tackle that to avoid that you become this (5:01) central ivory tower person who comes up with ideas and processes that look nice on paper but don't (5:08) really work in practice? When I started that role, that was an initial adjustment that I kind of (5:13) struggled with.

Essentially, the way I handled that problem is I recruited and led what was (5:19) called focus groups. Essentially, these groups comprised of the planners. There would be one (5:25) focus group that was committed to improving a process and ways of working with MPI, new product (5:32) launch in the EU market, what would comprise of at least three or four planners who are working (5:38) operationally in the business.

Okay, that's smart. I did eventually return to the United States. (5:44) Basically, I went home, moved back to my home country and I took a role to implement (5:52) an SNOP process in a very young startup company, essentially a very new business.

(5:58) At the time I joined the company, the headcount was probably only 50 people (6:02) and the industry was selling beauty and skincare products in the US. Were there (6:08) any existing processes then or did you have to set up everything from scratch there? (6:13) There was no SNOP process in place. This was a startup.

You had to go about the work in a very (6:19) scrappy level. There was limited even solutions or IT solutions in the business. I would say (6:26) from a supply chain perspective, the work was heavily reliant on Excel spreadsheets.

(6:32) Yeah, because your previous experience was a lot with JDA, working with it yourself, (6:36) getting other people trained in it also. But then that experience, (6:40) were you able to pull from that for that new role? (6:43) In achieving a successful SNOP process, data management and data exchange is very, (6:48) very crucial. That was, of course, a very challenging adjustment to pursue, to try and (6:54) have the right supporting documentation, the right information available and refreshed in a timely (7:01) matter.

And to be able to do that off of an Excel spreadsheet-driven process was, of course, (7:07) challenging. It did drive some hiccups. (7:11) Okay.

And where did you move after number seven? (7:15) So I moved to California, to the Silicon Valley area, and I managed to land a contract consulting (7:21) role at Apple. I joined their iPhone planning team, which was part of their AppleCare division. (7:28) And in that role, I was assigned one of the Apple iPhone models.

Since I was newer to the team, (7:36) I was given a couple older models to work with. And I was essentially having the responsibility (7:43) to be a global lead supply planner for one of their iPhone models. (7:48) Was that role a global role or focused on the North American market? (7:53) I was global lead responsible for the model that I planned.

It was my responsibility to (7:59) review all the work that was pulled and all the planning needs that was pulled together by the (8:04) regional planners in the other markets. And I was ultimately accountable for what decisions were (8:10) made as far as how much new orders or POs would be placed for products in the business. (8:16) Yeah.

I think IKEA and Apple both joined companies, but in completely different (8:22) sectors. But you spot the supply planning in both. Can you (8:26) see clear differences between both in processes or systems or organization? (8:31) The dynamics working in tech, at least in my experience from Apple, was different in a sense.

(8:37) Planning for consumer electronics was essentially a weekly process. It's very fast-paced, (8:44) very high-speed. Air freight is used for everything.

You place a PO, you're receiving (8:50) that within four to six weeks later via air freight. So it's a very fast-moving business. (8:55) Another change for me versus IKEA was the portfolio value of the products being planned.

(9:02) Did you notice any difference as to how supply chain was viewed internally? (9:06) I think the difference in dynamics is more concentrated in priority of the products. (9:12) If we talk in terms of like ABC segmentation, when you're planning iPhones, for example, (9:18) you of course have very key service level and KPI targets for the higher value items. So you're (9:25) planning a whole unit module, so a completely built iPhone, but you're also planning individual (9:32) components such as the screens, the motherboards.

The priority from a day-to-day work level (9:38) perspective was really on those higher value. Either you're planning a whole unit module phone (9:44) or those higher level components. While in IKEA, it was more flat.

Like the KPI targets, (9:51) the service level, everything was generally important. Whether it was you were planning (9:55) a bookcase, everything was kind of treated more equally as a priority. (10:01) Does it make it easier or more difficult if you have a similar approach for all the parts or you (10:07) have this very specific priority versus non-priority parts? What felt easier? (10:12) Basically, when you're meeting with the management team to review your plans, (10:18) there would just be laser focus on those higher priority SKUs, like a whole unit phone, (10:25) or whether it be the screens.

The screens were very expensive. The motherboards are very expensive. (10:32) You're basically given the unilateral discretion to act independently in planning those.

(10:39) After your time at Apple, you moved to Zoetis, a completely different line of business. (10:45) Can you explain that a bit? (10:47) Sure. Zoetis happens to be a pharmaceutical company that is technically a spinoff of Pfizer.

(10:53) They used to be part of Pfizer in their past. What was also interesting for me in this opportunity, (11:00) it was actually the first time I took a role outside of supply chain. Technically, (11:05) when I joined Zoetis, I worked directly in their marketing team.

(11:09) What was the role? What was it about? (11:12) The key focus area was to improve the financial forecasting process that the marketing team needed (11:20) to execute for the business, for their respective brand portfolios, and also ensure that there was, (11:27) from an S&OP perspective, to ensure that there was appropriate financial forecast input being provided (11:34) to the demand planning team in supply chain, because it was an opportunity to work with (11:40) developing and improving their financial forecasting process. (11:45) That was the crucial deliverable from the marketing side. (11:49) By financial forecasting, you mean the volumes translated to full values, (11:54) and to have that for budgeting also or for marketing to set campaigns? What's done with it? (11:59) With the support of the team I had there, we improved financial forecasting process to (12:04) incorporate customer insights to ensure that we had appropriate net revenue projections versus (12:11) the company's budget.

Those plans would be compared and used as inputs to the volume (12:19) forecast that the demand planning team was delivering from a supply chain perspective. (12:25) Those were separate. The demand generated forecast was translated into a financial number.

(12:31) You made your own financial number, and those two then got reconciled, more or less. (12:35) Yeah, it's two separate inputs. Then there was the opportunity to compare and say, okay, (12:41) assuming no change in price, for example, is year-over-year growth in that revenue (12:47) reasonably or directionally aligned to what the year-over-year growth in volume is for the (12:52) business? That's just one example that you could compare.

(12:55) Do you think it makes sense to combine those two, or did you find the split valuable? (13:02) Because you're doing somewhat similar things from the ground up, right? The end result is different, (13:07) value versus volumes. But the driver behind it is, yeah, what are customers buying? What (13:12) are the patterns? So it seems like double work being done to some extent. (13:16) In some companies, it's typically one input.

You can have your demand forecast in volume, (13:22) but you can also have that translated into a net revenue dollar value. So it's one input (13:28) that is presented in both volume and net revenue. But in some businesses, they do maintain and keep (13:36) separate inputs.

So each team is owning, the marketing team is owning the financial projection (13:42) aspect. Demand planning is owning the volume projection aspect. And then you have processes (13:47) in place to kind of run comparisons and make sure that there's reasonable directional alignment, (13:53) and you address any gaps, significant gaps or variances.

(13:56) Indeed, in many companies, there is indeed a split. And then, like you mentioned, sometimes (14:01) with marketing or sometimes just finance with the budget, then there's sales and marketing with (14:05) their targets and everything. And then the supply chain, just wondering if there is (14:10) more possibility to integrate? Yeah, that's a good question.

So what I can say from an S&OP (14:16) process perspective, in my opinion, you don't need two separate inputs. You can technically utilize (14:22) the demand forecast that the supply chain team is producing. And you can ensure that all S&OP (14:29) cross-functional collaboration is concentrating on what that demand forecast is.

And then that (14:35) demand forecast would be communicated to your S&OP team on a cross-functional basis in both (14:41) volume and net revenue. So everybody's kind of collaborating on one input that feeds into what (14:47) should ultimately be the consensus for the business. That's definitely the way it's done (14:51) in some companies.

That is a potential solution to delivering that goal. (14:56) And from a systems perspective, did you also all use different systems to create your plans? (15:02) Working in Zoetis, the financial forecasting process, some work was done in spreadsheets, (15:07) but they did have an IT solution. It was called Anaplan, which is pretty common in the industry.

(15:14) I even use Anaplan today in S&OP. Did the demand planning team also use Anaplan? Because (15:19) I've often seen indeed Anaplan being used in financial planning, also in supply chain planning, (15:25) but predominantly financial planning, if I'm not mistaken. No, the demand planning team utilized (15:31) SAP to manage the volume aspects, but essentially the numbers would be compiled and exported for (15:39) comparison purposes during presentations and so forth.

And after Zoetis, you moved on to Unilever, (15:47) where you're currently at. What are you doing there? (15:51) Yeah, I currently work in Unilever as an associate demand planning manager. I work (15:56) under the nutrition portfolio in the US retail business.

(16:01) This time, you don't have the supply element as part of your role. It's really just focused on (16:06) the demand planning and then putting everything together in the S&OP. Or does that S&OP contain (16:12) a lot of supply parts as well? I'm not working day-to-day operational (16:15) in the supply planning activities.

No. What is important in my current role is to, of course, (16:21) be cognizant of what is going on in the supply planning space. For example, if there are service (16:28) issues or stockouts, or if there's potential capacity constraints that could affect the (16:36) company's sales performance, those inputs and information would sometimes need to be accounted (16:41) for or recognized in going about the day-to-day work and ensuring that you're having an appropriate (16:48) demand forecast and that you're able to have an understanding of numbers and a narrative to (16:54) explain the forecast performance.

You've worked with quite a bit of systems, (16:59) right? Kinexis, Anaplan, JDA. What kind of things do you see that these systems are lacking or that (17:06) you think, oh, this could be improved still in the industry in general? (17:10) That's an interesting question. There seems to be sometimes a lot of emphasis in the supply chain (17:15) industry, for example, especially in the recruiting world, that they're sometimes looking (17:19) for candidates to have knowledge in certain systems like SAP or Kinexis or agility or whatever.

(17:26) Honestly, I do have experience of using many different systems now in my career. (17:31) Honestly, I would argue, in my opinion, that on a very general level, all the systems are (17:37) kind of fundamentally very similar. They're all good solutions to have versus having nothing at (17:43) all to support day-to-day planning.

Because you worked in the different sectors, (17:50) how do you make a selection? If you're a company envisioning implementing one of those systems, (17:55) if they're all so similar anyway in basic setup, what then helps you make a selection decision? (18:02) Except for the price element, of course. Is there nothing functional that will (18:05) help you move forward in a certain direction? I kind of feel that every company that I worked (18:12) with just kind of naturally happened to just adopt systems for different reasons. (18:17) I can't go into details on why they picked the systems they did.

(18:21) And to what extent do you feel like these systems have made a lot of progress (18:25) versus then 15 years ago? Do you feel like things are more automated these days? I mean, (18:30) there are a lot of terms being thrown around within the supply chain industry in general, (18:35) I think, about autonomous planning. Have you seen these things in practice? (18:40) I've definitely seen improvements in IT solutions over the years. It's well known in the industry (18:47) that there's more active engagement in using ML, machine learning, and AI technologies.

(18:53) If I go just for more specific examples to drive baseline forecasting as part of your demand (19:00) planning processes, I would say these technologies are very helpful in the job to kind of help (19:07) support automating and processing information in a more efficient, effective manner. (19:14) Have you seen a lot of changes in the supply planning side as well, with the use of models (19:19) or specific solvers? Or there you see less change and less progress of machine learning? (19:24) It's my understanding, and I have the impression that there's development in both spaces, (19:29) both in the demand and supply planning space. There's advanced models and there's AI (19:34) being developed.

I think both spaces of the planning area are benefiting from (19:41) these innovations in the IT solution space. Anything specific that you've seen better now, (19:47) that if you would be doing supply planning now, you think, oh, this is a lot easier than (19:52) how I had to do it 10 or 15 years ago? There was a certain amount of planning (19:56) activities that could be done in the systems, of course. I also felt that there was a lot of data (20:02) that sometimes needed to be exported and just managed in Excel to carry out the day-to-day (20:09) planning activities.

And what I can say on a general level is I feel like that is less of a (20:16) need today. I have the impression, at least from my experience with supply planners in Unilever, (20:22) for example, that they're able to do more of their work now in the systems and not having to (20:29) be reliant or pressured to export as much information and do work outside of the (20:34) system in Excel, for example. Finally, moving away from Excel.

Okay, (20:39) sounds good. You mentioned earlier in your role at Ikea, I think it was, that you were also (20:46) responsible for training the younger planners and the other people in the planning team. (20:50) These are not easy systems that you're using.

Any approach that you can share with the audience (20:56) that works particularly well for you to get these people up and running in these systems? (21:00) It's very important to develop close relationships with the planners who are, of course, (21:05) carrying out day-to-day operational work in the IT solutions. You have to stay close to your team. (21:12) You have to stay close to them, see what their questions are, learn what their feedback is, (21:18) learn what works well for them in the systems, where do they have problems, (21:23) where there are potential opportunities to pursue change requests, to develop changes to the system.

(21:31) And what I also did to support training of the planners was I would facilitate trainings. I would (21:38) create training content and materials and run training sessions periodically, at least a couple (21:44) times per year, to just kind of review how are this all the system functionality, how are things (21:49) working, how they can optimally go about their jobs to utilize the systems efficiently and (21:55) effectively. Basically develop close relationships with them and then also you set up specific (22:00) trainings for them.

What was the actual structure? Because you have the software provider, then there (22:07) is you who gets updated on the general training of the system and then you create these kinds of (22:13) manuals and training sessions for your planner specifically. So you're kind of this intermediate (22:18) person who knows everything about the system. I would imagine they also sometimes give trainings, (22:23) right, the system provider, but that happens to you and not to the planners directly.

(22:27) Yeah, in general the IT solution provider is providing information to what would be considered (22:34) the persons working in the business, like me for example, that on the business side with the (22:40) solutions. I had the capacity or a role of being a so-called solution expert for utilizing (22:47) JDA in the team. I was essentially, as you alluded to, that intermediary.

Did you then create (22:53) internal manuals or something? Basically internal training content, mostly PowerPoint presentation (23:01) based materials that would incorporate some screenshots of the systems with descriptions of (23:08) what every module or screen or functionality is for and so forth. (23:13) PowerPoints with screenshots, etc. is one way of bringing it, but in the meantime you have (23:17) other ways that sometimes are tried out with these web-based trainings.

The click trainings (23:22) where people need to click on a screen and then move to the next step and it's (23:26) really almost on a toddler level explaining what you need to do. Did you have experience with (23:31) those kind of trainings and then you see a difference in how people pick up how to use (23:36) the system or it was mainly the PowerPoint training? Great question. There was heavy (23:41) emphasis on utilizing those materials particularly for onboarding a new person joining the team.

(23:48) You may have to train or teach on the system from ground up. I still found it value-added (23:53) important to provide in-person more customized trainings. Moving to even more people-related (24:01) topic and culture.

I think you're the first U.S.-based guest in our podcast. You have a (24:07) unique perspective, at least for us, about cultural differences in between Europe where (24:12) you worked as well with IKEA and the U.S. How did that feel for you working in those (24:17) different environments? Was there something that really stood out? I would say in the U.S., (24:22) you're working in a supply chain job in American culture. There's probably a little bit more (24:28) heavier emphasis on results, on KPIs.

What was a little bit more unique for me for working in (24:36) Europe is while KPIs and results were still important in Europe as well, there was more (24:42) interest in people getting to know each other a little bit more on a personal level. So the (24:48) relational part was also important? Yeah, there was a little more time and resources allocated (24:54) to the relational aspect. I can cite a specific example to this.

For example, (24:59) when I started working for IKEA in North America's regional supply office, once a year, (25:06) the company would literally fly all the planners to Sweden to meet the global teams. We would spend (25:14) a few days in Sweden on an annual basis to interact and go over what are global priorities (25:21) for the business? What is the latest tactical and strategic goals in supply chain and whatnot? (25:28) There would be some time allocated to just get to know people that you're working with in offices (25:34) who are located in different countries. That was new for me.

I do believe that does happen (25:40) in U.S. companies as well, but I don't have the experience of feeling that that was as common. (25:45) In Sweden, they have this word hygge or hygge. I don't know how to pronounce it exactly, (25:51) but it's this coziness kind of description.

So I can imagine that they want to bring people (25:56) together also. Also, did you notice any differences in process or any non-supply (26:03) chain specific differences between U.S. and Europe? Or did you find there that knowledge (26:08) and systems were pretty similar? I think the knowledge, the work is very similar. The (26:13) capabilities, of course, very similar.

I don't think that one side had any kind of (26:18) advantage over the other. I think from a cultural perspective. (26:24) Our favorite ending question is, are there any recommendations that you would give to other (26:29) people in the supply planning or general planning world? The best advice that I can give and is (26:35) particularly for any younger professionals or people that are new to supply chain or still (26:41) relatively young in their supply chain career is, I strongly advocate pursuing trainings, (26:47) whether they be via an external provider or whether they're provided internally within (26:54) the company.

I can say, for example, I did attend an SNOP training once with an institution called (27:00) Oliver White. I found the training very beneficial. Do you have any other people that you follow, (27:06) try to read their articles or posts somewhere? I'm a very active person on LinkedIn.

Nicholas (27:12) Vandepute is one person who I enjoy his content that he puts on LinkedIn. I strongly recommend (27:18) to join supply chain focus groups on LinkedIn and discover the persons that you're most interested (27:25) to follow and connect with. If people have any questions for you, can they also reach out to you (27:31) via LinkedIn? No problem at all.

There are people that occasionally do reach out to me (27:36) and I do try to make an effort to actively respond and engage with other supply chain (27:42) professionals out there. Okay, that's great to hear. Then we'll share your profile with (27:47) the listeners out there.

Thanks a lot, George, for sharing all of your experience and to the (27:51) listeners out there. Until next time. Bye-bye.

Thanks, George. Thank you. Bye-bye.