Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Inside SAP Cloud ERP

Inside SAP Cloud ERP artwork

Episode 95: The Power of Advanced Variant Configuration with SAP S/4HANA

Nov 3, 202324 minAudio
ShareXLinkedInFacebookWhatsAppEmail

Markus Oertelt hosts this episode of Inside SAP S/4HANA, which focuses on the power of Advanced Variant Configuration solution. Marin Ukalovic, chief product owner for Digital Configuration Lifecycle at SAP and co-author of “Advanced Variant Configuration with SAP S/4HANA” and more books, is the guest and describes the functionalities of the solution, best practices approach, possibilities in the light of user experience and the tasks involving artificial intelligence. He also shares a glimpse of what’s on the roadmap for Advanced Variant Configuration. Show Notes: Download transcript as PDF file

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

Hi everyone and welcome to our podcast Inside SAPS or HANA. There's no customer success with our product and project success. In this episode we talk about the topic of variant configuration. And to answer all your questions around this topic, I have Maureen Okalovich, Chief Product Owner for Digital Configuration Life Cycle here at SAP on the show today Marine. Welcome to this podcast. Hi, Marcus. Thank you for having me, Marine.

The pleasure is also on my side. You are renowned author of SAP Press books and an engineer at heart. What gets you into the subject of variant configuration? Yeah, that's a long story, Marcus. Before my SAP time I was an engineer, so a mechanical engineer studied that. And when I joined company, which was end of the 90s in the last Millennium, my mentor was run this topic. He handed it over to me from the early days. So I started from the beginning of my SAP career with this topic.

And since I never got into this, now I know everything mode, I'm still with it and I'm continuing the topic now in a different role. I'm not longer in the sales area and presets area, I'm now in the Product Development Product Management role and I'm still happy with it. Great to hear Maureen. Here's a little secret of myself. My second job at SAP was also in free cells, but let's focus on the today's topic and the Variant Configurator. Advanced Variant Configurator is

a direct successor to the LOVC application component. And can you maybe explain why the SAP Advanced Variant Configuration solution was introduced and what makes it stronger in comparison to the old LOVC solution actually? Good question. So you know when we started or when SAP started the overall Espana story, we started also to investigate into very configuration to LVC how to move it to S4 and if it makes sense or now and we spent close to a

year together with well experienced partners and customers, it just makes sense and after this year. We saw that it's it's not the best way to do it and we decided together with the huge investment case from border area to invest in a completely new from scratch based G code based engine together with the Fraunhofer Institute. And the reason was yeah, all the LVC was old, was not integrated that good, was not the best engine and when we started doing

it, we of course focused on several areas. The one is. The user itself, the better guidance for sales user, for example, usability, how to use it. We are Fiori now, pure Fiori UI. We have filters for him to make the configuration easier. We have undo, redo functions. As you know it's from Microsoft applications in Excel or Word or PowerPoint. So the the guy who was using it was in focus of course. And then, most important, even more important, the modeler.

The guy who's. Building the product which needs to be configured so we you know. We have two user types. The one is standard user, the one is the professional user. The professional user is the guy building the product. We used to him a new trace, much much more better than the old 1A. New inspector is showing every object he wants to change. A new syntax is possible. Now we have new tables you can enter wild cards, negative tables. You don't need to maintain everything what's possibly could. Focus on the minority, meaning what is not possible and then

let the engine run. And of course we focused on the business processes behind barrier configuration steps which would help the the user to get faster than the result. For example, creating material variance is now possible via one click in the sales order. Can say, hey this is often sold, let's make a material variant, it's just one click. You can start with configure to order and then move also by one click into the engineer to order process. So if you find out that the customer's asking. Or something which is not possible in the current model.

You can then start the ETO process hand over to engineers, then they build something on bomb level or routing level that switch back. We have integration to planning advanced ATP product allocations running. We have now quite this year or what introduced this year a solution order and solution quote integration. So we offer capabilities which run into the solution business meaning. Configuring product, services, subscription and other objects combined into one.

We are part of the 2G scenario for customers running As for private cloud and public cloud and hand over objects here and we have for example cross plan procurement also for configured products. So a lot more than that. What I just showed were explained as said, it's enough to fill a book. Thank you, Marlene. That sounds great. I was able to test the new TRACE functionality as well and I have to agree, it's awesome. But let's move away maybe from the technical aspects, the

improvements we had on that side and maybe we go more into the business world and hear what I see in the market that organizations are currently facing trade-offs between standardization and also customer individuality. So last time I configured a car and I was able to choose a tremendous variety for a combustion engine car, but for an electric arm, my options were very limited. Where do you see the market going in terms of varying configuration? For me it's not clear and

probably for you, but do you also see an artificial intelligence or machine learning business cases in certain industries popping up or coming along? Yes indeed. Just an example, Recently I visited Daimler. In Sindufingen and the new Factory 56 where they built the S class and the Mybach and we had to end up in the whole several sessions regarding AI and machine learning integration into supply chain processes invited by a group of people called Operations Research

driven by universities in Stuttgart and Munich was really, nice to see. Really, really experienced people were intelligent people. And what came into my mind when you asked this question is that. Especially this topic was mentioned by OEM. How can we change, how can the sender change the behavior, the buying experience, the behavior in these times. And he especially mentioned COVID because a lot of things changed during COVID. You know, we had the impact in the spy chain. Not everything was deliverable. They had lack of material.

They needed to tell the customer, hey, you cannot get what you want, you need to wait, maybe you will not get it. And then they found out that they maybe need to change its strategy. For example, especially regarding like 2 cars, they now implement this more and more is they move away from this huge variety you just mentioned. When you can click on everything and start focusing, start guiding, create selling bundles as you know the package accepts on whatever you know bundles of take this and this and this. The business bundle we also know from SAP Cars.

Reduce complexity is really important. Not everybody is longer interested in having these huge sessions and configurations. Nobody wants to configure by the way. And moving away from just selling cars, meaning in numbers, sell as much as you can, moving towards, sell the margin related features, sell the more expensive parts, guide the customer towards what you want to sell, not what he wants to buy. And do this so intelligent that he doesn't recognize the way you're doing. So this is what I learned and it was really impressing because it was not only one company, was

two and three companies telling exactly the same thing is how to use AI. Because you of course you can write rules and say if you click on this then takes this, if you take red and choose large whatever. But use AI technologies to find out how the customer will end up with the order when he started in the specific way. And this is something where we need AI support. And machine learning. And this is where we also start investigating. So in in my case, the machine already knows what kind of car will configure in the future,

right? Yes, you should. Start with the biggest end and who will end up in the exports car, right? Great. When I talk to customers that use configurable materials, I usually hear that life cycle management is is very important to them. It's the first point that usually comes or questions in that direction and here I'm talking about changes for example to characteristics or attributes, rules, prices and as well as parts list or routing

structures within an existing data model. What is your advice on a best practice approach to handle this so that organizations stay or become actually competitive? Yeah, very good. As you may be noticing how you introduced me, you know my title is called Digital Configuration Life cycle and not very configuration or advanced very configuration. The reason behind we have an investment case called exactly the name digital configuration life cycle is the importance of this viewpoint. You mentioned. We're not talking about the

Configurator, we're not talking about the specific way an engine runs. For example, sales configuration engine does sales configuration. So he sells something, he may be derives the right pricing and that's it, right. Then you have a list of characteristic values and and that's it. You could hand over the printed out document to production. Safeties produce. A bomb, exploding engine, explode bombs and you have a huge Max bomb for example and you have rules behind every item and then based on the configuration engine result you have a bomb for your specific car for example.

Calculation of routings is made also in the so-called low level area. And what's important and I think this is the differentiator of SAP to other competitors which focus of 1 level what I just mentioned, for example, sales. Is that we connect each other and that's really crucial to see the topic very configuration as in process, not as an engine. Of course you need an engine, but more important and more value is coming out. If you understand that this is a

process which starts from the very first idea, you want to create a new product, a new feature, a new option, introduce a new pricing, introduce a Bundy combination, whatever. This first idea needs to be incorporated. For example by requirements management into the product model. Then you need to take care about the product, the material master, you need to add the description, you need to have pricing, you need to have bonds, you need to have a routing costs, you have marketing. So it's it's all a life cycle and then you will decide of

course this product is no longer available so you need to to yeah set it up. This the reason why we call since I joined this IP. I think this is really old. This this mantra is model once configured anywhere. See it as a process, model it one time but add everything. Every user, every consumer needs to run just life cycle process and not to calculate an engine great marine. I like your mantra and model once configure anywhere and I think that my next question goes a little bit also in that

direction. We all know that inefficiency or inefficient offers lead to lost customers and lost revenue. And personally, when I want to receive an offer, I configure products on a website or even an app and sometimes I get frustrated as the UX is horrible or the configuration is not working properly, whatever it is. And so how or what can SAP do to make the quantation configuration process a very good user experience for our customers? Yeah, coming back to this model

once configure anyway and I like it. Also this mantra, this concept. Yes, of course we. If possible, and I would recommend doing this, do it one time, use the same data all over because then you will get rid of any error and you will not sell something which cannot be produced and you will not offer something which is has a bigger cost in revenue. But of course it is super important that not that there's that one UI, not one UX. It always needs to be targeted towards the user of the

application. So if you have a back end user. And this guy wants to run really, really fast through a configuration. He wants to have fast data injury, for example, he wants to be able to switch immediately between CTO and ETO Stream wants to incorporate the engineer into the whole process and so back and forth. And he needs a special UI for that. So this is an S4. But if you are a sales user for example, and you just do sales and you have maybe CPQ application then you run you would like to run a CPQ or a configure price code oriented tool.

And UI running on the same master data, but having a different UI, different behavior. And even more important what you mentioned is if you are an end customer, AB2B consumer on the commerce platform have no clue about the product itself. You don't know which rules are firing, you don't know how many characteristic values you need to enter, You need to have a guidance. So what I want to say is, even having one model, it's super important not to have the same UI for every user. You need to address avenues of differently and this is I think

a task we solve with our different offerings in this life cycle business. They have worked with a client in the past and they have seen that they have a very complex, very configuration set up, mainly self developed for many, many years. So that means they have a multi level bill of materials with our configure to the yeah 25th or 26th level even. And here the challenge was that often customers were asking for changes in configuration to the

last moment. So the saying in the company there was till the machine is actually on the track and what do you think is the best approach? I mean, we have seen the AI case you described before where you go sort of like into the product recommendation area. But what would be your advice, especially for this company if they say they cannot reduce the complexity? How would you handle the changes in varying configurations? You know this is from my perspective typical German

statement. Yeah, we change, we change the piece on the track even if the foil is on it and it's, it's colored and and the customer calls, I want to have it differently. It's not easy to answer to be honest. I know because I'm that long with the company and and the systems even in there are three times we have. Solutions helping to solve this issue. It was called so-called Order change management, meaning simulating a change, see what the result of this change would be and then decide, maybe by rules or maybe by human interaction. If you go with the change out of the sales order, the standard

behavior is still the same, meaning that before you save or after saving the configuration and sales, you will open plan orders right with with. The subsequent documents and objects like bombs and routings and planned orders and in the moment you switch them to production orders and you fixate them, then they are fixed, right? Until then a change is not an issue. After that point it's getting hard and it's said not so easy to to answer. So I would say if you have a specific customer in mind, we should start talking maybe in a

different meeting to find a way how we can solve that. And maybe AI or machine learning could help because. I have a case in my mind where machine learning has helped to to explode bombs differently and routings and calculate routings. And maybe this is a chance to change last minute stuff, but it's independent from configuration. By the way, it is always hard to change. When you started producing, I can tell you how they developed it on their own, how they solved it, but painfully.

But let's come back maybe to, yeah, the SAP product recommendation product we just released here at SAP. And yeah, do you foresee other use cases for artificial intelligence, maybe also in the field of varying configuration? You already talked about a couple, but maybe you can also touch more on the SAP product recommendation, also explain that a little bit because I don't know if everybody out there knows what we really mean with that. Yeah, that's the first one which we release right now. And absolutely, yes, we all heard on an approach from SAP,

we are in AI. And I see a lot of opportunities, especially because of this complex process, learning configuration to help customers and just to name a few, which we started doing by mock ups and and proof concepts is maintenance rules right? Nobody wants to maintain rules. Why not asking AI or ChatGPT and say hey I want to have a rule making this and that and it works. Machine learning based calculations in the bomb areas of routings. We have partners already

delivered solutions to calculate routing times based on machine learning. Because you don't have time to to make a fast quote and the customer doesn't read the 4th quote, the first one who's sending the quote and it needs to be really accurate. And if you if you have a complex product you would need to have a lot of calculations Machine and can help here. So it's a lot of things we're starting to investigate and product recommendations. Also to find out the right way through the configuration to get you to the right product based on machine learning experiences,

right. You can as set you can do a lot of things with rules, but you will end up in huge lines of coding. And maybe you're going to the wrong direction and why not asking the experience? Machine learning is nothing else an experience is it? It's like you would ask 1000 people what they experienced the last five years, but now you have an engine giving an answer in a second, right? Or milliseconds. So we see a lot of opportunities, especially in variant configuration, in back end, low level, high level commerce integration, CPQ, even in 3D visualization.

How to visualize something? A lot of things are ongoing and I hope that we will deliver more in 2024 which fits into the all in approach with SAP and AI. Interesting times and let's stay with the times. What are your plans or SA PS plans for for the future road map for advanced varying configuration or other configuration products we offer? Yeah, that's of course a long list. I just had today a meeting. Because we started portfolio planning for 24 O 8 for next year, so just name a fuse.

We have besides the mentioned AI which is 1 hot topic right now to find applications use cases which we could make to a product or include into products. Is we open S4? Right, We we make it accessible. So we have a lot of Rappios ongoing. So you can write and read data like classes, characteristics, values and and rules and back and forth. So we open us towards partners which build applications on BTP. We are working at the UI again

the Fuel UI of course has had limitations to include things. Now we we offer changes that you can include pictures or icons make it more visual in the standard. We are working on process integration topics like the multi level set scenario. If you have a set meaning 3-4 components based on one configuration as an extract you can now do this multi level which wasn't possible in LVC. So we have now we're now setting the stage towards processes which haven't been there before, S4 and ADC because we don't want

to do exactly the same what we had in LVC. We are targeting every release some pieces of new syntax. And the reason behind this approach is that we have a lot of customers which want to move from LOVCERP to S4 ABC and they tell us say we had a lot of developments because we were missing this and this and this and so we'll build it on our own just name one that's called P functions which we don't support anymore and they say it's hard for us to move if we can't survive these P functions time and we would like to get back to

standard. So we asked them what do you need and if maybe syntax is missing, we deliver and this is what we do continuously. Then this bundling app will now ship in the in the beginning of next year is to offer a customer the possibility to bundle product, service subscription offerings, service orders into a solution quote solution or then don't use the classical modelling in the back end because it's an expert level. And the bundling guys are most often in marketing and they don't want to go that deep into

the topic of modelling. So we offer them now a bunting app based on that BTP which is easy to learn. You just need an hour to understand how it works and you can create a bundle and then sell solutions to your customers. So the road map itself is a huge list of topics. We move from year to year and we pick based on, yeah, different activities outside the market. We listen still to the customer Co innovation group where we are working together with them partners and customers to target our customer to ask them, hey, what do you think about this

idea? What do you think about that idea? Would you implement this? If we deliver so a ton of ideas, we could spend an hour talking about Rd. Maps. I would like to talk with you for hours. A lot of information for our listeners today, Marine. Unfortunately, yeah, we are already at the end. Marine, thanks for being on the show today. It was a real pleasure having you. Thank you for having me and if I could add an information which is important for the community and for me personally and for my colleagues is as you stated at the beginning.

We are our authors or co-authors of books and one of them is the SAP Advanced Parent configuration Esfahana which was released in the summer this year or in March, April, which was in German. It was by the way #1 on SAP Press best seller list in the German version and it's still in the top ten as far as I know. And we released the English version just two weeks ago. And it's also I I had a configuration work conference with 100 customers and every

book I had with me was sold out immediately. So if somebody wants a signed version, let me know. I think I have only one or two left. And if one of the listeners has interest, yeah, just write an e-mail to me marin.column@sap.com. Brilliant, brilliant and great. Marlin, Congratulations. I will bring my book to Valdorf and then maybe if you're there, you can also sign it. And for me, Yeah, of course. You're welcome. OK. Thank you very much. Thanks to you again and all the best to you. And as always, feel free to drop

us an e-mail via inside S4 at sap.com and let us know which topics are of your interest. TuneIn next time and be inside SAPS Fahana.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. TechEd #10: How Joule Studio and n8n Orchestrates Enterprise AI

    Nov 20, 202616 minAudio

    Recorded live at SAP TechEd 2026 in Berlin, this episode of Inside SAP Cloud ERP gets into one of the most significant partnerships in the SAP ecosystem right now. Host Yannick speaks with Vladimir Shakhov, Strategic...

    • Apple Podcasts

    Credit the people on this episode

  2. TechEd#9: How AI Transforms Cloud ERP Migrations

    Nov 18, 202614 minAudio

    Recorded live at SAP TechEd 2026 in Berlin, this episode takes a refreshingly grounded, practitioner-first look at what AI in SAP really means in the field today. Host Fernanda speaks with Wouter Lemaire, independent...

    • Apple Podcasts

    Credit the people on this episode

  3. TechEd #8: The Benefit of Laying the Groundwork for AI

    Nov 16, 202615 minAudio

    Recorded live at SAP TechEd 2026 in Berlin, this episode of Inside SAP Cloud ERP brings something the series hasn't had yet: the perspective of an actual large enterprise customer architect, not a consultant or partner....

    • Apple Podcasts

    Credit the people on this episode

  4. TechEd #7: The Hidden Risk in Every SAP Transformation — Organizational Change Management

    Nov 13, 202618 minAudio

    Recorded live at SAP TechEd, this episode of Inside SAP Cloud ERP zooms in on the dimension of SAP transformations that gets the least attention - and causes the most damage when ignored: Organizational Change...

    • Apple Podcasts

    Credit the people on this episode

  5. TechEd #6: Five Days to a Working AI Solution — Inside SAP's Forward Deployed Engineering Model

    Nov 11, 202623 minAudio

    Recorded live at SAP TechEd 2026 in Berlin, this episode of Inside SAP Cloud ERP gets to the heart of what the autonomous enterprise actually looks like when it's working. Host Yannick speaks with Sindhu Gangadharan,...

    • Apple Podcasts

    Credit the people on this episode