PI Congress video: Things to consider when considering PLM

March 25, 2015


I attended PI Congress in Dusseldorf last month. During the conference I had a chance to share my thoughts about PLM selection process. PLM landscape is changing these days. There are multiple opportunities to get PLM differently these days. Even some of my colleagues in PLM analytic community can call the complexity and cost of PLM implementation urban legend, I still believe many companies are asking how to develop right PLM strategy and implementation approach.

If you ask a company what they believe is the most important area of investment when planning and implementing PLM you are likely to get many answers. You can go between choosing the right vendor, setting up right PLM implementation team. You can hear about importance of cultural change management and many others. The truth is, there is no right answer. PLM is different and your company is an important factory you need to consider when selecting PLM system and strategy.

These are topics I covered my short commentary. .

1- What is the biggest force to change existing PLM paradigms.

PLM implementation is still considered as a complex task, which takes time, resources and cost. I believe, new paradigms should changed that. It is like waterfall development process. Agile development methods came to change it. The same with PLM. New paradigms – cloud, agile, etc. are coming to make a change in PLM.

2- What should every company thinking before taking first steps in PLM implementation.

Take a look in a mirror. You should learn your company and development processes. Company should understand how current processes are working. It sounds simple, but many companies are missing that. Without that PLM implementation doesn’t make sense – you will just automate bad processes.

3- How is vendor competition is changing PLM landscape

PLM space is dominated by  few very large vendors. It is a very competitive space with high barrier to entrance. Nevertheless, companies are looking for alternatives and I can see few interesting entrants into PLM space. Aras had fresh approach with enterprise open source. Aras claims “different PLM” with model based SOA platform and new business model. Autodesk came into PLM back in 2012 with cloud PLM. It was a change. These days cloud is transforming all vendors. This is probably the last big changing force in the industry.

4- What is the common misconception regarding PLM

The biggest misconception is that PLM can solve the problem of bad product development processes. PLM as a tool  and technology. It cannot solve bad organization process. Only company and people can do so.

5- Where do PLM implementation go wrong

There are few typical situations when PLM implementation process can go wrong. Some of them are generic and can fail any IT project. However, some of them are more specific. Here is my list – wrong expectations, no C-level support, no plan, integration between new PLM system and other systems wasn’t planned, company is planning to force a change in a short period of time.

What is my conclusion? Take a look in the mirror before starting PLM project. The PLM technology and product itself is not a silver bullet. You need to have deep understanding of your organization and its processes. Only by doing that you can develop right PLM strategy and manage PLM implementation step by step. It will help you to avoid PLM experience that is complicated and miserable. Just my thoughts.

Best, Oleg

PLM has a weak point in cloud-based supply chain

March 24, 2015


My attention was caught by Manufacturing Trends to watch in 2015 article written by Jeff Moad at Manufacturing Leadership Community. I missed that writing few months ago when it was published. I found one of them very interesting – The Rise of Cloud-based Supply Chain. Here is a passage from the article:

The Rise of Cloud-based Supply Chains. As the manufacturing landscape becomes more interconnected and interdependent, requiring close cooperative links with multiple supply chain partners in multiple locations for materials, parts production and the support of new multi-channel services, companies will increasingly adopt cloud and more predictive web-based supply chain software to help manage and swiftly reconfigure their networks to gain real-time visibility, cut time-to-market, and respond faster to customer changes and potentially disruptive political and natural risks.

It made me think again about new enterprise software reality for PLM vendors and changes in manufacturing eco-system. The interconnected manufacturing landscape is a key. It gives an interesting opportunity for software vendors thinking about cloud software as a platform, rather than a bunch of servers hosted elsewhere. At the same time, it raises many questions about how new generation of enterprise software will handle modern people and organization paradigm. One of the challenges for many PLM products and platforms is related to their ability to manage multiple organizations in distributed networks. Which can be a weak point for many of them to capture cloud-based supply chain opportunity.

What is my conclusion? Modern PLM software can embrace new paradigm of interconnected and interdependent manufacturing environment. This is quite different from traditional environments of OEMs and suppliers. The ability to manage distributed processes will become critical and can be one of the future differentiators for some PLM vendors. It looks like born in the cloud PLM technologies can gain some advantages here. Just my thoughts…

Best, Oleg

Image courtesy of cooldesign at FreeDigitalPhotos.net

PLM and VC firms are facing the same problem – differentiation

March 21, 2015


Platform is such a sweet word. We love platforms. In the past, the ultimate dream of every software developer was to build a platform. Today, word “platform” is applying to so many things. Uber is a platform for transportation and Airbnb is a platform for hospitality. You can bring some of your favorite examples here. Guess what PLM vendors are developing? Yes… PLM platforms. And these platforms are supposed to provide a solution for companies to build things. CIMdata, one of the largest analytic companies in PLM domain recently published a study about PLM platformization. The main point – PLM business is achieving a status of business platform. The things on a technology side are getting interesting too. I shared some of my thoughts about tech aspects of PLM platforms – Future PLM platforms: between a rock and hard place.

I had a great lunch reading earlier today – Are Newly Formed Roles In VC Firms Differentiators, Table Stakes Or Total BS? You can ask me how VC firms business is connected to PLM platforms? Here is the thing… My attention was caught by a word “differentiators”. The article speaks about VC firms building “platforms” to differentiate themselves among entrepreneurs. Interesting turn. Here is my favorite passage from the article.

So why is this happening? And more importantly, if the trend continues, how do we ensure this actually helps entrepreneurs? The answers all depend on how much VCs are willing to admit one problem: We all sell the same product. As a VC, you don’t want your platform to sound like every other competitor firm plus every incubator, accelerator, community group and graduate program in your city.

In my past roles in media and marketing, we relied on the “white space” in an industry to own something outright and ultimately build or market a product more successfully. But in VC, as one founder told me, everyone’s money is just as green. The best way I can describe the challenges facing all platform directors is that we each run a bakery on a street full of bakeries that all sell one thing: plain bagels. There are only so many ways to make a plain bagel sound different from another plain bagel.

Differentiators. I think this is a very important things and PLM companies are chasing that too. Nobody wants to sell “plain bagels” only. In my view, PLM industry has some challenges here. After all innovations and technologies that happened with PLM industry, there are lot of similarities in all PLM products these days. Dig inside of every PLM implementation. You can find it surprisingly similar. It comes down to five main components – (1) Design CAD data management; (2) BOM and change management; (3) Projects; (4) Portfolio and configurations; (5) Quality.

I can hear voices of PLM vendors saying that their platforms are different. Yes, they are. Different technologies were created in a different time. New user interface is nicer than one that was created 15 years ago. New administration tools are simpler. Cloud was a differentiation factor for some PLM vendors over the past few years, but these days cloud is everywhere. PLM companies are innovating in open source, business models and marketing. But, at the end of the day, five domains I mentioned above are representing problems customers are looking how to solve. And most of solutions related to these specific problems are very similar.

In the past, company built PDM and PLM platforms to solve a problem of complexity in product data. It was a layer on top of database to provide a solution how to store complex data, managing access and work together (collaborate). Modern world is changing fast. Manufacturing is very different theses days. Take a look on my blog – PLM and changes in manufacturing eco-system. How to build a different platform that can solve problems of manufacturing in 21st century?

What is my conclusion? This is a time for PLM vendors to think outside the box of known PLM solutions. Today’s CAD and PLM tools won’t become future platforms. Today’s PLM platforms were created for large OEM companies in aerospace, automotive and defense industries. Manufacturing world is changing. Large OEMs are changing too. How to build a different PLM platform? What will play a key differentiation role – cloud, big data, openness, business models, circular economy, anything else? These are questions to ask today. Just my thoughts…

Best, Oleg

New manufacturing business models will influence PLM software

March 19, 2015


Things are moving fast these days. Vendors are speeding up their plans to catch up with new development. The changes are coming from both technological and business sides. Today, I want to focus more on the side of business. Few years ago, I put some of my thoughts in the blog post – Faltered licenses and future business models. Fast forward from 2010 into 2015… there are lot changes in the way software companies are selling their software. Actually, many of them are not selling software anymore. Subscriptions became a reality.

The following article caught my attention earlier this week – Keurig accidentally created the perfect business model for hardware startups by Bolt (venture capital firm focusing on hardware). Take a read – there are some great examples and comparisons of business models in the article. Here are few of my favorite passages:

Recurring revenue matters because it fundamentally changes your business. There are good reasons investors are averse to hardware but love software. One of the leading reasons revolves around future revenue. Investors pay huge premiums to own stock in companies betting on the likelihood that future revenue will be drastically larger than current revenue. If you’re in a traditional hardware business, future revenue is confined to cyclic product sales. This roughly means you get one shot at revenue with each customer per product development cycle: each sale must be painfully acquired by building a new product every 18 months or so.

This is where the brilliance of the Keurig model shines. The initial sale of a $120 Keurig brewer isn’t that difficult or costly. Keurig doesn’t spend a lot on marketing or advertising and the product isn’t complex to manufacture or service. In my rough estimation, the BOM for a brewer is around $40, giving Keurig about a 25% gross margin on the product. Time from PO to FOB is likely less than 2 months, yet high-margin K-cup sales start immediately and continue for years. Keurig spends less than $0.015 on each K-cup and charges 100% more per unit than bagged, ground coffee. Yet few people complain about this cost.


The big thing is recurrent revenues. It is much easy to sell something to your customers. Especially, if the price is a comfortable zone. This trend is going up these days.

It made me think more about business models of engineering and manufacturing software. PLM is included. And PLM was guilty many years for very “uncomfortable” price points. Which in many situations lead to wrong set of expectations and failure of PLM adoption.

I can see a trend among software vendors today to change that. It might be a good question how to bring PLM price point to “comfortable zone” and connect it to the use and adoption of software. The more you use it- the more revenues you will be able to get from customer. On the other side, usage of software means value for customer. Some sort of win-win situations.

What is my conclusion? Business models revolution is coming to enterprise software. In the next few years, we are going to see many examples of innovation in engineering and manufacturing software business models. Freemium and subscription is only beginning. The latest example of Onshape, which is inventing pay for privacy model. Watch that space for innovation. Just my thoughts…

Best, Oleg

Picture credit Keurig and Bolt article

Image courtesy of ddpavumba at FreeDigitalPhotos.net

Why PLM should revise NPI products?

March 17, 2015


One of the challenges most of manufacturing companies are facing these days is related to ability to introduce new product to the market. Think about manufacturers of electronic gadgets that can miss holidays sales or supplier that misses deliveries on new components for OEM manufacturer. The cost of missed timeline sometimes is so high that it cannot be recovered when you development is going off schedule. Therefore, New Product Introduction (NPI) is one of the important elements of PLM strategies for many companies. These type of solutions are provided by most of PLM vendors and usually part of their Program and Project Management portfolio. You can follow few links to see examples – Aras PLM NPI, Autodesk PLM360 NPI, ENOVIA NPI, Siemens PLM NPI. I’m sure you can Google “PLM NPI” and fine more.

What caught my attention is that all NPI / NPDI products and methodologies are focusing on organizing of waterfall stage-gate process to follow up in order to achieve product development goals. Typically you can see it as a sequence of Concept, Development, Testing, Launch. The details can vary as you can see in few pictures below, but the essence of NPI process is the same.

Aras / Infor


Autodesk PLM 360


It occurred to me that traditional PLM NPI is a bit outdated in a modern world of lean and agile product development methods. Agile product development is focusing on a series of short development cycles (sprints) and it became hugely popular and successful in software development of the past decade. It goes opposite to something that we familiar as waterfall process. You can read more by learning about Scrum, Kanban, XP. The ideas of agile and lean are coming to manufacturing too. Last year, I shared some of my thoughts about that in a blog – Who will make PLM for eXtreme Manufacturing?

Here is the thing… All PLM products are focusing on NPI and project management implementation actually thinking about traditional NPI model. They are mostly focused on a workflow and waterfall process. Waterfall processes are dying in a modern development world. At the same it is a mainstream organization large companies are using for product development. But… for how long? Large companies are looking how to innovate and to work differently. At the same time, they are providing very little support for teams that want to follow agile methodologies.

So, what is wrong with waterfall NPI process? Below, I put some of ideas why waterfall NPI process can give you very little help in development of innovative product together with customers.

1- I know what customer wants.

2- I know what features to build.

3- Too much focus on launch date and stage gates.

4- No trial, no errors process.

What is my conclusion? Customer agile development can be a new way for PLM to support manufacturing companies. It is can be more efficient to innovate in small teams. It can bring new agile methods and change product development process. It can replace existing NPI with complexity of waterfall processes. Just my thoughts…

Best, Oleg

Image courtesy of ArtJSan at FreeDigitalPhotos.net


5 Attitudes that can kill PLM projects

March 16, 2015


In my experience, I can separate all manufacturing into two large categories with regards to what they want to do with PLM: those who don’t want to do anything with PLM and those who desperately want to implement that, but don’t know where to start. The border between them is blurred. Sometimes you can see two different groups of people in a company taking opposite opinions about how to deal (or not) with PLM.

I’ve been thinking about how company can build a right attitude to take on PLM projects. You can find many blog posts and analyst materials explaining how to implement PLM and not fail. I decided to try an opposite approach. Regardless on where any specific engineering or manufacturing company falls in the spectrum of their intent towards PLM project, I can identify several attitudes that can literally kill you PLM projects. Here are my selection of 5 attitudes:

#1 We don’t need PLM. We can do it differently.

The size of a company doesn’t matter. I’ve seen it in large and small companies. Usually, it happens with companies having established processes and/or systems. They are confident that they can manage everything they need using existing systems (CAD, Excel, ERP, CRM, etc.) and PLM as a concept is something artificial. My recipe to such type of company is to look for inefficiency of their processes and learn more about PLM. With some learning they can find a match between their problems and potential PLM systems can bring.

#2- This is only for big companies

A typical mindset of SME (small to medium) companies that afraid to see how many problems can be solved by organizing data about product and product development processes in a better way. Most of PLM companies are not spending enough educating small customers about how flexible and diversified approach in PLM can improve what they do.

#3- We want OOTB (Out-of-the-box)

This is a situation when customer wants vendor or service partner to solve their internal problems by bringing ready-to-be-used PLM systems. However, without learning about your own company, any PLM implementation will be a suicide. You can use OOTB as a starting point, but in no way to deploy system and change your company to operate accordingly. Such attitude usually ends up by abandoning of OOTB implementation and switching back to old processes.

4- We can do it as a side project

Another dangerous situation. Some people in a company can consider that PLM is an isolated project, which can be disconnected from rest of the company. As a result you can get PLM software installed and used for a specific group of people. It is obviously kills PLM value.s

5- It is for engineers only

A subset of previous attitude, which assumes PLM is only important for engineers. This approach usually downgrade PLM to basic set of CAD data management functions. Domains like BOM management, Change management, Quality, Manufacturing planning are going out PLM radar and it takes down overall PLM value for a company.

What is my conclusion? Think about PLM as a strategic approach to change the way you run product development. It helps to identify biggest pain points and find how to plan PLM implementation in a holistic way, but prioritizing on specific problems. It helps to build an overall plan, to discuss with people in all department, but build a gradual implementation. Just my thoughts…

Best, Oleg

How to break limits of existing PLM architectures

March 13, 2015


The conventional opinion of many people in PLM domain is that technology is not a main problem in PLM industry. At the same time, PLM vendors having significant challenges to convince customers to adopt new versions of their products. Manufacturing companies are replacing PLM platforms every 10 years (some people can even come with 20 years benchmark). When you think about technological improvement, even ten years is a number which can put any manufacturing company back in dinosaur era in terms of what tech they are using.

My attention was caught by article The Past and Future of Systems Management written by Ben Horowitz. Take some of your time during the weekend and read the article. If you have more time, I can recommend you Ben’s book – The hard thing about hard things. I found Ben’s insight about new cloud based architecture very important to understand for future development of PLM products. According to that, traditional systems would not work for modern, massive, cloud-based architectures. In fact, they would not work properly for cloud-based architectures of any scale. One of the most interesting points I captured is related to the move of system architecture “from servers to services” and the fact applications are now collection of micro services.

Traditional systems are server centric — Even relatively modern systems management products like New Relic treat servers as sacred resources which must be kept alive, but Facebook loses servers every day and it doesn’t matter. Facebook doesn’t care about servers; they care about services. Knowing when a cluster of services that provides, for example, an identity service is out of capacity is critical, but getting paged in the middle of the night because you lost one server in a cluster of 20 is asinine. Applications are now a collection of micro-services — These micro services are often managed by separate teams with all sorts of upstream and downstream dependencies. Having a solution that tracks all the relevant metrics across all the services fosters a much more collaborative environment where teams can communicate with one another (versus logs, where only the developer who wrote the app can really understand what’s going on).

It made me think again about existing PLM technologies and architectures. Most of them are 15-20 years old and they are completely server and database centric. Few years ago, I explained that in my Future of PLM databases article. In my view, the end of single PLM database architecture is coming . The new PLM system architectures can change a way customer can adopt and manage their PLM environments. Here is the idea to think about.


All existing PLM products are developed on top of existing database technological stacks. Nothing wrong with that, but here is a problem – the scale. The amount of data PLM systems have to handle is growing in scale and reach too. Manufacturing companies are dependent on significant amount of information originated and maintained outside of organization – product catalogs, supplier and other reference information. In addition to that, in many situations, the data is owned by multiple companies – not a single OEM. How traditional PLM platforms will handle that?

What is my conclusion? The conventional wisdom of PLM architectures and implementation is to put information in a single database. It must change. Modern engineering and manufacturing environments are different. It is more likely network of sacred resources rather than single PLM database. New product architectures and technologies should come to handle that. Just my thoughts…

Best, Oleg

Image courtesy of ddpavumba at FreeDigitalPhotos.net



Get every new post delivered to your Inbox.

Join 273 other followers