How To Fix PLM Industry Dissatisfaction?

COFES 2010 is just around the corner, and I had chance to discuss with Brad Holtz of Cyon Research the potential topics to think and discuss towards the event. We touched the issue of the top PLM industry problems and how it comes to the discussion on Daily PLM Think Tank. So, this morning, I made my first BLOG “Better listening on Google” research and shout “PLM dissatisfaction” search. The article with a complete match to my criteria came in the beginning – “Why is Industry Dissatisfied with PLM?” by Frank Lillehagen, and I decided to make a deep dive into this article and to compare it to my thoughts related to the current PLM industry problems.

The main introductory made by Frank is that PLM actually was too late to the “enterprise app dinner” and was built on top of either CAD, CRM, ERP or even PDM. It was interesting to me, since I have never seen PLM system built on top of CRM. However, I think the main point was clear – in the family of the enterprise engineering apps, PLM appears to be a guilty child who needs to explain his rights to exist and prove all the time his organizational and technological ROI. Later in the article divided into sections: 1- What is missed?, 2- PLM capabilities needed, 3- Towards Product Knowledge Architecture, 4- Where do we go from here? Each section presented from 10 to 20 bullets with lists of issues that need to be done or missed. Actually, in my view those are expected to be a single list of issues, but they were presented in different ways.  I made some write up based on the topics that got my special attention:

A PLM architecture should create visual collaborative scenes for proactive behavior and learning-by-doing, enhancing human capabilities and creativity, creating the real-time enterprise, and transforming industrial computing to practical services and work processes.

We need to combine the complementary strengths of computers and humans, in handling and managing data categories, object classes and product families, to provide methods for life-cycle management and reuse.

Huh… It is complex. I must admit, I personally failed to come to the conclusion about the topic after reading this article multiple times. I tried to analyze the list of 20 PLM requirements needs presented by author. However, issue presented there were either completely broad like “Transform IT systems development and industrial computing” or “Support holistic enterprise design of projects, products and work environments” or absolutely theoretical like “Redefine design theory to exploit role-oriented knowledge architectures and product families“.

I almost gave up, nevertheless, the topic of industry dissatisfaction kept me in the focus. Actually, I felt my complete dissatisfaction by my inability to write the conclusion and decided to put my top 3 priority list that can improve the level of PLM Industry Satisfaction. Here we go:

Clean the PLM Language from buzzwords and broad statements. In my view, lack of clarity is absolutely needed as a first step. Customers and industry audience needs to get feeling of simplicity and clarity in definition and implementation.

Open Data Foundation. The main issue here is that currently I observe a huge amount of disruptive data trends in different PLM implementations. Combined with a large amount of legacy data, apps and home grown implementations it created the situation where a significant effort need to move PLM data implementation forward. The potential idea for this foundation is open source data foundation alliance supported by all PLM vendors.

Flexible Integration Capabilities. The integration topic in PLM is currently in the very bad shape. In my view, vendors gave up in their effort to invest into integration. Customers are mostly solving these problems by themselves or by 3rd party products.

An additional issue is social. PLM as an industry needs to come to the level of maturity in their ability to organize PLM events with involvement of multiple mindshare PLM vendors and related to PLM providers and supporters. The understanding of such need is very important. There are an absolute absence of PLM events these days with some small exclusion. The healthy industry needs to be able to consolidate around strong industry leaders and additional technological, vertical, service and other commercial companies.

Just my thoughts… I will continue my “thinking research” and hope to come with some conclusions towards COFES 2010.

Best, Oleg



Share This Post

  • Hi Oleg,
    i was intending to send you a message and this post is completely related to it.
    When i talk with leads these days, as an Aras Innovator integrator i talk a lot about flexibility. But flexibility is a long term investment. You somehow see the ROI of flexibility 2 or 3 years after the implementation, when you want to change things in the PLM system. But on initial discussion, flexibility usually means development and money for the customer.
    So i have an idea which could be good for every editors. One of the few first request i make when i start proposing an implementation, is getting all the Excel files related to product management, and i ask many information about what’s in these documents, who use it, etc… And sometimes you see features that you can’t find in any PLM systems. And it’s ok, that would be a customizing. But, from client to clients, you see that what you thought to be specific to a client is common to many. But just a little bit different and with different names.
    So my point is, if we want to get PLM back on track for all industries, we should widely ask to every industry, to describe all their Excel files (i know it would be a million file, but i’m just talking about a concept). And then i’m sure we could build common development for companies and lower the dev cost. I guess the job is already done right for change management and FMEAs but many other processes should get the same treatment. The aim is to give an out of the box software with many more processes managed without customizing needed.

    Wish i’ve been clear enough, not easy to explain that!

    Best Regards,

    Yoann from Prodeos

  • just to add on my comment, on our codeplex (Microsoft Open Source Portal) page, we are starting some projects related to Excel files we found in more than one company. The aim is o get people involved in it to contribute on the different topics. We develop the requirements, the specs with contributors and then Prodeos develops it based on Aras Innovator. We are at the very beginning of this on our codeplex page, we are not looking for developpers or PLM consultant, we mainly would like to get industrials input in order to mind the gap between actual PLM Offer and industral needs.
    here is the link:

    best regards

    Yoann From Prodeos

  • Yoann, Thank you for sharing your thoughts! My take on this is a different than yours. In my view, the ROI of flexibility should be exhibited at the time of system implementation. I’m less believe in out-of-the-box stuff. In my view “best practices” is an important marketing activity. However, when it comes to the real implementation out-of-the-box system will be customized in the beginning. I wrote about that in my last post – PLM Best Practices Torpedo ( So, what I think, is doable. We need to have a flexible system that can adapt to the existing customer’s environments, etc. The enterprise organizations become very complex and to create a single data model (for an industry, organizations, etc.) is a hardly achievable task. Just my thoughts…Best, Oleg

  • Andy


    In my view the PLM industry has made a mistake of way overselling their product. If you listen to the sales pitch they’ll tell you that if you deploy PLM you’ll save millions, decrease time to market, improve quality, blah, blah, blah. The sales pitch is bogus and I think everyone knows that by now but the vendors keep using it. I suppose they don’t know how to be honest after so many years of puffery.
    The other issues is that all of the big PLM vendors use a strategy of holding your data hostage in their system. And the customer has to pay $M for this privelege.
    Where I work we’ve ditched the whole big PLM thing and are building our own solutions on the Aras framework. Our data, our schema, our tools fitting our process. It takes work to do that but at least I don’t have to constantly tell users that “the system doesn’t allow you to do it that way”. Oh, and “while you can’t run your business the way you want to, don’t forget to write the big check for the maintenance bill”.


  • I agree with Andy and you that “Open Data Foundation” is a key evolution that needs to happens by one way or another in order to put the company (Customer) in a situation where they can take full advantage of their data.
    I have recently stumble on [1] which is illustrated in these articles [2] and [3]. if found this approach interesting since it could enable the communication between alfresco, SmarTeam, Teamcenter, …

    Imagine a world where you could quickly build application (mashup) in a language agnostic way taking advantage to the data stored in all these repositories…


  • Oleg,
    you’re right for editors, but I think Integrators and communities (could be open source) are the key actors to mind the functionnal gap between out-of-the-box offer and implemented systems. I see some processes that are not implemented yet and which really are common to many industries. That’s why Prodeos launched this Open Source PLM portal.

    Best Regards,


  • Andy, Thank you for your insight! I think, the problem you outlined is not specific for PLM industry. I can imagine similar situations in other enterprise business domains.This is, in my view, the reality of value selling enterprise model – to get value and to pay money in return. However, in every situation, things can get to the extreme. And this is what happens. A huge amount of buzzwords is surrounding every PLM implementation. Why it is going that way? Partially, because overselling. However, it also happens because engineering word is complex, and it is very easy to get involved into various “naming convention” issues. Therefore, I see the need to clean PLM language as the number one target in changing PLM industry from the existing level of dissatisfaction. Best, Oleg

  • Yann, Thanks for the links. This is indeed interesting. It looks like big vendors are starting to pay for standards. At least, the presence of SAP, Oracle, Microsoft, etc. vendors in this CMIS initiative show, they (vendors) are positive. This is surely not the first and, probably, not the last “data standard” initiative. The ugly truth about standards – they are costly, hardly can bring results to a single vendor and requires “agreement” on their creation between many people. I’ve seen lots of standard initiatives. Some of them are still alive… However, the majority is dead. There is one (!) place that allows us (quote you) “quickly build applications in a language-agnostic way taking advantage of data” and this place is the internet. We need to learn from this space. Just my thoughts… Best, Oleg

  • Yoann, I agree. For the integration type of work, FOSS can be a very interesting solution. However, I see a problem in position between “out-of-the-box with gaps” and “services filling the gaps”. This is the fundamental root of the dissatisfaction. Customers have been feeling that they pay for “half-meal”. Don’t you think so? On the opposite site, I see FOSS as a very positive. However, the platform for this PLM FOSS is only in the initial stage of creation. Thanks for great discussion! Best, Oleg

  • Hello Oleg,

    I agree with you that Internet is a place where a lot of value is created with data mashup.
    It has been interesting to see in a slightly different world ‘publishing industry’ that some big TV network in US have decided to go down this road.
    they use Alfresco in the backend and build a Web portal to integrate it with other legacy systems and provide a unified front end during the lifecycle of the assets from the submission, edition, publication, distribution, …

  • Yann, You are right about Publishing business. I’m watching this business with a great interest. In my view, what is going in Publishing will be soon repeated in the enterprise. Great discussion! Best, Oleg

  • Frank

    Andy: I completely agree with the hostage issue you brought up. Closed, proprietary CAD data is a real problem in the industry, and much more in PLM than anywhere else (even MS opened up their file formats recently).

    What I really don’t get is your affection for Aras (I saw it in your previous comments). The Aras stuff is nothing but a customizable (PDM-)platform that’s kind of free (but *not* open source, BTW — smelly marketing pitch). Except the “free” there is no substantial difference to other data/process management platforms in the market, AFAIK.

  • Frank… Thanks for your comments! Agree on CAD data and openness. Even Microsoft is better and allows Google to load Office files into Google Apps.
    Re Aras – Let’s keep the stage for Aras’ folks to provide you with answers. Best, Oleg

  • @Frank – Wanted to provide you with response to statement: “Except the ‘free’ there is no substantial difference to other data/process management platforms in the market, AFAIK”

    There is actually quite a substantial difference in that Aras is “model-based”… means using XML solution models which for those non-tech people means real-time drag & drop customization without any complex programming. Its fast, its easy, its highly scalable and means that a company can use the out of the box solutions ‘as is’ or completely customize w/o making upgrades impossible…

    Make no mistake, this type of flexibility is not possible with Windchill, Agile, or any of the other legacy PLM systems because they are all Hard Coded (i.e. takes an army of programmers to make even small changes and then you’re orphaned and can’t upgrade).

    It’s easy for people to write things that are not believable… The only way to know for sure is to get proof… do a pilot, beat on it, compare it head-to-head with your company’s existing Windchill, Teamcenter, etc deployment…

    At Aras we make this possible for the first time… companies can prove out the system – without being forced to spend a lot of money on PLM licenses before you know it will work for your business, your processes, your PLM strategy. No one else has the guts to make their software freely available.

    Just my 2 cents, but don’t take my word for it… try for yourself (even if you work for another PLM company 🙂


    P.S. As far as being open source, all the solution projects provide full source using OSI compliant licenses… you can use forever, modify/change as needed, redistribute the modifications – whatever you want. Just because Aras runs on Microsoft doesn’t mean the solutions aren’t open source.

  • Hi Marc, Thanks for your comment and clarification! For the sake of discussion, and I’m trying to be very precise about technological terms, I know at least two PDM/PLM systems that are NOT hard coded from the data modeling standpoint (MatrixOne and SmarTeam). I cannot completely compare it to the XML model you are talking about, but I’d love to have the discussion separately… I’ll think how to post it during the next week. Best, Oleg

  • Frank

    Hi Marc, Oleg. I know for sure at least two other systems that are more or less “model based”, i.e. without a hard coded data model and extensive modelling features: Agile e6 and Contact’s CIM DATABASE (disclosure: I happen to work for Contact Software). Or take SAP (PLM), which is doubtless a quite capable modelling and development platform with countless tools surrounding it. Even a rather rigid system like Teamcenter UA can be adapted to a customer’s needs. I’d suspect that *all* “enterprisy” PLM systems have that platform characteristic, whereas simpler systems tend to focus on a closed solution scenario like managing CAD files.

    Maybe Aras has something special to offer in this area. I am all open for a fair race, comparing features and cost.

    What leaves me a little upset is hijacking the “open source” notion. This has already been discussed on Razor Leaf (see, where ChrisD put it nicely:

    “””You need only skim the Aras download license to realise that Innovator is an unapologetically closed application. Indeed, Aras PLM is not significantly different from Microsoft Excel: a proprietary engine that can, if you wish, execute sharable application-specific models. Like Excel, you can’t run any of those models without the closed application, nor do you need any of the community contributions to create an Aras PLM system. You can’t even get security updates or service packs for Innovator without a paid support subscription. Fair enough, but is this what OS means to you?”””

    Or read what Michael Tiemann wrote about that:

    I think Aras should fix their sensational “open source” communication and explain *exactly* what they do (which *has* its benefits).

    See you at COFES. Marc, I’ll buy you a drink for attacking you here in public … 😉

  • Frank, Thank you for such a great discussion! I’d like to join you and Mac in Arizona to discuss all “more or less scientific” definitions of OS as well as how appropriate Aras is using them. I’m recognizing Aras in innovative business models applied to PLM field. As I mentioned in my previous post- PLM Open Source: Business or Social Need? (, there are lots of benefits in such a model. So, we just need to find the name we can agree on :)… Best, Oleg