Last week I saw once more a post, where free PLM software was offered and combined with the open source aura it should be THE solution for companies that want to implement PLM during this economical downturn. I believe this is a big mistake and for the following reasons:
WYPIWYG (What You Pay Is What You Get)
I learned that the WYPIWYG rule usually applies in the software world. Free software is nice, but does not guarantee that in case some functionality is missing or corrupt, that it will be fixed. So in case a company wants to implement the free PLM software, what to do if you feel something important for your business is missing ? You can ask the software provider to implement it for you – but will this be done ? Probably only when it is easy to achieve it will be done, but no commitment as the software is for free.
To assure it can be done, the software vendor will say it is open source software, so it can be changed if you want it. But who is going to make the change ? The mid-market company that thought to have selected an economical solution is not an IT-company – so who to hire? The open source software development company ? And this is what their business model is based on – they have the expertise with their software, so probably they are the best to adapt the open source software – not for free of course – and they learn from that but the customer pays.
Conclusion: there is no such thing as a free lunch.
It does not mean that all open source software is bad. Linux has shown that for an operating system it makes sense. Operating systems are 100 % in the scope of IT. PLM is something different. PLM systems indeed need to provide an IT backbone to assure data collaboration and replication globally. However PLM is probably more focused on business process changes and NOT on IT.
PLM requires people with business skills and not software developers
From my experience, PLM projects fail in case there are no business knowledgeable people available. It did not only happen with free PLM software or open source software. Some years ago, ERP vendors started to provide free PLM software to their customers to keep PLM companies on a distance. Like free PLM software it looked nice business wise, the software is free when you buy their ERP system. But who is going to implement it ?
This free PLM software availability has changed in the past years for ERP vendors. Also ERP vendors see PLM as a growth market for their business, so they started also to invest in PLM, providing PLM consultancy and no longer for free PLM functionality. However in one of the projects I was involved, it is clear that PLM and ERP are complementary approaches. Interesting is that none of the PLM vendors focus on ERP, apparently ERP vendors believe they can master PLM. I won’t say it is impossible however I believe if there is no real PLM vision on the top level of an ERP company, you cannot expect the competitive focus to exist.
Are CAD vendors providing PLM ?
Some CAD vendors have an embedded data management solution to manage their own data. This is usually more a PDM system and often the word PDM (Product Data Management) is too much for that. These systems manage their own CAD data but have no foundation for a multi-discipline Engineering BOM. For me, this is the base for PDM, as most companies have several disciplines working with different tools all around the same product. So CAD data management for me is not a the base for PDM, so for sure not for PLM.
PLM vendors bring real PLM value !
For me it is clear having worked with different vendors in the past: an ERP vendor, several PDM and PLM vendors, it is clear for me in order to bring committed value to a customer, you need first of all people with PLM skills – the ones that can differentiate between business process adaptation and software development. In order to implement PLM successful companies need to change the way they were working (read many of my previous posts about this – in particular this one). Software developers tend not to take this approach, but they adapt or extend the software to support the old way of working.
Finally paying for PLM software guarantees that the development of this software has a continuation based on business drivers and best practices. A PLM software vendor has the drive to improve to stay in business, both by software capabilities but even more by providing industry best practices.
Therefor my conclusion is that free PLM software does not help mid-market companies.
Feel free to react as I believe it is an important topic in this market.
Jos Voskuil's Weblog 
11 comments
Comments feed for this article
February 16, 2009 at 4:27 pm
Jim Brown
Jos,
I enjoyed reading your post, but I am not sure how all of your points tie together. I agree with most of what you say, but I am not sure that it all adds up to saying open source PLM (or software of any kind) doesn’t provide value. Some thoughts:
– I agree the there is no such thing as a free lunch. Open source is not free. You pay for support. What you don’t pay are upfront license fees, and you don’t pay for support unless it is worthwhile
– PLM requires people with business skills – absolutely, but are you implying that open source PLM is written by pure technologists with no PLM experience? I don’t believe this is the case.
– CAD vendors providing PLM – CAD data management is an important aspect, but only a small part of PLM. But are you implying that open source software is only CAD data management?
– PLM vendors bring real PLM value – agreed, but again are you implying that an open source provider can’t offer real value? Particularly when you consider the vendor and the supporting ecosystem of consultants and integrators?
Again, I agree with what you say is important. What I don’t necessarily see is why they can’t be provided in an open source model. What am I missing?
What I would agree on is that no company, particularly a midmarket company, should jump on open source as an easy answer and not investigate the capabilities of the solution, the support, and the PLM expertise available. Companies should also develop a view on “total cost of ownership” because the license fee for the software is typically a relatively small portion of total cost, and open source only addresses that small portion (and can, in fact, raise the other portions). But I don’t see ruling out open source altogether without any consideration.
Jim,
thanks for your comment and I agree with your conclusion. When I wrote the post, I needed a short title that forces people to read the long story, so I used the newspaper style, which always tries to put things in black and white.
I understand the beauty of open-source PLM and conceptually it is like communism: if we all share our best thoughts it becomes the ideal world. However I believe that in situations where people are not commercially driven, they will not necessary share all their IP in order to make others succesful – the same debate you will find sometimes about the truth in a Wikipedia. For me PLM is software AND the knowledge to implement it, where till now my experience with open-source (or free) PLM was that it focussed mostly on the software side. I am looking forward to see it work in the mid-market, where the risk of working without bussiness skills is high due to the type of organizations.
Finaly I mentioned the CAD vendors (not in the relation of open source) as they claim to do PLM and I agree with you CAD is a small portion of PLM.
Jos
LikeLike
February 17, 2009 at 9:10 pm
Jim Brown
Jos,
Where we agree is that open source PLM does not solve the COST problem of PLM for midsize companies (or for any company for that matter). While there is a cost implication of open source, you can see from my post (if I can publish HTML the link is http://www.mbtmag.com/blog/1690000369/post/30040803.html) that there is much more to the decision than cost, and much more to the cost than the license fee.
But open source can be a business model for commercially minded companies, not just for non-for-profits or consortiums (or communist states for that matter). They choose to make their money from support and not license fees, so if they don’t provide value on the back end they fail.
Thanks for bringing up the topic, and I appreciate your perspective. In fact, I chose to respond because I respect your opinion and we usually see eye to eye so I wanted to test my own beliefs.
Best,
Jim
Thanks Jim for bringing it into a broader perspective. Looking forward and open-minded to how the future of open-source software will develop itself
Jos
LikeLike
February 18, 2009 at 12:15 pm
olegshilovitsky
Jos,
I think Open Source is pure 100% business model related and can help to a company or not in the same way as licensed software. My believe is that historically all PLM companies came from different place and therefore I don’t see them migrating to open source business model.
What is must for PLM technologies is to reduce overall technological cost – cheaper, simpler, more flexible and available. Today this is still too high. Try to delivery packaged solution failed in many cases – you can see it from today’s UGS announcement about adding development and solution tools for TC Express.
In addition, what will be helpful in PLM business is to reduce the overall upfront cost of a solution, make it more granular and dependent on values company can get from implementation. Possible ways are open source, SaaS or different licensing policy.
Best-Oleg.
Thanks Oleg you bring it to another perspective
Jos
LikeLike
February 19, 2009 at 1:44 pm
Tom Gill
Hi Jos,
Just to be up front, I’m currently working with Aras Corp. I think your
blog post is referring to a recent Aras post on linkedin.
I’ll try to address the issues you identified. As Jim stated above,
Aras is running a typical Open Source business, similar to Redhat. The
code is available for download, but if you want support you purchase a
service contract. Revenue is also generated from consulting, and
training. Peter Schroer, the CEO, has stated publically, he dramatically reduced the cost of sales by eliminating the typical PLM sales process. The rest of the business operates similarly to other PLM Software companies.
So far, Aras has released one major version per year, since converting to Open Source, and the third release will ship within the next few weeks. These releases contain enhancements and bug fixes developed since the previous release. Subscription customers get immediate access to the bug fixes and enhancements. This is the same process that proprietary software vendors use. If you are on subscription (maintenance), you will get at least as good a response from Aras as you would from other PLM Software or Solution providers. End users don’t have to fix the code, and in most cases don’t want to, but they can. You typically don’t even get this option with proprietary software. Even worse, what do you do when your vendor stops supporting your application, or even the version you are running in production.
Your conclusion is mostly correct, in the Open Source world, there is no free lunch. I say mostly, because there are several members of the PLM community who have downloaded the Aras software, customized and/or tailored to fit their needs, and are in full production, and did not pay Aras a dollar; no support contract, no consulting, no training. They definitely did however invest time to do their implementations. We hope to publish a case study on at least one of these in the future. Many of the production implementations are on subscription contract, for the “insurance” aspect, if nothing else.
A functional PLM solution is a large, complicated application. It requires a large variety of skill sets, including IT , programming, business skills, domain specific skills, and many others. As far as personnel, the Aras organization has an appropriate mix people. You can check out the executive teams backgrounds here: http://www.aras.com/company/management.aspx. It looks like a typical
enterprise PLM software company, the only thing obviously missing is the sales staff.
Are the CAD Vendors providing PLM? I agree that the mid range CAD packages are focused on PDM and that is not PLM. I don’t think those vendors would take issue either. I do believe that the PLM offerings from PTC, DS, and Siemens are solving PLM problems. They may approach problems from a CAD centric perspective, but, all address business problems way beyond managing CAD files. I do feel that properly managing CAD data is a core PLM function. [Guess I’m showing my roots ;-)] Geometric data and drawings are core to most products.
I don’t see why open source needs to be limited to infrastructure. Besides Linux, Open Office is a credible competitor to MS Office, Firefox is a credible web browser, and in the enterprise software space, SugarCRM http://www.sugarcrm.com, and Alfresco http://www.Alfresco.com are credible offerings in the enterprise software space..
I once had a boss that told me people don’t value what they don’t pay for. He was mostly right. I think that Open Source PLM software is a good choice for many companies, and is worth taking a look. Aras has live webinars every Monday. Take an hour to check it out. If it looks reasonable, download it and give it a try. It’s easy to install and get set up, and will even run in a virtual machine. If you do try it, we’d love to get your feedback.
Best Regards
Tom
Thanks for your response, and to be honest it was not only the Linked In discussion that triggered my post. I wanted to make the point that free software is not solving the real issue of implementing PLM – specially for the mid-market. A company has to go through a business change and this requires people with business skills. I saw a failed PLM implementation (not Aras), where after one and a half year the customer gave up, as their business partner was developing all kind of features for him, both the partner and the customer did not know where to go, they tried to fix the symptons.
I am looking foward to learn about and analyze success cases for open source PLM as I believe the knowledge issue remains the challenge. BTW, not implying that this is only valid for Open Source providers only 🙂
Best regards
Jos
LikeLike
February 19, 2009 at 6:41 pm
yml
Hello Jos,
I am going to surprised you, I agree with your article I think that you are 100% correct the fact that a small and medium company decide to implement an Open Source PLM will not warranty the success of their implementation. Do you imply that by buying the most expensive configuration (licence) from closed source PLM vendor he will be able to succeed in its implementation ?
We all know that one critical success factor is to have a well trained and experimented team to support the company in its effort of taking advantage of the chosen software. However I do not understand why you pretend that such profile does not exist ? It would be nice if you could develop a bit you thought on this.
I would also be interested to read in what the fact from buying from a software vendor will make this different :
”
Free software is nice, but does not guarantee that in case some functionality is missing or corrupt, that it will be fixed
”
Are you trying to tell that a closed source PLM solution does not have missing or corrupted functionality ? Or that all of them get fixed in a timely manner
The last bit is about the word Free often associated to Open Source it does not stand as Free as beer but rather as Free as in freedom. That means that you are free to fix, extend, modify the software.
I am sure you already know this but just decide to make a short tainted article before posting a more neutral and in depth one. This was a balloon to test your audience. In some domain now Open Source is the norm and you have to justify very hard the competitive advantage you will get at not using such approach. This is not yet the case for PLM but I do not see any technological reason why this will not change when the market will grow.
Regards,
–yml
Yann, thanks for the mind-reading and your comments. I agree, I was hoping to get some comments from other directions. I believe in Open Source in specific areas where business knowledge is less relevant, and I am trying to understand (and get the confirmation) that it will work for PLM. For sure PLM vendors all have their own agenda too, to capture their possible audience. I am looking for the solution that will serve the mid-market and have till now not found the solution that will make the difference
Thanks
Jos
LikeLike
March 9, 2009 at 12:45 am
Herve Kabla
Hi Jos, I’m sorry I didn’t read your post any sooner.
i think you’re partially right. Free s:w is great in only when it comes to the mass: I guess (or hope) you’ve been using Firefox (or Chrome) for the last few months. These free pieces of code are masterpiece, with regards to IE, which is almost free, but as it only runs on MS platforms, is partially fee-based…
I can’t see any real argument to defend free s/w when it comes to huge s/w, used by a few hundred or thousands of enterprise.
The real issue DS will have to fight some day is not free PLM, but free CAD. Ok, free to the limit of support.
Thanks Herve for your comment. I tend to believe that free sofware always requires a knowledge context – how to use it, how to integrate it. Additional free software would lead to open standards, question will be who will adapt.
Best regards
Jos
LikeLike
March 12, 2009 at 7:58 pm
chris
Your conclusions seem more like yelling at the scorebaord…
WYPIWYG – while there is alot of truth to this you example is not as true as you try to argue. Just becuase you paid for the software, installation and support, it doesn’t mean things will be fixed, nor does it mena they will be fixed on the customers timeline, nor does it mean you culd implement the fix… Many PLM implementations are stuck due to so much customization.
Linux – the argument that an operating system is so how less of an important infrastructure than PLM, just is hard to swallow.
PLM needs consultants – Certainly this is true, but this is not a postive, it is really a negative. But since PLM needs so much consulting, just what the consultant are availble to work on any application.
Thanks Chris for your opinion. Do you believe PLM can exist without consultancy or maybe to phrase it different would it be possible to realize the PLM goals in a kind of evolutionary approach where the software tools guide companies to work different or is there always change management required
Jos
LikeLike
March 19, 2009 at 5:33 pm
MarcL
Jos, personally I feel as if you hit the nail on the head with many of your comments. Just because you don’t pay for a PLM solution doesn’t mean there won’t be associated costs and time commitments necessary to achieve success.
To be up front, I work at Aras and as an enterprise open source business we are clearly not a charity. Our intent is to fundamentally change the relationship with the company implementing the PLM software. Instead of forcing a huge up-front license expense before the customer even knows if the PLM system will work for their company, the open source format eliminates that license cost and risk.
A company can prove out the Aras PLM solution before they commit. Then, they have the freedom to make changes as needed without waiting for the PLM software company. As you know, feature requests can take years to get implemented by any of the major PLM providers (and sometimes never do). With OSS changes – enhancements, fixes, innovative new processes, and such – can be contributed back so all companies using the system benefit. Of course, contributing back is optional in the Aras format. This makes the open source PLM solution advance much more quickly than conventional systems.
Just for the record, from a PLM knowledge perspective, I think you will find the Aras people quite experienced; Agile, Eigner, MatrixOne, PTC, Workgroup, and other PLM companies – and everyone has worked in the field at large and midsized manufactures during our careers. We run the business in a very professional manner and offer a high degree of responsiveness and expertise. Everything is optional, and effectively people just pay for our time at their discretion.
Overall, PLM success is dependent on a company’s commitment to solve a business problem like improving change management, product collaboration, new product introduction, etc. Spending a lot of money on PLM licenses only introduces cost and risk into the equation. Given the global economic conditions, the enterprise open source format for PLM means that companies of all sizes, particularly midsized enterprises, can take advantage of proven PLM software and avoid risk.
At some point you indicated that you’d like to learn more about successful implementations of open source PLM. If still interested, there are a number of PLM case studies and presentations from our community events available at http://www.aras.com/customers/
Appreciate your perspective on PLM topics and look forward to your thoughts.
Thanks Marc, for your comment and do not feel attacked as a company by my post. The point that I want to raise and perhaps not clear enough is that free software is not a reason for a succesfull implementation – and here we agree I believe. In the past I worked with some companies that got their software for free (as part of their ERP bundle) and the experience with these companies was that, because the PLM software was free it was chosen to be implemented. But there was no serious budget and vision to implement it.
If I had to implement PLM as a company I would first focus on finding the right partner that can assist me to implement, specially in mid-market companies you cannot afford yourself to be a guinea pig.
Next I would start small with either open source or licensed PLM to get experienced. And probably if I work with the right partner, chances are big that the initial solution will be used for expansion.
For sure there is a difference between the various PLM solutions, but I believe none of them is a real show stopper in case of implementation with the right expectations. References (industry validation) and preconfigured best practices are probably what you would expect as a customer from your PLM provider.
Open Source and a global deployement business model (local implementation support / knowledge available) can work as good as a PLM vendor with a global channel. Only the source of revenue is different per type of company.
Meanwhile I am appreiciating the discussion as it helps me and others to better understand the details
Thanks
Jos
LikeLike
October 22, 2009 at 8:27 am
Prasad
Jos,
Nice one on the open source. PLM implementation can go on for ever and eat all the money even when we buy, its the essance and undersnading and implementation perview which matters rather than which PLM you select. Iam not aware of any statistics collected on the number of organizations who are happy with the purchased software. We see PLM vendors releasing versions every 6 months with one or another bug fix rather than a new feature. Many of the implentation and maintanance costs are due to Bug fixes rather than new features.
Under these conditions woudn’t it be good to take a base model and built on it with strong testing. Any way the PLM users have to test the implementation. I do not know any software vendor who guarentees a impelemtation to be productionized directly without being tested with the user data.
Under these conditions a end-user any way does the following:
1) Record and monitor his business process.
2) Test the use cases suitable for his business process.
3) Customise the PLM as per the business requirement.
Many of the PLM vendors say OOTB instalation and setup can be used, Iam not sure how many of the PLM users live with OOTB setup’s.
Given these conditions only additional cost is the developer who can code. Is this cost too-much if we can avoid bugs and software releases are in our control rather than forced migration due to bugs created by the vendors.
Prasad, thanks for your comments. I believe you comment could be a start of a new discussion or post, as I understand the direction you are aiming, but i do not agree on the facts. I have the feeling that often PLM systems have too much functionality and not enough time and effort at the customer is spent on change the way companies should work. For me – software is 50 % of the success, the other part is the people involved, both the customer and implementor’s vision and experience. PLM should not be a software issue – it should have a focus on business value and ROI. And if the result of that is, that you could keep on implementing (as the company changes/ the focus changes) I have no problem with that as long as the value added is visible and justifiable. Change and flexibility are major strengths of mid-market companies. We do not want to force them in a fixed PLM implementation.
As said before I will put the points of this discussion as a target for a future post / discussion – it remains an interesting topic.
Thanks Jos
LikeLike
April 1, 2011 at 1:06 pm
Anoop Venugopal
Hi Jos,
Indeed a news paper style catchy header. But my experience with PLM implementations does not let me agree to you!
My humble experience implementing & maintaining PLM systems for large enterprises tells me:
(1) That licensing cost of a PLM system does not directly impact ‘success’ of a PLM project. This cost is always a big influencer for ‘product selection’ decision though.
(2) Cost of PLM implementation includes PLM license cost, PLM implementation cost charged by system integrator/consultancy company and cost of effort spent by human resources from IT, Engineering/Manufacturing & Management functions.
(3) Seed for success of a PLM implementation is a well thought and formulated idea of what business results are to be achieved. A robust PLM product capable of handling all processes involved provides the required environment for this success.
(4) An OOTB PLM solution that can be ‘installed and switched on’ is a dream. This might work to some extent if the objective of the implementation is only to address basic revisioning, CAD/document management etc. But then we will be talking about a PDM system and not a PLM system! PLM systems will need configuration, customization and maintenance effort put in to be successful. Whether this effort comes from the PLM product vendor, external consultants or internal resources of the company is a different discussion.
Regards,
Anoop Venugopal
Anoop hi, thanks for your reply – the news paper catchy title at that time was a response at some posts where some PLM software providers were claiming that with Free PLM software, companies now could realy implement PLM removing the barrier.
Agree with remark 1 and 2.
Remark 3 is correct for the ideal world I believe as only few companies are able to describe what they want to achieve – for me more important is that a company needs to have a PLM vision and then in a stepped approach implement this vision on a PLM platform.
Point 4 – I agree 200 %
Best regards
Jos
LikeLike
April 1, 2011 at 4:34 pm
Chris
Anoop as you say the most important is mapping the business objectives to be accomplished. This is the source of the problem. The objectives are very unclear… When customers purchased feature-based CAD the metrics were very clear, model faster. These metric drove/closed many of the deals and PTC was the best at helping customers create these metrics. Now the value of feature-based CAD is a given, but the market is saturated.
PLM vendors on the other hand have not provided clear business results. The pitch is more hopeful than finite when it comes to value creation. I will paraphrase what you said – customer should not let the wool be pulled over their eyes with a hope of value – they must define clear objectives and then implement with this in mind.
Implementation is also a difficult problem as it involves human change. Even with a simple and clear value of model faster many companies struggled with the change required and the change required to acheive the value of feature-based CAD was much less than the change required to achieve value with PLM.
Thanks Chris – fully agree
LikeLike