Author: Mark Lukianchuk

The paper is authored by Dr. Nicholas Khabbaz and François Cartier of e-Modelers, the preferred partner of Embarcadero Technologies (now part of Idera). Not exactly an impartial opinion. Keep that in mind when giving it a look.

Nonetheless, I want to thank Idera for coming out with a paper that actually compares current releases of each product, as the previous paper I read used a 5+ year old release of erwin DM that’s not even supported anymore. Thanks also to Nicholas and François for their efforts.

In my exciting new role heading up products at erwin, Inc., I lead a global group committed to growth and excellence. I am fortunate to have a team with decades of experience with the product, so I looked forward to their feedback on this new comparison. After all, we already know that our 64-bit erwin Data Modeler release is blazing the trail with the best performance in day-to-day usage and that we’re developing, selling, and supporting the #1 product in the market.

So, I was curious if they would come back with any inaccuracies. Surely there must be one or two, right?

I received 14.

Yes, 14.

Within an hour or two of the paper going out.

I was originally planning to provide a quick summary of each point, but in the interest of allowing you to do your own research, I’m providing more detail. Apologies in advance for the length.

Not a technical user? Short on time? Feel free to skip ahead to the summary and the 15th point.

Fourteen reasons why Idera has got it all wrong:

1. While most of the info regarding semantics is mostly correct (with some omissions/inaccuracies on the erwin DM side) erwin DM has a much more logical and generally accepted organization:MODEL – SUBJECT AREA – MODEL OBJECTSDIAGRAMS (which can exist at the MODEL and SUBJECT AREA levels) are strictly a vehicle for visualization and not a construct of the model itself. er/Studio does not have a comparable feature. erwin DM users have a much more flexible and richer capability for constructing and displaying model metadata for specific goals and/or purposes.So, the paper is wrong.
2. An erwin DM model is comparable to an er/Studio DIAGRAM, not an er/Studio MODEL. It represents either a Logical Only, Physical Only or Logical/Physical abstraction of metadata. There is no limitation (as in er/Studio) as to how erwin DM models can be related (e.g., 1 logical to many physical, 1 physical to many logical, many to many – can be logical or physical).Again, the paper is wrong.
3. Section: Data Model Organization, Model/Submodel Organization – this is incorrect. Metadata contained in Subject Areas in erwin DM are automatically synchronized. There seems to be some confusion between the concept of Subject Areas and Diagrams within erwin DM, most likely due to a lack of understanding of these erwin DM specific concepts. Auto-populate is a visualization function and not a synchronization function.Wrong again.
4. Despite statements to the contrary, erwin DM does provide a fully user configurable and flexible “reverse” naming standards capability (Physical to Logical).Yup, wrong.
5. Whilst the opinions detailed in the Section on Attachments / UDPs are factually correct, there is little to no compelling value (our opinion) in the er/Studio approach vs. the erwin DM approach. We’re not alone in this opinion, although we do acknowledge that it may be relevant to some people in some specific cases.Wrong to assume this is a true differentiator.
6. Despite statements to the contrary, erwin DM does support ALTER statement generation within Complete Compare.Wrong? Yes, wrong.
7. erwin DM fully supports reverse engineering of a database schema (or DDL) into a Physical-only model to which any number of defined naming standards can be applied in the derivation of a Logical model (from the existing Physical model) thereby automatically generating the desired logical names. No manual process involving spreadsheets, CSV files and the Bulk Editor is required.The paper is wrong as erwin DM supports this.
8. Automatic synchronization of model object mappings is accomplished using Design Layer models in erwin DM. Once established, these mappings are saved and synchronized via Complete Compare.So, wrong again.
9. Despite statements to the contrary, erwin DM maintains and manages a complete log of model changes at all levels entered by all modelers. As this log is a feature of an erwin DM model, this information is available within the mart.The statements are wrong.
10. Unlike er/Studio (and the lack of functionality described in the document), erwin DM can roll-back changes made to any model version saved in a mart – not just a Named (or Marked) version. Additionally, erwin DM has the ability to compare two model versions and to dynamically select metadata from either of the compared versions for inclusion/exclusion from a new version of the model. Although there is no specific branch/merge functionality, use of model versions, complete compare functionality and mart storage provides equivalent capability.Wrong. erwin can do this and more.
11. It appears that e-Modelers is not current with erwin DM functionality. For instance, our API supports .NET Basic and not the deprecated capabilities of Visual Basic. Furthermore, many of er/Studio’s macros are used to provide capabilities equivalent to “built-in” erwin DM functionality. erwin DM provides automation capabilities via our API as well as from within our product (e.g., FETs, user-defined reporting capabilities) using our well documented TLX macro language.Wrong as we provide more than what was written.
12. Section on Data to Business Process Alignment is misleading. It leverages a comparison of erwin DM to er/Studio’s Business Architect (a separate product with an additional cost) – which is truly a comparison of apples to asparagus. A better comparison, of course, would be their lackluster business process capability and ours when combined with our partner product (“Powered by erwin”) from Casewise. Our no-cost metadata bridges also provide the capability for erwin DM users to integrate their model metadata with best of breed products from vendors across the industry. Misleading, so wrong.
13. Despite statements to the contrary, erwin Web Portal provides a full range of flexible advanced search capabilities, metadata tagging, shared bookmarks, user defined comments to facilitate stakeholder collaboration, ability to provide attachments to managed metadata, a web-based data modeling feature (Data Documenter) and the Universal Data Model Harvesting capability which allows the inclusion of non-erwin DM modeled metadata (i.e., er/Studio models, relational database schemas, unstructured “Big Data” data sources). erwin Web Portal models may be viewed in a wide variety of user-selected modeling notations (IE, IDEF1x, UML, etc.).Wrong with respect to erwin Web Portal.
14. The list of databases supported (Appendix 3) is incorrect in that most databases left “blank” on the erwin DM side are supported via ODBC for FE and RE. Also, Microsoft SQL Azure is a natively supported database (NO “extension” is required although it is a separately licensed item.) Finally, the er/Studio side of this list is artificially enhanced by listing many databases that are no longer supported by the database vendors themselves. If you’re still running 1992’s SQL Server 4.2 on OS/2 1.3, please feel free to give me a call as I want to hear from you!Old news. And wrong.

To summarize (or TL;DR if the above was too much to read):

We found 14 things wrong with the analysis. In reality, some of these points have multiple facets so it’s more than 14, but that’s a minor point. Let’s just say 14.

Don’t take my word for it, please feel free to check it out for yourself.

As we’re busy developing our new release of erwin DM, we don’t have the time to dig for more, but if you can find something else that we missed, please bring it to my attention and I’ll send you a Starbucks gift card for your effort.

One last, and 15th point from me:

When judging the best solution for your business, you should look towards the leading solution on the market today.

The solution with #1 market share and used by more than 50,000 data management professionals in 60 countries around the world.

The one where the product name is also the company name. Where increasing investment in data management reflects 100% of its focus, not just a side business.

That solution isn’t er/Studio. It’s erwin.

And that’s why there are at least 15 reasons why Idera has got it all wrong.

Thanks to Neil Buchwalter, Danny Sandwell and the rest of the erwin product organization for their efforts here.

In other ways, over time you’ll see some key differences and not just with our erwin.com email addresses.

The way that our new company can build value is to invest in its solution, its partners, and its people. This means maintaining our leadership while also maturing our strategy. Adding new products and new partnerships. This is something that, based upon feedback, was missing for the past few years. We will be changing that at our new home.

But we can’t throw out the old, because that’s what made us the tremendous success that we are. We will be increasing our focus on our channel partners and also continue to deliver a superior level of customer service. Our passion remains as does our commitment to you, so stay tuned as things develop.

It’s a new day for erwin and the sun is shining!

To me though, it’s not enough just to compete. I want to do it in such a way that I can look my two sons in the eye when they ask if I had a great day at work and say “yes I did”. I know my team is the same way. It’s a great feeling to not only win, but to win with facts and not rhetoric.

Most of you know that CA is looking at where to focus its product portfolio, as any good company should strive to do. erwin, as a business unit within CA, is also being looked at to see how it might best fit – or perhaps be divested. This is not a reflection on the quality of erwin (far from it) but about how it fits strategically within CA’s portfolio. It’s something that few companies actively do well; GE is one that comes to mind that constantly does this. They are known for continually reviewing their portfolio to see what fits their vision for the future and which might be even more successful as a separate entity. I love GE – with my new home I purchased all GE appliances, only to find out that the division might be sold to Electrolux. So I called them and my mind was put at ease; their after-sale support has been excellent, so much so that I still plan to buy GE appliances in the future.

As it turns out, the Department of Justice may end up blocking the GE deal, which has overtones back to erwin and last November 2014 when they also “expressed continuing concern about the transaction’s potential for anticompetitive effects” when CA was looking to sell erwin. With the deal terminated, CA continued to look into alternative options going forward, and in doing so as a public company, listed erwin as remaining in discontinued operations on its income statement. This is required from a regulatory and compliance standpoint. Nothing nefarious there.

As I’ve traveled the globe meeting with partners and customers I’ve been very transparent in what this really means, and that it’s essentially “business as usual” for erwin while we look towards our future path. Whatever decision is ultimately made will be for the benefit of our customers, partners, employees and the products, and everyone I’ve spoken with has appreciated my candor, transparency and approach in this manner.

This is how we at erwin operate, and as soon as I’m in a position to provide an update on this I will, but no matter the outcome I’m confident it will be good for erwin as we have accomplished a lot this year.

In March we released the erwin Web Portal Data Governance Edition, which enters a new and important segment for us. I have been adding new staff across many areas and geographies in the business to support our customers and partners. And, of course, we have been busy making significant updates to erwin Data Modeler, from the 9.6 release in March through the 64-bit release that’s in beta now. CA has been super supportive of erwin.

The feedback from our customers and partners is that we’re moving in the right direction strategically and are executing really well. Thanks for your support and loyalty, we work hard to earn and maintain it.

Due to the loyalty of our partners and customers I also regularly get forwarded e-mails from one of our competitors which tend to say something else, such as that CA is “killing” erwin, that the only releases coming out are “bug fixes”, that we are laying off staff (in some area, geography or function) and that due to this uncertainty customers should switch to their solution. Usually these e-mails make me laugh because I know better, but I do realize that some people might actually take them seriously so I personally follow up on each and every one that I hear about.

I’ve hesitated to say much about this to date because I’d rather that my team focus on improving erwin than to address these petty issues. After all, I have to believe that our customers and partners know differently, right?

I’ve hesitated to say much. Until today.

I just read a report on Reuters that this same competitor may be sold. Given this, it is extremely ironic that they have been telling customers to leave the #1 data modeling solution for their solution due to corporate uncertainty.

Food for thought.

The reality is that there is always change and consolidation in the software market. erwin itself started with one company and went through a couple of acquisitions, and remains the leader.

That’s why customers should always focus on partnering with a company who values loyalty and who works hard to earn the respect of its partners and customers.

If a customer buys a competitive solution due to technical superiority, then I will ask (and thank) them for their feedback and take it back to the team so we can work harder. If, however, they want to move off of the #1 data modeling solution because of fear, uncertainty and doubt spread by a competitor, I would ask that they look past the rhetoric.

On another note, why do I keep on saying we’re the #1 or “top” solution? It’s not because I think so. It’s not because one of our customers says so. It’s because we’ve been validated as the #1 solution by revenue in the data modeling market by a major analyst firm. It’s also because we were named as the top data modeling solution by a recent poll (see details of that award on erwin.com) – of which I have you, our customers, to thank for that.

You may see other references from our competition that they’re “top” or “#1”. All I can say is that if it’s said without any attribution or backup you should ask for the details. I saw one a while back that claimed ridiculous performance differences between their product and erwin. Naturally, that concerned me, because I want to make sure we’re as good as we can be. I also read that same company’s white paper describing reasons why customers should switch to their solution from erwin.

What did I find? First, the comparison was between their current release and a version of erwin that is 5 major releases ago (version 7.3). I’m admittedly not a technical expert these days but one would think that an “apples to apples” comparison might be best. Even a run of the mill Ford sedan from 2015 will beat a 50 year old muscle car.

Second, any mention of performance differences didn’t list the use cases. Or the specifics. Of anything. I’m the kind of guy that likes to see the data that backs up any assertions, and I’m sure you are as well. When you see such comparisons, you should ask for information such as relative model size (large models behave differently than small ones), the use cases and releases used. This then lets you make your mind up based upon facts.

To conclude, I’m all for competition. My team and I love to compete, but we like to do it by working hard and producing the best product possible instead of resorting to rhetoric. I think we’re doing a great job of this but of course, we always can do better. Feel free to e-mail me at mark.lukianchuk@ca.com if you have any ideas.