CMS Watch Trends and Features

Microsoft leaves Linux-based FAST customers stranded

kthomas@cmswatch.com(Kas Thomas) — Tue, 9 Feb 2010 08:00:00 -0500

Buyers of the Linux and UNIX versions of FAST's Enterprise Search Platform (ESP) got some bad news the other day: It'll now be necessary to switch to a Windows server platform, or else move to some other product for enterprise search.

Microsoft Corporation, which as you may recall acquired FAST two years ago for $1.2 billion, has confirmed it will stop development of FAST's ESP core on Linux and Unix after the release of SharePoint 2010 later this year. Microsoft has in effect said it will now build its FAST search engine only for Windows.

FAST chief technology officer Bjørn Olstad hinted there will be an upgrade program to move customers to a hosted search service or FAST on Windows. He also said Microsoft will support ESP 5.3 throughout its product lifecycle of five years' mainstream support and five years' extended support. (Extended support for ESP 5.3 will end on July 16, 2018.)

We suspect that some FAST customers will use this as an opportunity to look at OS-neutral search solutions, of which there are many. Let's put it this way, it would be surprising if FAST's Linux customers did not look for alternatives to FAST at this point, because presumably these customers chose Linux for well-considered reasons (most likely involving security or performance) in the first place. They won't be easily persuaded to invest in Windows servers now if that was not the original plan.

It's not just a matter of swapping out boxes. It's also a matter of having appropriately trained personnel in place to set up and administer a Windows-server cluster instead of a Linux one -- and accepting the security trade-offs that come with moving from Linux to Windows. That's not something a Linux customer relishes thinking about.

For more on FAST and your alternatives, consider our Search and Information Access Research. We'll keep our subscribers informed of developments as soon as we learn about them.

Search fares well in sluggish economy

bloem@radagio.com(Adriaan Bloem) — Mon, 8 Feb 2010 08:47:00 -0500

At times I get the feeling the content technologies aren't affected by the economic downturn. Some of the vendors we cover have come out with early reports on their performance over 2009, and so far the news is good.

For instance, French vendor Exalead proudly claims $22.7M in revenue, and 50 new customers. Canadian vendor Coveo isn't quite as open about the financial specifics, but notes 37 deals with new and existing customers in Q4 alone. I'll accept that there's at least some truth in the "record Q4 2009 results" and "significant growth" the company announced. (Though you can never be certain, even with publicly traded companies -- which Coveo is not.)

Then there's Autonomy, which has now branched out way beyond the pure enterprise search business. However, the English company's strategy still revolves around the IDOL search technology, so for good measure, let's add in their figures too: revenues of $740 million, up 47% from 2008, "including strong organic growth of 16%." And a staggering $323 million profit before tax, up 55% from 2008. Many stock traders may not know much about enterprise search, but they'll certainly know about LSE: AU.L.

Of course, this is just scattered evidence, and I wouldn't dare say that all of the vendors we cover in our Search & Information Access Research are breaking records in revenues or profit. (If only because companies that aren't doing so well aren't going to announce their figures with great fanfare.)

But it illustrates the sense I had this past year: that many of these products continue to be in high demand, and the content technologies are doing a lot better than the economy at large. At the beginning of 2009, we could still dismiss this as a delayed effect (projects already set in motion before the crunch hit), but evidently enterprise search continues to grow against the tide, at least for another year.

I can think of many reasons why this would be the case. (Feel free to add your thoughts in the comments below.) I believe the principal reason is that for many enterprises, search and information access technologies have become either just a cost of doing business (much as you wouldn't turn off the electricity in a shop when sales are slow), or a means to increase efficiency and get an edge over the competition (because unlike electricity, you can get better search). It's not something you cut corners on.

So even though the core technologies have been around for several decades, I tend to think this shows that there's still a lot of untapped potential for search in the enterprise. If you still think of search as a simple commodity, you might want to rethink -- you're being overtaken left and right.

ROI calculations are a joke

aps@cmswatch.com(Alan Pelz-Sharpe) — Mon, 8 Feb 2010 07:47:00 -0500

Our colleagues over at Forrester recently undertook some interesting research regarding content management investment attitudes in 2010 (DM, RM and WCM). The overall finding was along the lines one might expect, "72% of respondents intend to expand their use of ECM technologies"... but there was an intriguing second key conclusion in the report: "49% could not estimate the ROI for any of their ECM systems."

Let me state my take on ROI calculations as clearly as I can. ROI calculations for information technology are junk calculations, a fraud, a nonsense, and a complete waste of time. Clear enough for you? Oh and by the way, ROI calculations from software vendors are even worse...

ROI assessments are based on the simplistic formula of benefits minus costs to calculate the return on your investment. But simple is not always smart, and most if not all the of the benefits in such calculations are by nature predictive. In other words they are guesses, and in my experience, almost always overly optimistic -- and fatuous guesses at that.

There is a cost to any new system, and there is also always a return (sometimes a good one, sometimes bad, often a bit of both) from the system. Far better and more honest I believe to just build a valid business case that details these costs and potential returns, and utilize more concrete and verifiable calculations such as TCO (Total Cost of Ownership) or CDB (Cost of Doing Business). That way you have a business case that details the investment and what you hope and expect to achieve as a result, and at least have a business case based on facts. One that when it strays into the predictive (as it must at times) is clear about its limitations and values.

The industry analyst guru's guru, Paul Strassman, has written extensively on this topic. For those who want to know more about the pointlessness of ROI calculations, I highly recommend you read Paul's bestselling book, "The Squandered Computer", a work that IMHO should be on the required reading list for any IT or Business related degree course.

But to return to the Forrester research findings, I would respectfully argue that they asked the wrong question. For rather than asking buyers of content technologies whether they can build a valid ROI for new investments, they could have been asked whether they believed new investments in content technology would deliver worthwhile benefits. The results from that would have been very interesting indeed, and I suspect quite different from the question that was asked.

Just like writing an RFP, developing a solid business case can be detailed and tricky work - and it is work that we help our advisory clients with on a daily basis. But in essence it is the core purpose of our work: we help buyers make the right and best technology investments possible. Yet over the years I have seen so many nonsensical ROI calculations, so many works of fiction claiming to be business cases, and so many buyers misled by farcical ROI calculations, that the very sight of those three letters is enough to make my blood boil...

An iPad for DM and RM?

aps@cmswatch.com(Alan Pelz-Sharpe) — Fri, 5 Feb 2010 07:40:00 -0500

The launch of Apple's iPad last week has caught the imagination of armchair critics worldwide:

"What's it for?"

"What a funny name (snigger snigger)"

"It's just a big iPhone"

And so on. But whether the Apple device itself is a success or not, we seem to be approaching a tipping point for improved user interfaces to manage documents.

Be it the Kindle, Tablet, Nook, or iPad, keyboardless document reader devices are on the verge of becoming mainstream -- at least for consumers. For browsing and reading through large volumes of files, or large documents containing multiple pages (books and libraries), such devices are infinitely more user friendly than the current desktop or laptop paradigm.

Take that one step further and it is logical to reach the conclusion that such devices might work well for reading through airliner maintenance manuals, consulting documentary evidence in court, searching archives, or accessing a patient's medical records and images? So while the pundits mock the iPad, I see real potential here for the world of case, document and records management. That said, I already own too many Apple devices, and may have just drunk too much Apple-flavored Kool Aid. But surely it can only be time before some enterprising vendor starts to deliver secure organizational and access applications for these devices.

I for one wish them luck, for as somebody who has spent his career digging through virtual crates, accessing electronic files that I then need to print to actually read, I know for sure that there has to be a better way.

Alterian drops Immediacy

bloem@radagio.com(Adriaan Bloem) — Thu, 4 Feb 2010 11:42:00 -0500

Software vendor Alterian has cleared up which Web CMS software it wants to try to sell you in the future. In short, all of their alphabet soups going forward will taste like Morello.

The U.K.-based vendor will be dropping the "Content Manager Corporate" edition product (CMC, formerly known as Immediacy) and concentrate its WCM efforts on the Enterprise Edition (CME, formerly Mediasurface Morello). The company expects to release "CM7" this fall as a successor to both CME and CMC, but this will really be an updated CME, not an integration of CME and CMC.

Readers of our Web CMS Research will know that when Alterian bought Mediasurface, the company was juggling three systems, ranging from

Hosted Pepperio
.NET and page-based Immediacy
Java/.NET object-oriented Morello

Pepperio quickly found a new home. But two distinct, full-blown WCM solutions is still a lot to carry (just ask Open Text).

Alterian would much rather focus on providing a one-stop online marketing solution, so rallying around a single WCM tool as part of that broader package makes a lot of sense -- for the vendor.

For customers, it's a different story. No doubt CME/Morello users will be happy to see Alterian focus on that system, which -- like all the tools we cover -- has had its fair share of problems. On the other hand, Alterian says it will support CMC/Immediacy "indefinitely," but the "upgrade" to CM7 will really be a wholesale migration. (Alterian promises scripts to ease the transition.) Current CMC customers and integrators will have to get up to speed on a new system that is architecturally completely different. Custom plugins won't work anymore, and the move from a page-based system to objects and fragments is far from trivial -- for authors as well as developers.

If you're currently looking for a CMS, it's probably fair to say that Alterian's soon-to-be defunct CMC should drop off your shortlist. If you want a system like CMC/Immediacy -- that is to say, page-based and .NET -- you should probably consider Ektron, EPiServer, GOSS, or Kentico instead. If you want to place a larger bet with Alterian, then your option is now simplified to CME. Just don't forget, that's a Mid-Range Platform, and a lot more complex than a Simpler Product.

Non-disclosure is a non-starter

kthomas@cmswatch.com(Kas Thomas) — Thu, 4 Feb 2010 10:15:00 -0500

We've mentioned this a time or two before, but in recent dealings with a well-known WCM and DAM vendor there seemed to be some confusion about it, so perhaps it bears discussing one more time.

We have a policy at CMS Watch of not signing NDAs with the vendors we evaluate. Nondisclosure agreements are antithetical to what we do; fundamentally, we're in the disclosure business. We perform research and then reveal what we've learned. We're not in the business of keeping vendors' secrets.

We do, on the other hand, honor news embargoes. Commonly, a vendor will have a new-product announcement that carries a specific go-live date. The go-live date might be a week in the future, but the press release is written and the vendor wants to show it to us. As long as the embargo period is reasonable, that's fine. We'll agree (verbally) to look at the press release and not write about it for publication ahead of the embargo date. We do that sort of thing all the time.

NDAs are a different beast. An NDA is a legal contract. NDAs typically carry all kinds of fine print that inhibit disclosure in various ways even after the information in question has entered the public domain. Vendors sometimes choose the NDA mechanism to share with analysts when the information in question involves business strategy or an acquisition. Public corporations in particular have to be very careful about how they disclose certain kinds of information. We don't want to be a party to those kinds of disclosures. We want to be free to explain to you the customer whatever we need to say, based on the research findings we dig up.

So if you're a vendor, please don't ask us to sign an NDA. We understand the need for secrecy on certain matters, but in that case, please keep it to yourself, and don't brief us on what you don't want made known. From a buyer's perspective, a real demo is worth a thousand words anyway...

The trouble with DAM and your corporate laptops

tregli@cmswatch.com(Theresa Regli) — Thu, 4 Feb 2010 08:00:00 -0500

The net-wide discussion over the Apple iPad's lack of Adobe Flash support has brought back a few memories for me: ones in which DAM tools with sexy Flash interfaces can't run in large, corporate locked-down environments, where there's restrictions against downloading just about anything (such as a special plug-in, or the latest version of Flash) to a corporate machine.

As we wrote about last year and cover extensively in our Digital & Media Asset Management Research, many DAM tools have refreshed their UIs in the last year, while other DAM vendors have continuously promised they will but still haven't released. Those interfaces are largely Flash, Flex, or AJAX-based, and definitely make DAM interfaces look like they belong in the 21st century.

That doesn't mean they're perfect. In the past several months, while assisting three Fortune 500 companies with their DAM procurements, I've watched the excitement in a potential buyer's eyes when they see a well-designed, fluid interface that's Flash or AJAX-driven, only to be followed by dismay when they realize they can't run the tool because of their locked down corporate machine.

It's something that often gets brazenly ignored by a vendor during a demo. The vendor may know that an entire corporation is still running Windows XP and the enterprise standard is still Internet Explorer 6, and a 2-year-old version of Flash, but that won't stop them from showing you something that's designed for IE 8+ with the latest version of Flash.

When you go to procure a new DAM package, be sure to include your enterprise OS and browser standards in your RFP. Your current corporate standards may well hold you back from investing in the latest, most feature-rich, and fully-supported version of a software product. (This can also prevent existing customers from upgrading.) So now is also a good time to start pushing for a re-visit of those standards -- and an upgrade to something more modern where needed.

How Does Your Portal Expose its Services?

adurga@cmswatch.com(Apoorv Durga) — Thu, 4 Feb 2010 05:43:00 -0500

Enterprises have traditionally used portal technology to aggregate content and functionality from different applications. In that sense, portals have typically been consumers of services.

Nevertheless, despite what vendors may tell you, it's very rare that a single portal instance becomes the only place where you access all your enterprise information. Most enterprises support multiple portals (or delivery environments) in an organization, and so a single portal server becomes just one component of the enterprise technology landscape.

This means that any portal must not only consume services but also expose its own services in a manner that other components of your enterprise architecture can consume.

Why would you want to do that? Well, many enterprise portal software products have services or functionality that could work well across your enterprise (i.e., not just within the portal). As an example, you might want to use a portal's in-built search engine to be able to index your Web CMS or you might want to use your portal's wiki services more broadly across your intranet.

Of course there may be licensing implications here, but in general you get the idea: de-coupling a portal service so that you can "white label" it or extend its functionality to suit a different environment.

How do you do this? For starters, you can choose among many standard integration mechanisms to integrate portal services with other applications. Among others, you can use Web Services, iFrames, Web Clipping, directly access a Portal's data source, or employ an Enterprise Service Bus (ESB) to access the relevant functionality. All of these have their drawbacks and are typically very complex to implement, especially if you want to access a "service" rather than just a "page."

Other approaches are potentially promising. The Oasis standard, Web Services for Remote Portlets (WSRP), provides a mechanism to expose portlets to other delivery engines. However, while many portal tools can consume WSRP, they do not often comply with WSRP for exposing their own functionality.

Some portal products now expose their services using simpler mechanisms like REST-based APIs, which provide a URI-based way to access a specific feature or subset of a functionality For example, you could obtain a list of top 10 blog posts using a URL from a portal's built-in blogging engine.

The other alternative that some vendors have implemented is the ability to expose every portlet as a "gadget" that you can then include in another web page using simple JavaScript.

As you can see, there are many ways of exposing a Portal's functionality and I'll not go into all the details in a short blog post. Obviously you face trade-offs, as some of these approaches are more complex than others, while some are more presentation-oriented than others.

So if you're evaluating portal products, make sure to understand how the product exposes its services and not just how it consumes services from other applications. We delve into the pros and cons of these different approaches -- and how each vendor covers them -- in our Portals research.

Vendor Tip: perform some basic research before disqualifying prospects

aps@cmswatch.com(Alan Pelz-Sharpe) — Wed, 3 Feb 2010 09:14:00 -0500

Our job here is to advise buyers of technology, never vendors. But today I am going to make an exception and give the vendors out there a solid tip.

Do your homework before responding to RFPs.

Now that we have search tools like Bing and Google, ten minutes desk research can return a lot of information. If you receive an RFI or RFP from a firm you have not previously worked with, then it makes good sense to check them out and get a feel for their structure, size, and potential long-term needs.

I only mention this as recently one of our clients received a (belated) response to an RFP from a very well known vendor, one that essentially whined about the thoroughness of the buyer's procurement process, and (not very subtly) implied the buyer could not really afford their products anyway.

In fact the buyer is a sophisticated organization with advanced requirements, one that can more than afford the product. On paper at least, that vendor offered an excellent fit. Rather than asking the buyer to justify themselves, a quick bit of desk research to qualify them would have done the job. But one lazy and somewhat insulting e-mail later, this particular vendor now likely lost all hope of landing what could have been a very nice deal indeed.

Many Portal Products Getting Refreshed

adurga@cmswatch.com(Apoorv Durga) — Tue, 2 Feb 2010 10:23:00 -0500

Last week we released some significant updates to our Enterprise Portals Research. The latest research reflects significant changes among new versions of several portal platforms, as well as additional details for all the products we cover.

Our research found the market in a particularly unusual state. Buyers should exercise particular caution as many portal software vendors are significantly updating their software. You can read more about the market’s state of limbo in the press release. As our Cross-Check below illustrates, IBM, Oracle, and Microsoft, along with open source options, are all at an overhaul and refresh stage.

Click to enlarge

IT Executives who are procuring a new portal solution typically do not want to be the first to deploy brand new software. Rather, they prefer a product that has somewhat matured. With so many major vendors in transition, the market has entered a kind of limbo phase.

Our research also found a healthy open source market (almost half of the portal products we cover are open source). We also track some new technical developments, such as whether and how different portal tools integrate with other applications as both consumers and producers of services. If you are not currently a subscriber, you can download a free sample here.

SAP to resell EMC Documentum

aps@cmswatch.com(Alan Pelz-Sharpe) — Tue, 2 Feb 2010 08:45:00 -0500

Last week SAP announced that it would begin reselling EMC Documentum products to the Insurance and Finance industries. It's not a world-shaking announcement, but it is interesting for one simple reason: buyers in those sectors now have a choice. Whereas previously they would have only had the choice of Open Text (SAP's preferred partner), now they can opt for Documentum.

It's worth pointing out that theoretically, SAP customers have always had options other than Open Text, insofar as every major document management vendor markets a connector to SAP, typically for many years now. But when Open Text is the only document management and archiving option on the SAP price list, and is actively sold by SAP reps, then it created the misimpression of "Open Text or nothing."

So moving forward buyers in these sectors have a choice, and choices are a good thing. You have more leverage to negotiate and more chance of finding the right fit. OK, so it's only one other vendor, but when the choice previously was something (Open Text) or nothing, any choice is surely better than none.

Would EMC really buy FatWire?

kthomas@cmswatch.com(Kas Thomas) — Sun, 31 Jan 2010 08:47:00 -0500

From time to time, rumors surface about Web CMS vendor FatWire being up for purchase. Also, from time to time, we hear rumors about EMC Corporation being ready to acquire this or that content technology. Now there's a rumor going around fusing the two: that EMC might acquire FatWire.

As with most rumors about alleged impending acquisitions, we tend to discount this one. I certainly have no inside information. On the surface, it would seem to make little sense. EMC already markets a Web Content Management offering in Documentum. And FatWire's (roughly) $40 million-per-year revenues would add only a drop or two to EMC's $4 billion-per-quarter stream.

But I'll play devil's advocate for a moment. We do know that FatWire and other upper tier WCM vendors have competed successfully against Documentum for new deals, occasionally displacing Documentum as incumbent. As with Open Text buying Vignette, EMC taking over FatWire would remove a competitor from the market, and give them more credibility in a strategic area where their own product seriously lags.

But the key factor for EMC is always storage. EMC is a storage company first and foremost. Almost every acquisition EMC makes (and it has made plenty: at least 38 companies since the December 2003 acquisition of VMWare) plays into its cloud or SAN storage stories. From a direct revenue-acquisition point of view, acquiring FatWire would not be strategic. But from a downstream storage-revenue point of view, it might very well be. EMC may know something we don't about how much money FatWire's customers are spending -- and plan to spend -- on storage.

Again, I have nothing specific with which to corroborate the EMC-FatWire rumor. However, it's interesting to speculate based on what we do know. We know that EMC is on a multi-year, multi-dozen-company, multiple-billions-of-dollars acquisition spree; we do know that FatWire and others have been taking business from Documentum in the WCM market; and we know that FatWire has lately seen some key people leave (among them, CTO Dmitri Tcherevik and Director of Corporate Communications Rita O'Brien), which is sometimes a prelude to M&A activity. We know, too, that web content has been getting richer and bigger (driving people to spend more money on storage). And finally, FatWire does have a Documentum connector.

But still. EMC? Buy FatWire? Really?

Let's put it this way. Stranger things have happened.

New Enterprise Portals Research

jgingras@cmswatch.com(Jarrod Gingras) — Thu, 28 Jan 2010 09:40:00 -0500

Yesterday we launched our most recent Enterprise Portals Research. This report critically evaluates 13 portal vendors and products, which we break up into 2 categories:

Infrastructure Vendors

IBM: WebSphere Portal Server 6.1.5
Microsoft: Microsoft Office SharePoint Server 2007 SP2
Oracle: WebLogic Portal 10gR3
Oracle: WebCenter Suite 11gR1
SAP: SAP NetWeaver 7.0 Enhancement Package 1
RedHat: JBoss Enterprise Portal Platform 4.3

Specialist Portal Products

Open Text: Vignette Portal 8.0
Broadvision Portal 8.1
eXo Portal 2.5
GateIn 3.0
Liferay: Liferay Portal 5.2.3
Plone: Plone 3.3.2
uPortal 3.1.2

Our annual subscribers and recent customers will receive their updates automatically. You can also download a free sample of any of our evaluations.

By the way, this release marks the end of an era for CMS Watch -- and the beginning of a new one. This will be the last "big-bang" update that CMS Watch releases in the current format. With your feedback, we are embarking on new research delivery model that emphasizes online access to the latest updates, along with more avenues for obtaining practical advice for your specific problems.

Rest assured that buyers of this report and any of our existing research in the weeks to come will automatically receive access to our new services. We're extremely excited about the new model and we think you will be too. In the meantime, I hope you'll join the club by signing on to one of our research streams.

Oracle Sun Update

adurga@cmswatch.com(Apoorv Durga) — Thu, 28 Jan 2010 08:10:00 -0500

While Steve Jobs was introducing Apple's iPad, Oracle was explaining its own approach to hybrid hardware/software offerings. Oracle yesterday announced that it has completed its acquisition of Sun Microsystems. Oracle now provides a complete stack consisting of Storage, Server Hardware, Operating Systems (Linux and Solaris), Database (Oracle and MySQL), Middleware (Fusion), Programing Language (Java), and Applications.

In general across the Java and Middleware areas, other than Java, most incumbent Oracle offerings will remain "strategic," whereas their counterparts from Sun will be merely supported. Here's a quick take on some of the announcements that might impact content technology users:

Oracle will keep investing in Java and plans to unify J2ME (for mobile) and J2SE. Currently it is not easily possible to create applications in Java that will run uniformly on computers and mobile devices. So this will help take Java closer to the "write once, run anywhere" promise. Oracle also plans to include support for IPTV and Blueray devices in Java.
On the application server front, Oracle will share some technology pieces between WebLogic and Glassfish. Oracle WebLogic continues as the strategic application server platform and Sun's Glassfish will serve as the reference implementation for "tactical applications." Similarly, Sun's NetBeans will focus on "lightweight development," whereas Oracle's JDeveloper will be Oracle's IDE of choice for enterprise application development.
As expected, Oracle's WebCenter Suite (we cover it in our Enterprise Portals Research) will remain the company's strategic Portal offering. Sun had previously announced a partnership with Liferay for their own version of Portal (called Sun Glassfish Webspace Server) based on Liferay's codebase. Oracle says it will continue to maintain that initiative and contribute extensions back to the community.
Oracle will also retain Open Office and come up with a web-based office suite. There is obvious potential to integrate with Oracle's Content Management suite here.
MySQL will become part of a separate Open Source business unit within Oracle. The company discussed plans for some integration between the Oracle database and MySQL, probably in the area of management services.

These announcements did not hold any big surprises. Oracle's approach is edging closer to "everything in a box," something I mentioned on my personal blog last year. From Oracle's perspective, it gives them an opportunity to align engineering efforts to make their middleware perform optimally on their hardware and OS.

For customers, it could work both ways -- depending on your exposure to Oracle. On the one hand, you only have to deal with one vendor for support of your entire IT landscape. But on the other hand, by hedging all your bets on one vendor, you're increasing vendor lock-in.

In his address, Thomas Kurian of Oracle mentioned "...our middleware strategy has been simple and clear..." Well, if by "simple" Oracle means they will support every acquired product, resulting in a potpourri of applications, it is indeed straightforward. For customers, the line between "supported" and "promoted and enhanced" is quite important.

As with every acquisitive vendor, Oracle is talking a lot about interoperability among its various components. In fact, Oracle says it will spend $4.3 Billion on R&D in the first year to try to make sure the complete stack delivers. We will be watching, and sharing key advice with our research customers.

When your systems integrator picks your vendor

tregli@cmswatch.com(Theresa Regli) — Wed, 27 Jan 2010 08:10:00 -0500

Earlier this week I had an advisory call with one of our gold subscribers who's in the process of creating a short list of vendors for Web Content Management. A US-based health care company, they're looking for potentially one solution for both their public-facing website and their intranet: two rather divergent scenarios, where not all WCM tools necessarily present a good fit.

At the same time, the company will be working with a systems integrator / creative agency to redesign their web properties. The conversation turned to the question of whether or not it made sense to get the agency's input on what CMS they should pick, or even guide the process. The answer is one so often uttered by advisors such as myself: "it depends."

Most systems integrators have tight relationships with specific content management vendors. They train their developers on specific tools. They have teams of specialists focused on implementing it, and SIs send that team to specific vendor conferences and training. Their CEOs and salespeople play golf together. In some cases, systems integrators are even financially incented to recommend a particular vendor, even if it's just a lavish dinner on the town. Or, an SI might specialize in open source, and wouldn't think to recommend a commercial product that's a particularly good fit (or vice-versa).

As a former Director of Content Management at a systems integrator myself, I was occasionally pressured to recommend Interwoven products (my former employer's primary CM partner at the time) in any and all situations, because it was the tool we knew best. (I'm pleased to report that since then, my successor has cultivated a greater diversity of relationships with CMS vendors). I recall one behind-closed-door argument between myself and an Interwoven salesperson, because a colleague and I knew a particular client requirement could be fulfilled with TeamSite (a product our client already owned), but the salesperson wanted us to recommend Interwoven's WorkSite document management product as an upsell, because it was "a better tool for the job." We stood firm in the best interest of the client, and Interwoven was furious. Sorry guys, but that nice round of golf we played together wasn't going to turn me into your quota-reaching vehicle.

So while systems integrator partnerships with content management vendors may skew who and what systems integrators recommend, it doesn't always. If you get a systems integrator or creative agency involved before you pick a product, be sure to understand who they tend to work with. Have an up-front discussion about who they tend to recommend, and why. Ask them if they're incented by any vendors to make specific recommendations. Check to see who sponsors their events. You may be surprised to find they're the same vendors they tend to recommend.

At the same time, you don't want your website to become the testing ground for an outside firm's first experience with a particular piece of software. Integrators generally get adept with a particular tool after three or more implementations. Of course, this doesn't stop their developers from wanting to learn new tools -- perhaps at your expense. So experience matters too. In the end, your comfort level with your preferred SI may trump your choice of the best-fitting software.

Above all, make sure your systems integrator holds your best interests above those of their partners -- and their developers.

EPiServer goes public

bloem@radagio.com(Adriaan Bloem) — Wed, 27 Jan 2010 07:13:00 -0500

Like nearly all its competitors, Swedish Web CMS and Social Software vendor EPiServer is doing well financially. The company reports nearly $30m in revenues over 2009 and now boasts almost 3,000 customers. So perhaps it's no surprise that last week, EPi announced its intention to go public (on the Stockholm Exchange).

Of course, we're technology analysts, not financial analysts. But in our reviews (we cover EPiServer CMS in our Web CMS Report, and a new EPiServer Community evaluation will soon be added to the Enterprise Collaboration & Community Software Report) we always pay attention to the figures, as well: these can be essential to judge the viability of a company and its products. In that respect, EPi's public offering is good news, since it means the company will be required to publish growth and revenue data. (That doesn't mean there won't be spin on the numbers: the press release mentions EPiServer "has grown twice as fast as the WCM market in general," which seems hard to quantify.)

Of course, it's quite possible that for EPi itself, a much more important benefit is that a public company has a lot more maneuverability. Readers of our reports will know that EPiServer has steadily been expanding its portfolio with acquisitions over the past few years. That's going to be a lot easier once EPi's venture capitalists have cashed in and the company can trade on its shares. So watch this space -- I have a feeling EPiServer will soon be adding more .NET products to its line-up.

But also, watch your relationships at the vendor.Sometimes after a respectable number of months following an IPO key company staff cash in and leave. That's not typically a disaster, but EPiServer customers should maintain multiple points of contact just in case.