As a buyer it can be confusing and difficult to stay updated on the rapid market developments, so I made a slide showing who the big vendors were in 2007 and what the picture looks like today for a recent J. Boye group meeting.

Large, global and complex organisations tend to gravitate towards the big vendors. As one of our members put it:

Elephants buy from elephants

The elephants: The big CMS vendors in 2007 and 2012. Click for a larger version

Consolidation has only happened to the extent that vendors have bought other vendors. With just a few exceptions all the products have been kept alive, so today several of the above have more than one Web CMS as a part of their offering.

Finally, a look at the big vendor websites won't get you far in terms of figuring out more about their CMS offerings. They may be big vendors, but they also offer many other solutions and CMS is apparently not on top of their list. Here's the CMS products from each:

• Adobe: CQ5 (I gave up looking for it on adobe.com and instead turned to day.com, which is the CMS vendor they acquired in 2010)
• Autonomy: TeamSite (which came via the Interwoven acquisition in 2009 and Autonomy is now actually owned by HP)
• EMC: Documentum (which they acquired in 2003)
• IBM: Web Content Manager (which belongs to the Lotus family in IBM software)
• Microsoft: SharePoint
• Open Text: Web Experience Management (which came via the Vignette acquisition in 2009)
• Oracle: WebCenter Sites (formerly known as FatWire which they acquired in 2011)

Learn more

Read about who's big in the European CMS marketplace.

If you are working with CMS as a big part of your job, then consider joining our CMS Expert groups. CMS and all the related challenges is also a regular topic in many other J. Boye groups for web & intranet professionals.

CMS is also a hot topic on the web content management conference track at J. Boye Philadelphia 12 on May 10.

I recently had a conversation with Thomas Ørgaard Bredgaard who is Department Head, Web Management at the Denmark-based headquarters and a member of our groups for online professionals. He shared several critical and hard-earned lessons on performance and how it relates to a global, complex and large business-to-business company.

The Grundfos website performance challenge

Unlike media or transportation companies who experience extreme peaks due to breaking news or inclement weather, Grundfos as a B2B company did not have to worry so much about load balancing to handle peaks. With up to 6 million visitors projected next year to Grundfos websites and a distributed IT infrastructure led from the headquarters in Bjerringbro, Denmark, the key challenge in the recent website redesign was to ensure that the website was equally fast for a customer or distributor visiting the site from Beijing, Boston or Brisbane.

As illustrated by this recent tweet from user experience expert Jared Spool, slow performance can lead to bad publicity. For Grundfos, the main concern was users leaving the site for competitors or resorting to more expensive phone calls or e-mails for product inquiries.

Besides the global reach, Grundfos also has a website packed with several different content types, including videos and several applications. This required extensive testing and several layers of performance improvements to get right.

Further complicating the matter was the lack of industry best practices when it comes to performance guidelines. As Thomas said:

We were unable to find any standards or reports detailing performance requirements for a large, complex and global B2B company. To establish a working definition of acceptable performance requirements we had to get inspiration from the consumer space and vendors specializing in content delivery.

According to Thomas, Grundfos used the below 2 sources as main inspiration to define acceptable performance:

• Guidelines from Jakob Nielsen on response times, where 1 second is used as the limit for the user's flow of thought to stay uninterrupted,
• Findings from a study commissioned by Akamai, where between 2 and 4 seconds are defined as the "threshold of acceptability for eCommerce Web Page Response Times"

The many elements of the solution

To make sure that the new website would be delivered within a time-frame of 2,5 to 5 seconds, depending on content type, Grundfos had to undertake several actions, some complicated and technically demanding, while others were more about project management.

A key step according to Thomas was creating a cross-functional web performance team including members from IT and business units. As Thomas said:

This team meets regularly to review performance improving initiatives and also discuss any issues, such as areas of the website which are performing poorly. The main focus of the team is to find potential improvements, whether governance initiatives to prevent large images on the frontpage or purely technical issues

Here some other steps which Grundfos have implemented:

• Performance goals have been built into the service level agreement (SLA) with the IT department, profiling both up-times and load-times
• To measure and benchmark performance globally, Grundfos decided to work with Gomez, a vendor specializing in web load testing, cross-browser testing, and web performance management
• The relaunch was based on a new Web CMS (Day CQ5). During the Day CQ5 CMS implementation Grundfos used web performance best practices from Google and Yahoo Yslow to keep the focus on performance from the beginning. Testing happened often and early to make sure that potential slow areas could quickly be identified and addressed
• During the project planning the concept of static vs. dynamic delivery, also known as "frying vs. baking", was carefully considered. In brief, the idea is to ensure that only those website elements which are truly dynamic, e.g. application output, are generated at every request, while other static elements (e.g. the Grundfos logo and press releases) are published to a server for quick delivery. Day CQ5 has several performance guidelines and Grundfos implemented caching as well as static publishing as much as possible
• jQuery, a fast and concise JavaScript Library, was implemented using the Google Libraries API, which is a content distribution network and loading architecture for the most popular, open-source JavaScript libraries

In addition, Grundfos is currently considering using a content delivery network, such as Akamai, to leverage a globally distributed hosting infrastructure.

Learn more

If you don't have a large and globally distributed IT infrastructure like Grundfos, another option for performance improvements may be cloud computing.

Besides quick response-times and doing what you can to makes your website fast, you should also consider up-times, to ensure that the website is actually up and running most of the time.

In a recent eConsultancy article, slow load times were cited as one of 25 reasons why I'll leave your website in 10 seconds

Join our community for online professionals and benefit from the insights of experienced online professionals whilst cutting out the time consuming digging and the awkward small-talk.

As usual when this happens, the vendor to be acquired is very optimistic about the future prospects. To quote from a message to Day Software customer and partners:

this will serve as a new catalyst for Day to accelerate our investment in product innovation, our customer community, and our global ecosystem of channel and technology partners

The Day management team is understandably in a happy mood, but when it comes to product improvements, Day really has a case to prove over the coming months. Most in our community did not see this as good news for the product or for the standards that Day has been championing. I've been talking to customers in the past weeks and here is our take on what the acquisition could mean to customers:

• Until the deal has been closed, Day will continue to operate as a separate company with its own partner and customer community
• Product updates are likely to be delayed: Day is known for letting itself distract by such things as open source and standard involvements with less focus on delivering a solid and easy-to-use product. It will now inevitably be seriously distracted for a while, whilst getting to know the ins and outs of the new owner and whilst trying to teach a large sales force how to work enterprise web content management deals.
• Don't expect CQ5, Day's main product, to receive much attention inside Adobe. To put things in perspective,  Day is a very small fish indeed next to the whale that is Adobe. Adobe has 8,600 employees compared to Day's less than 150. Compared to previous acquisitions by Adobe, e.g. Macromedia (1,400 employees) and Omniture (>1,000 employees), this is quite a small deal.
• No overlap with existing Adobe products. Quite unlike recent industry acquisitions, e.g. when Open Text bought Vignette or when Oracle bought Sun, there seems to be no product overlap. As part of the expected integration of the two companies, Day will operate as a product line within Adobe's Digital Enterprise Solutions Business Unit
• Adobe emphasised integrating proprietary technologies such as Air and Flash in the acquisition announcement and did not mention anything about Day's work on standards. This was quite worrying to several Day customers.
• CQ5 moving down in the market. At the moment CQ5 is overkill for anything but complex and global websites. Expect that to change when Adobe begins to put its engineering footprint on the product.

What should you do as a customer?

1. Use the change and added uncertainty to get discounted rates or freebies. Day has an expensive product but is known to be flexible on licensing models
2. Keep reminding your Day Software contacts of your existance, while they are busy closing the deal. Expect some new faces in 2011 as the organisational integration kicks in and as always make sure you get an experienced team on your project.
3. Prepare to listen to some funny stories from the competitors. Usually some use an opportunity like this to spread some good old FUD (fear, uncertainty and doubt). Fellow Swiss-vendor Magnolia have already started this with a public blog: Day to be acquired by Adobe - implications?

More detailed and interesting coverage of the deal:

• CMS Wire: Perspectives: What the Adobe + Day Software Deal Means, Part 1
• ComputerworldUK: Will Adobe See the Light (of Day)?
• Jon Marks: A Fine Day For Adobe
• mwd advisors: Adobe acquires Day Software to flesh out enterprise portfolio
• Real Story Group: ECM Perspective and DAM Perspective

and finally for extra credits read: In defence of PDF, an interesting perspective from a new Day employee saying that "Day has a strong prejudice in favor of HTML as the one true and proper Web format for documents".

Based on our extensive experience with CMS selection, we have created the below Top 10 list with vendors you should always consider. This is geared towards buyers from large and complex organisations with significant web demands.

What it requires to be on the list:

• Significant dedication to CMS. It does not have to be everything the vendor does, but to mitigate your risk, CMS has to be very important to them. This includes a history of relatively smooth upgrades combined with on-going technology investments in improving the system.
• Global footprint. You can either find direct vendor representation or experienced partners in almost all parts of the world to help you with the implementation. There are also successful references around the world for you to learn from.
• The vendor has something very significant to offer. This easily turns into yet another unhelpful long list, so we kept the list short and predict that the list will change in 2010. A vendor can only get on the list if we can remove another one. This means that many vendors, even though they might have interesting references, are not on this list.

You can reduce the list further by considering licensing and technology. Some on the list might also not have local partners in your region. If you feel troubled by suddenly having too few vendors, remember that you also need to find a good implementation partner to support you. To find the right one, you should send to more than one implementation partner for each vendor; this way you will easily end up with 10 - 12 qualified companies on your list.

Here a few comments about some of those missing from the list:

• Microsoft is not on the list as neither SharePoint nor Oxite are good fits for Web CMS. Despite tremendous adoption, SharePoint is often chosen for the wrong reasons. Also, as mentioned on this blog, content management does not seem important to Microsoft. For additional details, you can consult our research on Best Practices for Using SharePoint for Public Websites.
• Several other large vendors are absent, eg. Autonomy, IBM, Oracle, as they are often simply overkill for Web CMS. Not only are their products very expensive, but they are also very complex to implement and use. We challenge buyers who insist on adding them, that they carry additional risk due to the CMS being acquired from smaller vendors and their diminutive focus on WCM in the overall picture.
• Many significant, but still regional vendors, eg. CoreMedia, e-Spirit and Terminalfour are left out as they do not yet have a global footprint. There are regional differences in the market, which we will cover in separate forthcoming blogs.
• Alfresco has very good marketing, in particular for an open source vendor. The actual product is quite complex with weak usability and many on-going architectural changes.
• Joomla lacks a few important features such as workflow, custom roles and custom content types. This combined with security concerns means that we do not always recommend Joomla.
• WordPress is a very popular blogging platform, which might slowly be morphing into a CMS, but is still lacking in many enterprise features, including security. In too many regions it is also quite difficult to find any significant SI that offers WordPress implementation support.

Most CMS vendors are having a great time, c.f. recent earnings from Day Software, FatWire and Sitecore, but I'm hoping this list will help you save some time and confusion while navigating a still very crowded marketplace.

I welcome your feedback and stay tuned for regional shortlists soon!

Thanks to James Hoskins (@jameshoskins), John Goode (@johngoode), Jon Marks (@McBoof) and Mark Morell (@markmorrell) for valuable input.

UPDATE:  Aug 18 - In response to popular demand, I've released a wrap-up with additional background on the shortlist

Revenue was up 33 % and license revenue accounted for 42 % of revenue, maintenance and support (M&S) revenue for 32 % and services for the remaining 26 %. A few months ago, Michael Marth, a friendly Day engineer, shared an interesting insight into Day's customer retention success when he documented the lifetime of a CMS installation to be more than 6 years for Day customers. This certainly helps in relation to the M&S revenue.

While Day has been known to continuously invest in research & development (R&D), a closer look at the operating expenses reveal that R&D costs are down roughly 5 %. Sales and marketing costs are up more than 10 % and general and administrative costs are up more than 20 %. Perhaps Day is hoping that the increased spend on sales and marketing will translate technical leadership into market leadership?

After Day released what analyst Kas Thomas at CMS Watch characterized as not-so-sunny financial results in February, quite a few of our community of practice members have asked us to keep a watchful eye on the Swiss CMS vendor. Interestingly, CMS Wire had a more positive take in a story titled Bright and Sunny 2008 Financial Results. Irrespective of your interpretation of the numbers, Day deserves recognition for being transparent and more open about numbers than most competing vendors.

With software sales at €4,7 million for the 6 months, Day has quite a global footprint with dual headquarters (in Switzerland and the US), 4 regional offices in Europe and 1 in Singapore. Based on software revenue, Day is a slightly smaller than Sitecore, which has experienced rapid growth and Day is significantly smaller than FatWire, which recently reported a profitable year and revenue increase. If you look at the recent and (too) influential Forrester Wave for WCM vendors, Day was indeed the smallest dot on the chart.

Many potential buyers, particularly in Scandinavia, have so far been reluctant to consider Day, mostly due to the lack of local experienced partners. According to Kevin Cochrane, Chief Marketing Officer at Day, who I recently interviewed, a revamped partner program will soon be introduced in order to win new system integrators and agencies in regions where Day has historically been weak.

Cochrane also reflected briefly on the management changes that Day experienced in 2008 with a new CEO and a new CFO. He saw this as a very positive change for the company and he mentioned that there had been no change in R&D, even with the cost reduction. According to Cochrane, the number of engineers actually went up as Day had some quite extraordinary costs related to the CQ5 launch in late 2008.

As always, great financial results from a vendor is not necessarily an indicator that the company's products fit the specific requirements of your organisation. When evaluating vendors, ensure you compare features as well as intangibles, e.g. roadmap, community, partners and finances. If you are very close to a decision, consider doing reference calls to some of the 27 new customers to check in on their progress.

Disclaimer: I’m no financial analyst, but the financial standing of a vendor is important when evaluating. So remember to look beyond the product features.