Collaboration and Content Strategies Blog

Why Google Apps Hasn't Taken Off in Large Enterprises

Blogger: Guy Creese

Last week, Fortune magazine published an article ("Google's tough sell to Corporate America") and blog entry ("Gen Y takes to Google Apps") on Google Apps, noting that it was popular with Gen Y but not with large enterprises. To put it in perspective, Fortune noted that Microsoft made 3,000 times more revenue than Google off of its Office software: Microsoft sold $12.2 billion of Office software in 2007; Google sold $4 million of Google Apps. Large enterprise sales have been few and far between. Fortune notes:

So far, though, the largest Fortune 500 company to use Google Apps is Sanmina-SCI (SANM, Fortune 500), an electronics manufacturing company that has 900 of its 45,000 employees using the full package of applications. Another paying customer is Valeo, a publicly-traded French automotive supplier. And General Electric (GE, Fortune 500) is running Postini, an anti-spam technology that Google acquired last year, for its 300,000 employees.

I'd predicted this would be the case when I published my first report on Google Apps a year ago (Burton Group clients can access the April 2008 revision here). So why is that? Google Apps has done well within small- and medium-size businesses (SMBs); large universities such as Arizona State and Northwestern have also adopted it.

It comes down to four reasons. First, Gen Y is not yet in charge of IT strategy at large corporations. Google is banking on Gen Y employees demanding Google Apps--"'We’re not waiting 15 to 20 years. The generation of new [office apps] users is already here,' says Glotzbach"--but there's a difference between a minority of employees (without seniority) pushing for them and the older generation of CIOs mandating them. While small companies headed up by Gen Y workers are happily using Google Apps, large enterprises headed by 50- and 60-year olds who are used to Microsoft Office are taking a wait-and-see attitude.

Second, Google's consumer-centricity has done it a disservice here. Google basically took a consumer service, slapped an Enterprise name on it, and assumed companies would lap it up. Unfortunately, Google neglected to add features (so far) that competing packages have offered for more than a decade, such as tables of contents, footnotes, sophisticated e-mail distribution lists, and records management capabilities. (Although Fortune quoted me as saying that the lack of footnoting was a major deficit [I'm sure because it was an example everyone would understand without needing an explanation], the lack of records management capabilities for documents and the need for better distribution lists are the bigger problems.)

Third, the incremental benefits of using Google Apps are not that great. True, you can edit documents concurrently--which you can't do in Microsoft Office--but otherwise Google Apps is pretty much a straight copy of base Microsoft Office functionality. In other words, Google Apps makes do by copying the market leader in a rudimentary way, rather than taking the gloves off and pounding away at the things Office doesn't do: keeping track of who's reading what, suggesting other documents it might be useful to refer to, recognizing that a document similar to this one has already been written and suggesting that it might be easier just to modify the earlier document. Software as a Service (SaaS) office productivity suites can do so much more than software because they can watch the community working, and thereby help the community leverage what it already knows. Amazingly, Google doesn't take advantage of that architectural difference.

Fourth, large enterprises are behemoths--enterprises aren't inclined to change and there's nothing compelling in Google Apps at the moment to make them change (and the attractive low price isn't enough). Word processors replaced typewriters not because they did the same thing, but because storing digital documents made revisions so much easier--word processors offered a quantum leap in productivity. Google Apps doesn't offer that productivity leap yet.

When it does--or a Google competitor does--I think large enterprises will do well to adopt the new technology. Until then, more nimble SMBs and cash-crunched universities will be the sweet spot for Google Apps.

Uh, Has Anyone Thought About Licensing?

Blogger: Guy Creese

It's been an interesting week or so in terms of movement towards, "I don't care where the system is, I just want it to run." Microsoft announced it was relaxing its virtual machine licensing requirements (CCS commentary here) and Cisco announced it was acquiring PostPath (CCS commentary here) as a way to offer Outlook-friendly SaaS-based e-mail.

However, as we move towards a world where IT departments deliver solutions that can run anywhere--in-house or in a condo (on a virtual machine) or in the cloud--one thing that all vendors quickly need to figure out is how to license all of this in a coherent fashion.

Back when the dominant model was software upgrades that came out every 9 or 18 months, enterprises and vendors could afford the luxury of sitting down and spending days or even months hashing out the licensing details. (I had one client recently tell me that he's spent nine months in his latest licensing go round with Microsoft and it's still not finished.) However, virtualization means that virtual servers can come and go within minutes, and workers can sign up for SaaS solutions within minutes as well. Enterprises also want to move towards an instant provisioning model, where they may have n servers in house but then contract with the cloud to augment those servers for a month or so for a specific project.

In other words, when it comes to licensing, rigid paper contracts have to give way to online checkboxes. The business decides, "OK, we need 250 SaaS-based e-mail mailboxes for the company we acquired yesterday," it ticks off the required checkboxes, and the online mailboxes get provisioned within the hour, the licensing dashboard is updated automatically, and a correct bill goes out the next month. In less than a day, the system and affiliated licensing is taken care of.

It may seem like a dream now, but this is where companies such as Cisco, Google, IBM, and Microsoft all need to get to. And I would argue the company that does it first and well will gain a huge competitive advantage.

Another 800 LB Gorilla Jumps In - Cisco Acquiring PostPath

Blogger:  Bill Pray

Microsoft and IBM have long dominated the enterprise messaging market, with Novell and Oracle holding onto a small and loyal set of customers. Google continues to threaten to enter this market but, as my colleague Guy Creese points out, still has a ways to go. This market hasn't changed much in several years, other than it's slow evolution in becoming integrated with hot new technologies like social networking. Now Cisco has jumped into the enterprise messaging market with the announcement of the acquisition of PostPath.

On the surface, Cisco has the size and resources to potentially challenge Microsoft's and IBM's dominance of the enterprise e-mail market. However, like Google, Cisco has a ways to go and displacing Microsoft and IBM as a preferred e-mail solution vendor will be very difficult. 

The e-mail market is established. To win, you have to take market share from Microsoft or IBM.  Their solutions are mature, integrated, and entrenched. Once in place, e-mail solutions tend to stay. 

Cisco's software-as-a-service (SaaS) approach, ability to tie the PostPath solution to their existing collaborative technologies (notably WebEx), and knowledge of enterprise IT give them a shot taking market share. Initially, it is clear that Cisco is targeting Microsoft. 

It won't be easy. Cisco has to answer the questions of why and how for enterprises. Why would an organization want to incur the cost to leave an e-mail solution that works for them today and how would an organization actually migrate to the Cisco solution?   

Both Microsoft and IBM are launching SaaS offerings that include e-mail and are meeting their customers' enterprise messaging needs. Cisco is going to need a strong "Why" story. And if they don't offer an easy "How" answer, then even a good "Why" story might not be enough to persuade many customers to leave their current solution. But, hey, this is technology. Anything can happen.

Note:  This is a cross-posting from the Le Collaborateur blog.

Email Overload: A Little Help From Microsoft

Blogger: Craig Roth

In June, Google announced an "Email addict" feature that was kind of a gag response to people complaining about email overload.  When you press a "take a break" button, the screen turns gray and locked the user out of email until you clicked again.  I had posted my own suggestions of how an email tool could help with email overload at http://knowledgeforward.wordpress.com/2008/06/08/google-lands-crushing-blow-to-email-addiction-with-new-feature/.

I was just turned on to Email Prioritizer from Office Labs which seems like a nice (and real) response to Google’s gag approach with its “Email Addict” feature. It hits on one of the features I wrote about: mail arrival schedules. I’d also recommend that Microsoft add automated scheduling options (hourly, morning/noon/evening, etc) to the manual option provided. This would simulate the cycle of the postman coming to deliver the mail, and leave your brain free outside those times to concentrate.

One nit: the description of the tool on their website annoyingly equated “do not disturb” as allowing you to “work without interruptions”. Unless you have toasts turned on email doesn’t interrupt you. And if that bothers you, the feature is already there to turn them off. I’d say email is a distraction or a temptation, not an interruption. The reason I’m picky is that there is a lot of great research around “interruption science” (for example, see interruptions.net) that mostly can’t help or doesn't apply to this situation.  One needs different approaches and has different goals and metrics when dealing with distractions versus interruptions.

Note: This is a cross-posting from the KnowledgeForward blog

Virtualizing Exchange? Maybe not...

Blogger: Bill Pray

Microsoft recently made changes to its licensing program for Microsoft Server Products in Virtual Environments. My colleagues, Chris Wolf and Richard Jones, commented on the changes in licensing from the data center perspective. However, as part of the announcement, Microsoft also published a support statement for virtualizing Exchange. While the statement is very detailed, it led me to an important question for Microsoft – or as one of the Microsoft folks I talked with aptly put it: “This announcement begs the question of whether MSFT recommends virtualizing the Mailbox server role.”

The answer? “The answer is that while the Mailbox server role is supported in a virtual environment and customers are permitted to do it, we generally recommend customers run Mailbox servers on physical hardware.” 

The reasons are understandable – Mailbox servers have high server transactional costs that lead to high CPU utilization, heavy disk IO, and large quantities of attached data. 

The support statement doesn’t contain this advice, so after further conversation with the folks at Microsoft, I received the following clarifications: 

• Virtualizing Exchange in test, development, or small deployments (branch office, for example) is not a bad idea because the load is lighter.
• The role of the Exchange server needs to be considered – a mailbox server is a role where issues may arise, but an edge transport or client access server might be a good candidate, depending upon the architecture of the implementation.
• You can throw more hardware power at the mailbox server role, depending upon the load, but the drawback may be that you end up paying more for the hardware than if you didn’t virtualize.  The mathematics of the TCO need to be examined in this scenario.
• Virtualizing Exchange narrows your storage options, like deploying on a SAN, and may force smaller mailbox quotas. 
• Avoid the mistake of making the virtualized Exchange a single point of failure. 
• Note that support for virtualized Exchange 2007 is for SP1 or better.  The Microsoft folks verified that this is not because of specific fixes in SP1, but rather to take advantage of the fact that 2007 SP1 is the version supported on Windows Server 2008 with Hyper-V. 

The folks at Microsoft offered some recommendations for virtualized Exchange also:

• Follow the guidelines for implementing that are contained in the support statement.
• Use tools to load test before putting into production.
• Cluster continuous replication is a good idea.

The sum? Virtualizing production Exchange mailbox servers is, generally, not a good idea from a technical perspective. When I asked the folks at Microsoft if they were going to use virtualization for Exchange Online, their answer was no. They said it would increase their management load and tweaking Exchange and AD to run multi-tenant was more cost effective.

On the Internet, Nobody Knows You're a LAMP

Blogger: Larry Cannell

My recently published Burton Group overview, which discusses open source 3C (communication, collaboration, and content management) solutions, is entitled "Open Source Communication, Collaboration, and Content Management: Cutting-Edge Innovation, Low-Cost Imitation, or Both?".

The overview provides examples of open source 3C solutions which are innovative and responsive to market needs. It also identifies key signals and project traits to look for when evaluating open source 3C products as well as the challenges they still face.

The paragraph below is from the introduction:

"The need for IT innovation has never been greater but the cost of cutting-edge solutions from established enterprise vendors is often prohibitive. Given today's globally competitive markets and tight budgets, IT organizations can often rationalize, to themselves and their business customers, that these solutions are not affordable and decide to wait until the financial situation improves before moving forward. Unfortunately, these delays impact IT's ability to support critical business initiatives and strain credibility with business partners.

"However, these proprietary solutions enterprise IT organizations are waiting to buy have rarely taken the pounding that some open source solutions are taking today. While software vendors will proudly point to 200 installations with thousands of users, some open source 3C solutions are being used by millions of users through thousands of websites across the Internet."

Millions of people are using open source 3C products on the Internet every day. Readers of this blog probably use a number of them regularly, although may not be aware of it.

So this begs the question: if a product is good enough to serve the needs of a few million users why wouldn't it work on your intranet?

btw: the title of this blog post refers to a famous cartoon caption and the bundle of software used by many open source 3C solutions.

Participation In Social Networking Contextual Research Project

Blogger: Mike Gotta

Social Networking within the Enterprise

Burton Group is conducting a qualitative research project to assess organizational and technology trends driving interest in social networking within the enterprise. A mix of face-to-face and telephone interviews will be the methods used for data collection. Our goal is to have an open-ended dialog with participants from business units and IT groups. The project is open to Burton Group clients and non-clients alike. Based on the response, there may be a limit to the number of candidates selected. Some participant roles we are looking to interview include:

• HR managers
• Line-of-business managers
• End users
• IT managers, architects and project leaders involved in such efforts
• Executive sponsors of initiatives related to social networks (e.g., retiree/alumni networks)

This effort is being coordinated by Mike Gotta, Principal Analyst within Burton Group’s Collaboration & Content Strategies service. Information obtained during the interview process will be held in confidence. Any publication or presentation arising from this field research project will not mention any participating enterprise or individual by name without the express written permission of the organization.

Those enterprises involved in the project will have the option to take part in a peer review process prior to publication of the research (e.g., provide comments and factual corrections). Participating organizations will also receive a copy of the resulting research document and have the opportunity to discuss its findings with Burton Group.

If you have a question, need additional information, or are interesting in participating, please leave a comment on this blog post and I will get back to you.

Going Green with Web Conferencing

While doing research for my upcoming web conferencing overview document, I was discussing with the folks at the Adobe the perceived value of web conferencing in relation to being “green.”  With new announcements almost daily from companies announcing their “green” efforts (For example, IBM’s latest announcement in it’s Big Green initiative), being green in IT is growing in popularity and trendy. 

The folks at Adobe showed me a nifty little application pod for Adobe Connect Pro from RefinedData that calculates, based on the IP address of attendees, the carbon footprint savings by attending a web conference meeting instead of traveling to the host location.  The add-in is called Footprints.  The screenshot below gives you a feel for it:

The idea is that the Footprints add-in will show attendees and the organization how much carbon they have saved the planet through the use of the Connect Pro conferencing tool.  If you are a Connect Pro customer, the add-in might be a way to tout your “greenness” –  if that is worth the $295 license fee from RefinedData to you.  You can download a free trial version here.

Another web conferencing vendor, iLinc, is expending a fairly significant effort to tout it's iLinc Green Meter:

 

The gist of these tools is to provide some method to track and measure the benefit of web conferencing in relation to Green initiatives by organizations.  While the scientific accuracy of the real environmental benefit could probably be argued, the marketing value seems to be on the rise with the trendy and politially correct efforts to "show your greenness."

Where Did All The Collaborative Authoring Researchers Go?

Blogger: Craig Roth

My job as an industry analyst sometimes requires me to be a detective.  I'm in the midst of researching my upcoming document on "Content Authoring in the Enterprise 2.0 Age" and uncovered an interesting mystery. 

Most of my day-to-day interactions are with end user clients, with a smattering of vendor conversations thrown in.  But when researching a new topic I like to see what research is going on in academia, which is where I noticed an interesting phenomenon.  One of the trends in content creation I'll be writing about is "collaborative authoring".  This is the idea that more and more documents are being created as a collaboration between many authors, which introduces procedural and technical challenges.  My research uncovered quite a bit of academic work in this area, but the lists of papers I found all mysteriously stopped around 2000.  It's as if an academic meteorite hit the earth at the end of 2000, wiping out all the collaborative authoring researchers without a trace! 

Did humanity solve the collaborative authoring problem rendering further research unnecessary?  Or was a more nefarious hand at play?  I had some theories, but this was just too curious to ignore, so I contacted some of the academics who were involved in this space in the late '90s to find out what happened.  I'm happy to say they are still alive and well.

Dr. Sylvie Noël, an HCI research scientist for the government of Canada, fingered "free collaborative authoring tools such as Wikipedia" as a culprit.  And since quite a few commercial products offering collaboration started coming out after 2000, researchers weren't as interested.  Dr. Noël did point out that work continues under the rubric of "collaborative editing" (more encompassing than just authoring).  Regarding collaborative authoring, she still hopes for "a popular product that meets the large corporations' needs and is as simple to use as email."  Me too.

Dr. Michael Spring, Associate Professor of Information Science at the University of Pittsburgh, pointed out that while the research assumed people author content together, in reality there is generally one owner with others just commenting.  And getting information workers to be a bit more structured and maybe - gasp! - look beyond Microsoft Word is often futile.  Like Dr. Noël he points out a profusion of "good enough" tools like wikis and better reviewing features in Word.  Once theory starts showing up in real, commercial products like word processors and wikis the grants and academic interest dries up pretty quick.

Dr. Spring had another observation that carries over into my research on attention management and improving employee productivity.  After exploring the potential time and cost savings that technology could yield for distributed collaborative authoring for engineering standards, he wonders if "the senior engineers really didn't want to be that efficient."  They liked getting together in first-class global cities to hang out together rather than efficiently exchanging snippets of content using web-based collaboration.  In fact, these efficiencies could threaten the staff and budgets of their departments.

To me, it's unfortunate that this research has died down. Even if the theoretical level is now understood, it hasn't all been turned into practice and technology yet.  Large vendors like IBM and Microsoft do have research groups, but I haven't confirmed they have picked up the research now that academia has handed it off.  It's clear there is still more than enough room for some good ideas.

Conversation View: Lotus Notes 8

In response to my post about conversation view, Carl Tyler provided screenshots of Notes 8 and how it organizes messages in conversations.

Below is an opened conversation. Notice how the subject field in the replies contains the first line of the message body. This provides additional context to the e-mail conversation.

This next screenshot shows the mouse hovering over a message and and a pop-up displaying more of the first line. This looks like an effective way to scan an entire conversation.

These screenshots of Lotus Notes 8 are great examples of how conversation view can make a big difference in managing e-mail. Thanks Carl!