Scalability Redefined

Scalability matters. Obviously. But scalability seems to mean many things to different people. Most commonly, scalability is equated to the number of users - people - accessing a particular system. That's a good start, but we we have to of course consider the degree of concurrency. There is a difference between a solution that 1,000 people use once or twice a week versus a solution that 1,000 people pound upon continuously.

In large deployments, there is also the question of how many systems are actually really being used. Early in my career, I spent several years working at Novell. It was the heyday before Windows NT and the company enjoyed a commanding market share. Large and small companies used Novell networks. But we knew, that an average (software) server always had only about 100-200 users working on it. Sure, there were some companies with 100,000 employees that used Novell NetWare. Those were big deals, big deployments, and examples of high scalability. Right? Well, not really.

Those large companies were in reality running hundreds of separate Novell server instances and each one of them would still get just a couple hundred users. I see a similar pattern with today's deployments of Microsoft SharePoint and, to some extent, also with Exchange. It is quite a different kind of scalability when 100,000 users all share a single instance of an OpenText repository - even if that repository runs physically across multiple hardware servers.

There are more dimensions to scalability than just the number of users, though. The number of operations or transactions comes to mind - from MIPS to the number of credit card purchases. How about the number of objects under management? Examples could be the number of documents, number of customers, number of transactions, number of contracts, number of relationships, number of suppliers, etc.

To be scalable, enterprise software must not be just capable of holding the number of objects in a database or repository. It also has to provide the ability to efficiently view and manipulate the data. If your web-based user interface shows the objects in blocks of 20 while you have millions of objects to sort through, your application might not be very scalable! By the way, having a single data container capable of holding millions of objects is another dimension of scalability.

There is yet another factor that needs to be added to the definition of scalability: the metadata. This is the data that describes your data. Without metadata, your data only contains what is explicitly stated in it. Sometimes, that may be useful by itself but in most cases, we want to add lots of enterprise related context. We want to add information about people and teams who work with the data. Information about the organizational structure and approval hierarchies. Information about projects to which the data belongs. The deadlines, the cost centers, the retention requirements, etc. The metadata can be often richer (= bigger) than the data itself. But it is absolutely critical in the enterprise and your application needs to scale to accommodate it.

There are many factors that define scalability and looking just at the number of users can be often insufficient or even misleading.

The Social Groupthink

The groupthink is a well documented psychological phenomenon where a group of people, usually insulated from any counterbalancing point of view, ends up making gravely irrational decisions. During the phenomenon, the group is driven by the spirit of harmony, collaboration and the desire to reach consensus. The more the reasoning progresses, the more the group is mutually reinforcing its one-sided perspective while it completely discounts any alternate point of view. The results can be disastrous and many events in history have been attributed to the groupthink, including the US Navy negligence prior to the attack on Pearl Harbor or the US invasion of the Bay of Pigs.

The John F. Kennedy cabinet during the Cuban Missile Crises - another "groupthink" at work

In the world of social media, groupthink is very common. On social media, we tend to follow people who share views that are consistent with our own views. Hockey fans follow other hockey fans, single mothers follow single mothers, wine connoisseurs follow wine connoisseurs, and democrats follow democrats. This natural selection is the result of a rational behavior - we engage with people with whom we share common interests.

On top of that, social media like Facebook employ filtering algorithms to reduce the torrent of updates we get exposed to. These filters are based on explicit personal preferences (i.e. interests stated in our profiles) as well as on the results of our interactions. If you like a post about kittens the algorithm will reason that you like kittens and chances are you will be seeing more posts about kittens. Over time, you end up ‘liking’ various comments, pictures, and pages. Based on your liking, Facebook starts presenting you more of the stuff you like from the people who’ve shared liked news before. That happens at the cost of all other news in your newsfeed. As a result, you get exposed only to views from friends who you “like" more and more and you won’t get exposed to anything else. This is a fertile breeding ground for a groupthink with all its shortcomings.

Now, consider social software in the enterprise. Its promise was to stimulate employee effectiveness and foster innovation by bringing together diverse groups of employees who bring in different expertise and who share different points of view. Yet, if we end up with conversations where only the employees thinking the same way talk to each other, the results of social software will be greatly reduced.  

There are, of course, many other great uses for social software in the enterprise i.e. collective decision making, process collaboration, or customer service. Yet with the need to drive the corporate innovation agenda on top of the priority list for many CEOs, the promise to use social software for ideation is very compelling.

To make that happen, we must avoid the social groupthink. We have to be very careful about the filtering algorithms we employ and we have to devise strategies that encourage employees to engage with others beyond their existing teams and functions. Tribal interactions are good, but engaging across a variety of employees is what stimulates corporate innovation.

Gesture Control in the Enterprise and the Consumerization Chasm

When Microsoft first shipped Kinect as an add-on for the XBox 360, I thought: “Wow, there is a new way to interact with information!” Sure, Kinect was designed for ‘full body gaming’ as Microsoft calls it but the ability to use gestures to find, access and view information seemed very promising. Ever since the 2002 hit movie Minority Report, we are yearning to work with information the way the Tom Cruise character did: using gestures.

The original - Steven Spielberg's Minority Report 

The use cases in the consumer space are primarily focused on gaming and the interaction with entertainment media. Using iTunes on AppleTV or Netflix on Xbox is great but, let’s face it, searching for movies using a remote control with no keyboard is a pain. Gestures could help with browsing the content while voice recognition could solve the typing problem.

Microsoft Kinect

The use cases in the enterprise, though, are far more promising. Just think about the surgeon with sterile hands who needs to flip through a series of X-rays, zoom in, start and pause a video recording from a echocardiograph, and quickly query a drug database. Think about the aircraft mechanic with oily hands who needs to access a repair manual for the latest model of a jet engine. How about the teachers explaining the latest material in front of a class of students? Or the speaker on stage using his hands instead of a geeky laser pointer...or instead of a fork lift like Al Gore did in The Inconvenient Truth? There are many possible professional uses for the gesture technology!

Yet, how come I don’t see any of this in the real life? Maybe Kinect isn’t good enough? Maybe it is sold only through the same stores that sell the gaming consoles and ignore the enterprise? Does Microsoft Marketing perhaps need help? There is a Kinect for the Windows web site promoting a software development kit (SDK) but there are no business examples featured on that site.

Google Glass, those hip looking glasses with a built-in computer screen (and a computer) have a similar potential in the enterprise. There are many professions that would greatly benefit from this kind of “always on display”. However, Google’s primary concern right now is making sure that a lot of celebrities get their picture taken with the Glass on their nose. They don’t even talk about business use cases. I worry now that Google will spend all its energy on devising schemes on how to push ads to people while they walk down the mall. Sure, we have seen that too in Minority Report but, honestly, that part of the movie sucked.

Google co-founder Sergey Brin wearing Google Glass

Microsoft Kinect, Google Glass, and other interactive devices such as the MYO wrist device or the Leap Motion Controller, combined with the Siri-like voice recognition are the future of computing. Touchscreen has its limitations. People have only so much tolerance for the small screen size of a smartphone - which is why the so-called phablets have become so popular. The interaction with a computer of the future will likely not involve fingers on glass but rather gestures, voice and perhaps even thoughts.

MYO is a gesture control armband

While using such interactive devices to browse movies is cool, using them in the enterprise can result in some really powerful benefits. Unfortunately, the leading vendors such as Google, Apple, and Microsoft are all chasing the consumers right now. Consumerization is hitting the enterprise but the vendors only think about the consumers and not about the enterprise. The innovation in enterprise computing is stagnating today and there is a chasm. And where there is a chasm, new opportunities open up for new entrants...

How Come Fax Isn't Dead?

When was the last time you sent a fax? I bet your answer is probably going to suggest that you don’t use fax machines much anymore. Yet the statistics are surprising. According to the industry analysts Davidson Consulting, there are almost 100 billion faxes sent each year worldwide. CouponChilli estimates a smaller number - 17 billion faxes annually - but either way, it is still a lot. The market for fax services is growing at impressive 15.2% CAGR as new technologies such as Fax over IP and cloud-based fax are dramatically reducing the the cost of faxing.

But let’s face it - fax? In the age of email, interactive web sites, and omnipresent mobile devices? There are so many alternatives - from email, FTP, managed file transfer, to interactive web sites, sophisticated BPM solutions and most recently even the easy-to-use cloud-based shared folders. How come we are still using faxes so much?

One of the arguments is usually the legality of the “wet signature”. Supposedly, our legal system is perfectly satisfied with an illegible scribble transmitted at 204×98 dpi but an electronic or digital signature on an electronic document is not good enough. I’d argue that’s a bogus argument - after all, a digital signature can use a much stronger authentication of the signatory which should make the digital signature much more legally binding. Only when notarized, do wet signatures come close in terms of security and legal admissibility.

Another argument is the confirmation that you get when you send a fax. That fax confirmation page (sometimes called the Transmission Verification Report) can act as legal proof that the recipient actually received the fax. That might come handy, for instance, when you fax an invoice. However, it also assumes that the right person has actually picked up the received fax and that it didn’t end up in the waste bin by accident.

This argument is also not very convincing. Electronic transmissions via email, ftp, managed file transfer, or shared folder usually all come with an audit trail. There is also no reason why an email system could not be setup to automatically send a confirmation - most e-commerce and customer service solutions do it today. A secure audit trail should be a much better proof of delivery than the easily falsifiable fax transmission report.

Then there is the cultural argument. We are all used to faxing, right? The New York Times reported recently that this is a big issue in Japan. Japan has by far the highest number of fax machines per capita. But give me a break. Japan is a country with thousands of years of tradition - from Zen gardens to calligraphy to the sword swinging samurai. And all the sudden,  they are culturally attached to a fax? Or a technology that was adopted at the end of the 1980s? Ah, come on!

Maybe, it is the ease of use - everybody knows how to use a fax, right? Right. To sign a document, I have to print it, sign it by hand, and stand next to a fax machine to wait to see if the transmission was successful. If the line is busy, I have to try again later. Yeah, right...that’s really easy and convenient...

I suspect that the reason is complacency. Complacency of organizations in banking, insurance, healthcare, government and other sectors. The Customer Service group would probably like to replace faxes but they are scared of Legal. The Legal group is scared of technology. And, IT does what Customer Services asks them to do after checking with Legal. In a way, this is actually about culture. But not in a good way.

Picture from the 1999 cult movie Office Space. I'm sure it's copyrighted but it fits so well here.
You have to see it, by the way! (That should get me off the hook with 20th Century Fox).

The result is that you can hardly open a bank account, refinance a mortgage, buy a car, or submit an insurance claim without having to send a fax. In some countries like Canada and Germany, it gets even harder as they require signed documents for stock trades, change of  address, and other relatively mundane tasks. No you can’t just call them. They need it in writing and signed...

But don’t despair, if you need a fax solution, you don’t need a fax machine on every floor anymore. There are some pretty cool network and cloud-based fax systems out there and OpenText (my employer) happens to be the market leader in this space. Just check out that Davidson Consulting page.

As consumer, though, I hope that the fax will soon die. It ought to be replaced by the interactive web or mobile BPM solutions. In any case, OpenText has a solution for you here ;-)

Maybe, There Are No Systems of Engagement

A couple of years ago, I took part in an AIIM project that was examining the recent changes impacting enterprise software systems - consumarization and mobile, social, and cloud based software. Spearheaded by none other than Geoffrey Moore (yes, the one of Crossing the Chasm fame), we ended up defining the terms systems of record and systems of engagement. You can find the AIIM paper by Geoffrey Moore here.

Geoffrey Moore

Since then, the terminology of systems of record and systems of engagement entered the mainstream nomenclature and it is used regularly in various discussions. The systems of record are the traditional (yes, old-fashioned and boring) applications that enterprises have been relying on to manage their critical information. The systems of engagement are those hip, new applications (with, ehm, sometimes questionable ROI) that were supposedly the future of enterprise software.

But the more I think about it, the more I’m convinced that we got it wrong back then. When I look at the current landscape, the traditional enterprise applications have all taken on the aspects of systems of engagement. Pretty much every application has a mobile story. There is no mobile enterprise software market. All decent enterprise software has mobile capabilities today. Most enterprise software vendors have added social capabilities to their software. Many of them have launched their cloud initiatives. In fact, I am not sure that there are many viable systems of engagement left out there. Even the vendors who started in the systems of engagement world are rushing to add some of those boring system-of-record features like a repository, security, and governance in an attempt to look more like true enterprise software.

Take the Customer Experience Management market as an example. Those were supposed to be the ultimate systems of engagement - the web based applications engaging with the company’s customers, partners, and employees. But to a Chief Marketing Officer (CMO), these applications are not just about engagement. They are about leads, opportunities, and deals. The Marketing department is being measured by the strength of the pipeline and the opportunity-to-deal conversion ratio. To the CMO, these systems are his systems of record as much as systems of engagement.

In reality, there are no two separate worlds. No systems of record and systems of engagement. What used to be referred to as systems of engagement are a new set of capabilities that have, greatly improved the traditional enterprise applications. When done right, they can significantly augment the usability and adoption rates of enterprise software. But they are not a separate market. They are features.

There is only one type of enterprise software - systems that manage enterprise information.

The Maslow's Hierarchy of a Strategic CIO

Not that long ago, IT departments were responsible mainly for making sure that knowledge workers got access to whatever information was available and for keeping the lights on. That latter responsibility was all consuming. PCs and enterprise systems were still going through their growing pains and varieties of system failures were all too common. The job of the chief information officer (CIO) was more about troubleshooting and firefighting than anything else. Information Technology (IT) was simply a cost center - just like travel or communication.

Well, things have changed. The systems we work with are, for the most part, reasonably reliable. We don't have to reboot our PCs twice a day just to flush out the memory leaks. 99.999% uptime is not that big of a differentiator for servers, routers, and switches anymore. Users don't call the help desk daily and access to information is not the challenge at hand. Today, we have information. We have a lot of information. In fact, we have way more information than we could ever consume.

Herein comes the change in the mission of IT departments. As organizations came to realize that information emerged as a key source of competitive advantage, they were increasingly looking at their IT departments as a key stakeholder in corporate strategy. All of the sudden, the CIOs got what they were always dreaming about. No longer the troubleshooter, no more the firefighter - the CIO is now the strategist.

What organizations need is the ability to make better decisions - using the right information. They need insight. They also need to apply information to create an impact on their business - to grow revenue, attract new customers, enter new markets, and generate innovation. And finally, they also need to drive productivity and continuously optimize their business processes.

Not to forget, enterprise information must also be secured. As it represents significant intellectual property, it has to be protected from intentional or unintentional misappropriation by internal or external actors. And let us not forget the need to address compliance and information governance requirements and to protect the company from legal exposure.

All these requirements reminded me of the Maslow's Hierarchy of Needs and so I have attempted to map the CIO needs into a similar model:

With all of this, the strategic CIOs have their hands full. They still need to keep the lights on but now, they are major stakeholders in defining corporate strategy. Combine that with all the new technology trends such as mobile devices, social software, and cloud computing and you get the picture of the magnitude of their challenge. But I guess that being strategic is way better than fighting fires all day long, right?

ECM - Fish or Fowl?

Enterprise software companies usually fall into one of two main categories: application software and infrastructure software. The common wisdom suggests that enterprise software vendors are either building infrastructure that enables multiple applications to operate or they are delivering applications that solve actual business problems.

ERP is a typical application software. The ERP applications solve specific business problems such as how to optimize the use of enterprise resources, how to accelerate HR processes, or how to reduce the amount of materials in stock without slowing down production.

Platforms such as operating systems, databases, and application servers are infrastructure software. Communication software ranging from EDI to messaging, file transfer, and email is infrastructure software. Virtualization software and administration/management software are infrastructure. Infrastructure does not solve any particular business problem - it can be applied to solve any number of them.

That begs the question, what is enterprise content management (ECM). As it turns out, ECM is rather unusual. Unlike almost any other enterprise software category, ECM can be deployed as infrastructure or as applications. Many customers I have seen, have deployed their ECM as a platform with multiple applications built upon it. They clearly consider ECM part of their infrastructure. Many customers have also deployed their ECM at the heart of their information sharing, archiving, or retention infrastructure. Gartner assigned ECM to what they used to call the Knowledge Worker Infrastructure category. There is no doubt about it - ECM is infrastructure software!

Yet, there are many deployments of ECM software devoted to applications - solutions for very specific business problems. Examples include contracts management, early case assessment, invoice processing, customer onboarding, plant asset management, digital marketing, and technical publishing - to name just one example for each of the seven types of content applications that I have introduced in my recent blog post. These deployments are not infrastructure, they are applications. Clearly, ECM is application software!

That makes enterprise content management quite unique - it can be deployed as an infrastructure and it can be used as an application (or multiple applications). There is hardly any other enterprise software category that comes with this level of complexity. The only other category I could think off is business process management (BPM) - a software category that shares a lot of synergies with ECM.

This uniqueness may explain some of the identity challenges that ECM has been having over the last two decades. Who owns ECM? Who’s in charge of the ECM architecture in the enterprise? The answers have never been very clean cut. Depending on the organization, the ‘owner’ may be in IT or in any number of corporate functions.

Most enterprise software vendors fall clearly on one side of the dividing line - they are either application vendors or infrastructure vendors. They may attempt to reach across that line, but their DNA is usually pretty hard-wired. SAP is an application vendor. VMware is an infrastructure vendor. IBM doesn’t do applications. Salesforce may claim to be a platform but it is really an application with the ability to be extended by add-ons - nobody would refer to their SFDC deployment as infrastructure. There are hardly any mixed vendors outside of ECM.

OK, people usually mention Oracle as a company that provides both - database software (infrastructure) and application software (ERP, CRM, etc.). That is true. However, Oracle has gotten to this point via a very aggressive acquisition strategy and its database, middleware, and application businesses are operating independently from each other. They are effectively several enterprise software companies under a single brand.

I believe that we have to accept the fact that ECM may mean a completely different thing to different people. It doesn’t make it any easier to describe (or sell) but it also makes it a much more exciting market. The versatility of ECM is something that both the vendors and customers often appreciate.

Because, ECM is unique!

Seven Types of Content Applications

The enterprise content management (ECM) industry has been talking about content applications for many years. A while back, Gartner coined the term “composite content applications” while Forrester talks about “content-centric applications”. What I mean are applications that primarily deal with unstructured data (content) rather than structured data applications, such as enterprise resource planning (ERP), customer relationship management (CRM), supply chain management (SCM), and product lifecycle management (PLM).

We all can usually come up with many examples of such applications but, to my knowledge, they have never been properly classified. What are the types of content applications out there? Sure, Forrester introduced new categories for ECM in their groundbreaking report Transactional, Business, and Persuasive Content: A Better Way to Look at Enterprise Content back in 2005. But that was really dealing with the different technologies rather than the application types. So, I decided to give it a shot myself. I would love to get your help with it, though. Please do comment if you agree or disagree and if you can think of applications that don’t fit into my categories.

Before I start, I should make it clear that while all the application types below use content as the primary data type, they go beyond ECM. They also involve business process management (BPM), customer experience management (CEM), and discovery. At the same time, I am not trying to cover all BPM or CEM applications, but rather only those that use content. For instance, I am not including straight-through processing (STP) applications in BPM such as payment transactions or capital markets transactions because those don’t use content. Basically, I am categorizing applications that span the enterprise information management (EIM) space, as we define it at OpenText. So, here are the seven types of content applications:

1. Productivity Applications
I’ve labeled the first group ‘productivity applications’ because they are all aimed at increasing employee productivity, which is sometimes very difficult to measure. These applications usually involve sharing business documents, sifting through vast volumes of information, collaborative authoring, document libraries, and collaboration/social software as a means of improving employee communication and effectiveness.

Examples of such applications include corporate policy libraries, knowledge management, contracts management, idea management, etc. These applications are often considered part of the knowledge worker infrastructure as they require relatively little customization and they are typically not department specific or industry specific. IT usually selects and owns these applications.

2. Compliance Applications
Compliance applications are the bread and butter of the ECM industry. They are addressing the requirements for regulatory compliance and corporate information governance, and they are reducing the legal risks related to content used as possible evidence in a court of law. These applications focus on access control, auditing, information retention, and mandated tasks, approvals, and sign-offs.

Examples of such applications include records management, legal discovery (eDiscovery) and early case assessment, as well as many regulated document applications used to address specific requirements such as the Sarbanes-Oxley Act as well as self-imposed requirements such as Six-Sigma or ISO 14001. Such applications are almost always function or industry specific, i.e. applications dealing with the FDA 21 CFR Part 11 regulations in life sciences, the OSHA material safety data sheets in chemical process manufacturing, or the Dodd-Frank Act in financial services. A wide variety of functions can be selecting and owning such applications, although the heavily regulated industries often have a Chief Compliance Officer while companies in highly litigious industries lean strongly on the Chief Legal Counsel here.

3. Process Applications
There is a group of applications that are very process-oriented, and yet they depend heavily on content as the information used for decisions that determine the process routing, tasks, and results. Such applications usually involve electronic forms and the capture of incoming paper documents. They also take advantage of process modeling and analysis, process simulation and optimization, rules engines, and process engines as well as process reporting and analytics. Frequently, the process applications integrate with other applications such as ERP and CRM.

Examples include invoice processing (a.k.a. accounts payable), travel expense management, and many vertical applications such as engineering change orders, dispute resolution, and authorization for expenditure. Usually, the process applications have an easily measurable ROI. These solutions are typically selected and owned by their respective functions and can span both the core and the supporting functions of the organization.

4. Case Management Applications
Case management came about as an use case of BPM but it deals with quite different types of applications. Gartner believes that case management is just a use case of BPM while Forrester declared case management a separate market - and a very fast growing one too. The case management applications are different from traditional BPM applications as they don’t just use content as a payload - they are much more about the content. They typically involve a case file which is a smart repository container accommodating many content assets and the logic governing their use. Besides a stronger dependence on a content management repository, case management applications can include many of the process application technologies for all the big and small processes required to manage the case file.  

Examples of case management applications include customer onboarding, employee file management, and fraud investigation. Vertical case management applications include insurance claims processing, loan origination, legal case management, and patient care management. The selection and ownership of case management applications falls - just like the process applications - onto respective corporate functions.

5. Resource Management Applications
Resource management applications are, as my label suggests, managing various resources - from human resources to customers and suppliers, from products to plant assets. These applications are frequently used in tight integration with structured data applications such as ERP, PLM, or CRM. The main purpose of these applications is to systematically organize large volumes of content assets that often need to be retrieved very quickly based on a complex set of metadata - i.e., all material with a warranty expiring in the next 30 days found in a specific geographic location. The resource management applications need to accommodate a rich variety of content formats: from documents and images, to CAD drawings and digital X-rays.

Examples of resource management applications include customer information management, product information management, plant asset management as well as vertical applications such as patient records, and legal matter management. The resource management applications are selected and owned by the responsible function in the organization (line of business).

6. Go-to-Market Applications
As the name suggests, these sales and marketing applications are used to support the organization’s go-to-market efforts. Their main job is to capture attention, persuasively convey a message and solicit a desired call to action. They typically involve rich media assets, multi-channel delivery of content, social engagement and the need to measure and monitor their effectiveness. The sales and marketing applications also need to account for geographic differences - from language, local examples, and local trends, to different customs, ways of conducting business, and customer privacy laws.

Examples include digital marketing, e-commerce, marketing libraries, marketing campaign management, sales collateral libraries, and virtual pitch books. While there are vertical flavors to such applications (i.e., the pharma companies have to add some compliance capabilities to their digital marketing), the go-to-market applications are applicable across industries. They are almost always purchased and owned by the sales and marketing departments.

7. Publishing Applications
I use the label ‘publishing applications’ for all types of applications where content is the actual product or a product component. The published product can come in many formats and increasingly multiple formats are being combined into a single final product. For example, books are now typically published on paper (hard and soft cover), for consumption in different e-readers (at least three formats are necessary: Kindle, iBooks, PDF), and as an audiobook. Increasingly, the content delivery needs to take in account the consumption device with its screen size, resolution, and bandwidth. Although the products are becoming more consistent worldwide, translation, localization and distribution rights are a major factor here.

Examples of such applications include technical publishing, catalog publishing, different types of media publishing (i.e., book publishing, magazine publishing, motion picture and video production), radio and television programming, and learning material publishing. Even non-corporate blogs like the ones published on WordPress, Blogger, or Tumblr fall into this category, although many of them are used in the consumer domain outside the scope of ECM. The buyers of such solutions are almost always the heads of the publishing production who often carry different titles depending on the industry.

So, that’s it. These are the seven types of content applications. Before anyone brings it up, I do realize that many of these applications started reaching across the categories. In regulated industries, pretty much every application has to include compliance. You can make an argument that BPM should be included in everything. Same for collaboration or social software. But what decides the categorization is the original goal for deploying the application. Are we deploying the application to handle more customer requests and compliance just happens to be a required feature? Then it is primarily a process application rather than a compliance application. The primary stakeholder is usually the telltale too. Different owners have a different purpose for their application, which usually determines the type of solution they will select.

While this is my view of the landscape, I would love to hear from you. Do you agree with my categories? Do you see any other categories or a different way of categorization? Have you encountered any applications that don’t fit? Please share your comments and help me make it better. If there are substantial changes as a result of your comments, I will publish an updated version of this post.

Thank you!

Is Dropbox Making us Stupid?

Yes, I’m paraphrasing the famous article by Nicholas Carr with the title of this blog post. But I do have a point to make. For years, the document management and content management solutions suffered a bit from the stigma of being too complex and difficult to use. Then, the nouveau bread of solutions stormed in, by propelled by the power of consumerization: cloud-based file sharing with Dropbox leading the charge. Suddenly, all the problems were solved...

Well, not quite. The consumer-grade software seduced users with its simplicity. It usually did just one thing but it did it well and easily. Before the CIO could blink twice, the Dropbox or some similar software was all over the enterprise. Of course, we all have heard the stories about the lack of security, the compliance issues, and the legal exposure this caused. But that’s not my point today.

My point is that users started using this software to do actual work. Work, by definition, is a more serious matter than sharing Christmas pictures with your relatives. When one knowledge worker needs to share a document with another one, Dropbox or a similar file sharing system might be a great way to do it. Probably better than email. But when a team of 10 or more people start using file sharing to jointly work on a set of documents, it may result in chaos.

The workers start mixing up document versions. The wrong people get to see the documents when they shouldn’t. People attempt to make edits at the same time only to overwrite each other’s changes. A manager critiques the document before the subject-matter expert got to review it. And, nobody can find anything... All that because they use file sharing when what they really need is a solution with versioning, access control, library services, workflow, and search. In short, they need a team collaboration or document management solution - not file sharing!

There is nothing wrong with file sharing. It is great. It saves a lot of time, hassle and money. But we need to continue educating ourselves about the right tools for the right job. Some activities can be done easily using file sharing. In the enterprise, it should be an enterprise-class file sharing solution with security, compliance, and under the control of the company. But again, that’s not my point today. My point is that to share documents, file sharing is great. To collaborate on documents, you need team collaboration software.

You may be able to loosen a screw with a pocket knife but, let’s face it, a screw driver is a far better tool for the job.

Sometimes, Common Sense Beats ROI

Return on investment (ROI) is supposed to be the magic that makes any software sell instantly - the ultimate silver bullet for any sales person. You simply offer a logical, monetary justification of the solution’s value and the customer has no alternative but to buy. Right? Yeah, right...

The reality is that ROI can be a very elusive value proposition. For one, it rarely considers any competition. You have a problem and you can solve it by reducing your cost by x% for the price of $y which represents the initial and ongoing fees for the solution. Well, chances are that the competition has a similar solution and - what an audacity - they claim the same cost savings! It will be rare that your solution has a quantifiable capability that results in a unique and competitively differentiated ROI.

More often than not, your competitive advantage lies on a different level - product architecture, user experience, quality of service, or...ehm...price. Sure you can argue that each one of those characteristics too has an ROI but now you start overlaying one ROI calculation of a solution characteristic on top of business value ROI which is probably going to be too complex and less and less credible.

Let me give you an example. Say that the solution in question has to do with accounts payable where we know the cost for processing each invoice and a proposed solution results in a 15% cost reduction per invoice. Every CFO usually knows the number of invoices processed each month and the resulting savings can be compared with the cost of the solution and voila - we have an ROI! Pretty easy, right? But of course, every competitor can claim similar savings, assuming the cost of the solution is comparable.

We could in theory add a set of ROI calculations that estimate the savings of a particular user experience compared to the competition, or the savings resulting from a better product architecture but let’s face it, this will be a tough one to make credible.

That brings me to the main issue with ROI. Some solutions - like my accounts payable example - have a very measurable, credible ROI. Others don’t. Usually, anything to do with knowledge worker productivity or customer experience is difficult to measure. Often, it is impossible - or impossible to make credible.

Just take knowledge worker productivity (or effectiveness) which is a frequently quoted value proposition. What’s the ROI of the telephone on your desk? Or the wi-fi network in your office? How about the ROI of email? It’s hard to argue that these tools don’t contribute to the knowledge worker productivity but the benefit is so obvious that we just don’t bother calculating it. Sure, we can calculate the ROI moving from traditional PBX phone system to Voice-over-IP (VoIP) phones. But we don’t need any formula to justify purchasing phones for every employee.

Any calculation that deals with knowledge worker’s time savings is dubious. Sure, a transactional type of work is really measured in hard numbers but most of the knowledge workers are switching between tasks, going to meetings, spending time brainstorming, thinking, creating, and communicating with others. It is not credible to argue that because a particular solution saved 30 minutes a day, it results is a measurable productivity increase.  

The bottom line is that ROI is not a magic bullet. It may very much help to establish the need or even justify the purchase for certain types of solutions. But there are many solutions purchased not because of an ROI: communication, collaboration and social software, knowledge management, document sharing, office and personal productivity applications - all such offerings are usually purchased because they just make sense rather than as a result of a hard ROI.

Don't sweat the ROI too much. Because sometimes, common sense beats ROI.