In my time in this business (enterprise content management, records management, document management, etc.) I have been asked to help with a ‘shared drive problem’ more times than I can remember. This particular issue is analogous with the paperless office problem. Thirty years ago when I started my company I naively thought that both problems would be long gone by now but they are not.

I still get requests for purely physical records management solutions and I still get requests to assist customers in sorting out their shared drives problems.

The tools and procedures to solve both problems have been around for a long time but for whatever reason (I suspect lack of management focus) the problems still persist and could be described as systemic across most industry segments.

Yes, I know that you can implement an electronic document and records management system (we have one called RecFind 6) and take away the need for shared drives and physical records management systems completely but most organizations don’t and most organizations still struggle with shared drives and physical records. This post addresses the reality.

Unfortunately, the most important ingredient in any solution is ‘ownership’ and that is as hard to find as it ever was. Someone with authority, or someone who is prepared to assume authority, needs to take ownership of the problem in a benevolent dictator way and just steam-roll a solution through the enterprise. It isn’t solvable by committees and it requires a committed, driven person to make it happen. These kind of people are in short supply so if you don’t have one, bring one in.

In a nutshell there are three basic problems apart from ownership of the problem.

1.     How to delete all redundant information;

2.     How to structure the ‘new’ shared drives; and

3.     How to make the new system work to most people’s satisfaction.

Deleting redundant Information

Rule number one is don’t ever ask staff to delete the information they regard as redundant. It will never happen. Instead, tell staff that you will delete all documents in your shared drives with a created or last updated date greater than a nominated date (say one-year into the past) unless they tell you specifically which ‘older’ documents they need to retain. Just saying “all of them” is not an acceptable response. Give staff advance notice of a month and then delete everything that has not been nominated as important enough to retain.  Of course, take a backup of everything before you delete, just in case. This is tough love, not stupidity.

Structuring the new shared drives

If your records manager insists on using your already overly complex, hierarchical corporate classification scheme or taxonomy as the model for the new shared drive structure politely ask them to look for another job. Do you want this to work or not?

Records managers and archivists and librarians (and scientists) understand and love complex classification systems. However, end users don’t understand them, don’t like them and won’t use them. End users have no wish to become part-time records managers, they have their own work to do thank you.

By all means make the new structure a subset of the classification system, major headings only and no more than two levels if possible. If it takes longer than a few seconds to decide where to save something or to find something then it is too complex. If three people save the same document in three different places then it is too complex. If a senior manager can’t find something instantly then it is too complex. The staff aren’t to blame, you are.

I have written about this issue previously and you can reference a white paper at this link, “Do you really need a Taxonomy?”

The shared drives aren’t where we classify documents, it is where we make it as easy and as fast as possible to save, retrieve and work on documents; no more, no less. Proper classification (if I can use that term) happens later when you use intelligent software to automatically capture, analyse and store documents in your document management system.

Please note, shared drives are not a document management system and a document management system should never just be a copy of your shared drives. They have different jobs to do.

Making the new system work

Let’s fall back on one of the oldest acronyms in business, KISS, “Keep It Simple Stupid!” Simple is good and elegant, complex is bad and unfathomable.

Testing is a good example of where the KISS principle must be applied. Asking all staff to participate in the testing process may be diplomatic but it is also suicidal. You need to select your testers. You need to pick a small number of smart people from all levels of your organization. Don’t ask for volunteers, you will get the wrong people applying. Do you want participants who are committed to the system working, or those who are committed to it failing? Do you want this to succeed or not?

If I am pressed for time I use what I call the straight-line-method. Imagine all staff in a straight line from the most junior to the most senior. Select from both ends, the most junior and the most senior. Chances are that if the system works for this subset that it will also work for all the staff in between.

Make it clear to all that the shared drives are not your document management system. The shared drives are there for ease of access and to work on documents. The document management system has business rules to ensure that you have inviolate copies of important documents plus all relevant contextual information. The document management system is where you apply business rules and workflow. The document management system is all about business process management and compliance. The shared drives and the document management system are related and integrated but they have different jobs to do.

We have shared drives so staff don’t work on documents on ‘private’ drives, inaccessible and invisible to others. We provide a shared drive resource so staff can collaborate and share information and easily work on documents. We have shared drives so that when someone leaves we still have all their documents and work-in-process.

Please do all the complex processes required in your document management system using intelligent software, automate as much as possible. Productivity gains come about when you take work off staff, not when you load them up with more work. Give your staff as much time as possible so they can use their expertise to do the core job they were hired for.

If you don’t force extra work on your staff and if you make it as easy and as fast as possible to use the shared drives then your system will work. Do the opposite and I guarantee it will not work.

Document imaging (scanning) has been a part of most business processing since the early 1980s. We for example, produced our first document imaging enabled version of RecFind in 1987. So it isn’t new technology and it is now low risk, tried and proven technology.

Even in this age of electronic documents most of us still receive and have to read, analyse and process mountains of paper.

I don’t know of any organization that doesn’t use some form of document imaging to help process paper documents. Conversely, I know of very few organizations that take full advantage of document imaging to gain maximum value from document imaging.

For example, just scanning a document as a TIFF file and then storing it on a hard drive somewhere is almost a waste of time. Sure, you can then get rid of the original paper (but most don’t) but you have added very little value to your business.

Similarly, capturing a paper document without contextual information (Metadata) is not smart because you have the document but none of the important transactional information. Even converting a TIFF document to a PDF isn’t smart unless you first OCR (Optical Character Recognition) it to release the important text ‘hidden’ in the TIFF file.

I would go even further and say that if you are not taking the opportunity to ‘read’ and ‘capture’ key information from the scanned document during the scanning process (Forms Processing) then you aren’t adding anywhere near as much value as you could.

And finally, if you aren’t automatically initiating workflow as the document is stored in your database then you are criminally missing an opportunity to automate and speed up your internal business processes.

To give it a rating scale, just scanning and storing TIFF files is a 2 out of 10. If this is your score you should be ashamed to be taking a pay packet. If you are scanning, capturing contextual data, OCRing, Forms Processing, storing as a text-searchable PDF and initiating workflow then you get a 10 out of 10 and you should be asking your boss for a substantial raise and a promotion.

How do you rate on a scale of 0 to 10? How satisfied is your boss with your work? Are you in line for a raise and a promotion?

Back in the 1980s the technology was high-risk, expensive and proprietary and few organizations could afford the substantial investment required to scan and process information with workflow.

Today the technology is low cost and ubiquitous. There is no excuse for not taking full advantage of document imaging functionality.

So, where do you start?

As always, you should begin with a paper-flow analysis. Someone needs to do an inventory of all the paper you receive and produce and then document the business processes it becomes part of.

For every piece of paper you produce you should be asking “why?” Why are you producing paper when you could be producing an electronic document or an electronic form?

In addition, why are you producing multiple copies? Why are you filing multiple copies? What do your staff actually do with the paper? What happens to the paper when it has been processed? Why is it sitting in boxes in expensive off-site storage? Why are you paying to rent space for that paper month after month after month? Is there anything stored there that could cause you pain in any future legal action?

And most importantly, what paper can you dispose of?

For the paper you receive you need to work out what is essential and what can be discarded. You should also talk to your customers, partners and suppliers and investigate if paper can be replaced by electronic documents or electronic forms. Weed out the non-essential and replace whatever you can with electronic documents and electronic forms. For example, provide your customers, partners and suppliers with Adobe electronic forms to complete, sign and return or provide electronic forms on your website for them to complete and submit.

Paper is the enemy, don’t let it win!

Once you have culled all the paper you can, you then need to work out how to process the remaining paper in the most efficient and effective manner possible and that always ends up as a Business Process Management (BPM) exercise. The objectives are speed, accuracy, productivity and automation.

Don’t do anything manually if you can possibly automate it. This isn’t 30 years ago when staff were relatively cheap and computers were very expensive. This is now when staff are very expensive and computers are very cheap (or should I say low-cost?).

If you have to process paper the only time it should be handled is when it is taken from the envelope and fed into a document scanner. After that, everything should be automated and electronic. Yes, your records management department will dutifully want to file paper in file folders and archive boxes but even that may not be necessary.  Don’t accept the mystical term ‘compliance’ as a reason for storing paper until you really do understand the compliance legislation that applies to your business. In most cases, electronic copies, given certain safeguards, are acceptable.

I am willing to bet that your records manager will be operating off a retention schedule that is old, out-of-date, modified from another schedule, copied, modified again and ‘made-to-fit’ your needs. It won’t be his/her fault because I can probably guarantee that no budget was allocated to update the retention schedule on an ongoing basis. I am also willing to bet that no one has a copy of all of the current compliance rules that apply to your business.

In my experience, ninety-percent plus of the retention schedules in use are old, out-of-date and inappropriate for the business processes they are being applied to. Most are also way too complicated and crying out for simplification. Bad retention schedules (and bad retention practices – are you really destroying everything as soon as you are allowed?) are the main reason you are wasting thousands or millions of dollars a year on redundant offsite storage.

Do your research and save a fortune! Yes, records are very important and do deserve your attention because if they don’t get your attention you will waste a lot of money and sooner or later you will be penalised for holding information you could have legally destroyed a long time ago. A good records practice is an essential part of any corporate risk management regime. Ignore this advice at your peril.

Obviously, processing records efficiently requires software. You need a software package that can:

1. Scan, OCR and Forms Process paper documents.
2. Capture and store scanned images and associated Metadata plus any other kind of electronic document.
3. Define and execute workflow.
4. Provide search and inquiry capabilities
5. Provide reporting capabilities.
6. Audit all transactions.

The above is obviously a ‘short-list’ of the functionality required but you get the idea. There must be at least several hundred proven software packages in the world that have the functionality required. Look under the categories of:

1. Enterprise Content Management (ECM, ECMS)
2. Records Management (RM, RMS)
3. Records and Document Management
4. Document Management (DM, DMS)
5. Electronic Document and Records Management (EDRMS)
6. Business Process Management (BPM)

You need to define your business processing requirements beginning with the paper flow analysis mentioned earlier. Then convert your business processing requirements into workflows in your software package. Design any electronic forms required and where possible, re-design input paper forms to facilitate forms processing. Draw up procedures, train your staff and then test and go live.

The above paragraph is obviously a little short on detail but I am not writing a “how-to” textbook, just a simple guide. If you don’t have the necessary expertise then hire a suitably qualified and experienced consultant (someone who has done it before many times) and get productive.

Or, you can just put it off again and hope that you don’t get caught.

Many records managers and librarians still hold on to the old paradigm that says if a user wants something they should come though the information management professional. They either believe that the end user can’t be trusted to locate the information or that the task is so complex that only an information professional can do it in a proper and professional manner.

This approach to tightly controlled access to information has been with us for a very long time; unfortunately, not always to the benefit of the end user. It is often interpreted as a vice-grip on power rather than a service by the end users.

In my experience, (many years designing information and knowledge management solutions), most end users would like the option of searching for themselves and then deciding whether or not to request assistance.

Of course it may also be true that the system in use is so complex or so awkward to use that most end users (usually bereft of training) find it too hard to use and so have to fall back on asking the information professional. However, if this is the case then there will invariably be a backlog of requests and end users will be frustrated because they have to wait days or weeks for a response. In this environment, end users begin to feel like victims rather than valued customers or ‘clients’.

The obvious answer is to make it easy for end users to find what they are looking for but this obvious answer seems to escape most of us as we continue to struggle with the obscure vagaries of the existing system and an often impenetrable wall of mandated policies, processes and official procedures.

If we really want a solution, it’s time to step outside of the old and accepted model and provide a service to end users that end users actually want, can use and appreciate. If we don’t take a wholly new approach and adopt a very different attitude and set of procedures then nothing will improve and end user dissatisfaction (and anger) will grow until it reaches the point where they simply refuse to use the system.

End users are not stupid; end users are dissatisfied.

One of the core problems in my experience is an absence of an acceptance of the fact that the requirements of the core, professional users are very different to the requirements of the end users. At the risk of oversimplifying it, end users only need to know what they need to know. End users need a ‘fast-path’ into the system that allows them to find out what they need to know (and nothing more) in the shortest possible time and via the absolutely minimum amount of keystrokes, mouse-clicks or swipes.

End users need a different interface to a system than professional users.

This isn’t because they are less smart, it is because the ‘system’ is just one of the many things they have to contend with during a working day, it is not their core focus. They don’t have time (or the interest) to become experts and nor should they have to become experts.

If end users can’t find stuff it isn’t their fault; it is the system’s fault.

The system of course, is more than just the software. It is the way menus and options are configured and made available, it is the policy and procedures that govern access and rights to information. It is the attitude of those ‘in-power’ to those that are not empowered.

If you want happy and satisfied end users, give them what they need.

Make sure that the choices available to an end user are entirely appropriate to each class of end user. Don’t show them more options then they need and don’t give them more information than they are asking for. Don’t ask them to navigate down multiple levels of menus before they can ask the question they want to ask; let them ask the question as the very first thing they do in the system. Then please don’t overwhelm them with information; just provide exactly and precisely what they asked for.

If you want the end users off your back, give them what they need.

I fall back on my original definition of a Knowledge Management system from 1997, “A Knowledge Management system is one that provides the user with the explicit information required, in exactly the form required, at precisely the time the user needs it.”

With hindsight, my simple definition can be applied to any end user’s needs. That is, please provide a system that provides the end user with the explicit information required, in exactly the form required, at precisely the time the end user needs it.

What could be more simple?

More references:

The IDEA – 1995

Knowledge Management, the Next Challenge? - 1997

Whatever happened to the Knowledge Management Revolution?  – 2006

A Knowledge Management System – A Discourse – 2008

I may be wrong but I think it was AIIM that first coined the phrase Enterprise Content Management to describe both our industry and our application solutions.

Whereas the term isn’t as nebulous as Knowledge Management it is nevertheless about as useful when trying to understand what organizations in this space actually do. At its simplest level it is a collective term for a number of related business applications like records management, document management, imaging, workflow, business process management, email management and archiving, digital asset management, web site content management, etc.

To simple people like me the more appropriate term or label would be Information Management but as I have already covered this in a previous Blog I won’t beleaguer the point in this one.

When trying to define what enterprise content management actually means or stands for we can discard the words ‘enterprise’ and ‘management’ as superfluous to our needs and just concentrate on the key word ‘content’. That is, we are talking about systems that in some way create and manage content.

So, what exactly is meant by the term ‘content’?

In the early days of content management discussions we classified content into two broad categories, structured and unstructured. Basically, structured content had named sections or labels and unstructured content did not. Generalising even further we can say that an email is an example of structured content because it has commonly named, standardised and accessible sections or labels like ‘Sender’, ‘Recipient’, ‘Subject’ etc., that we can interrogate and rely on to carry a particular class or type of information. The same general approach would regard a Word document as unstructured because the content of a Word document does not have commonly named and standardised sections or labels. Basically a Word document is an irregular collection of characters that you have to parse and examine to determine content.

Like Newtonian physics, the above generalisations do not apply to everything and can be argued until the cows come home. In truth, every document has an accessible structure of some kind. For example, a Word document has an author, a size, a date written, etc. It is just that it is far easier to find out who the recipient of an email was than the recipient of a Word document. This is because there is a common and standard ‘Tag’ that tells us who the recipient is of an email and there is no such common and standard tag for a Word document.

In our business we call ‘information about information’ (e.g., the recipient and date fields on an email) Metadata. If an object has recognizable Metadata then it is far easier to process than an object without recognizable Metadata. We may then say that adding Metadata to an object is the same as adding structure.

Adding structure is what we do when we create a Word document using a template or when we add tags to a Word document. We are normalizing the standard information we require in our business processes so the objects we deal with have the structure we require to easily and accurately identify and process them.

This is of course one of the long-standing problems in our industry, we spend far too much time and money trying to parse and interpret unstructured objects when we should be going back to the coal face and adding structure when the object is first created. This is of course relatively easy to do if we are creating the objects (e.g., a Word document) but not easy to achieve if we are receiving documents from foreign sources like our customers, our suppliers or the government. Unless you are the eight-hundred pound gorilla (like Walmart) it is very difficult to force your partners to add the structure you require to make processing as fast and as easy and as accurate as possible.

There have been attempts in the past to come up with common ‘standards’ that would have regulated document structure but none have been successful. The last one was when XML was the bright new kid on the block and the XML industry rushed headlong into defining XML standards for every conceivable industry to facilitate common structures and to make data transfer between different organizations as easy and as standard as possible. The various XML standardisation projects sucked up millions or even billions of dollars but did not produce the desired results; we are still spending billions of dollars each year parsing unstructured documents trying to determine content.

So, back to the original question, what exactly is Enterprise Content Management? The simple answer is that it is the business or process of extracting useful information from objects such as emails and PDFs and Word documents and then using that information in a business process. It is all about the process of capturing Metadata and content in the most accurate and expeditious manner possible so we can automate business processes as much as possible.

If done properly, it makes your job more pleasant and saves your organization money and it makes your customers and suppliers happier. As such it sounds a lot like motherhood (who is going to argue against it?) but it certainly isn’t like manna from heaven. There is always a cost and it is usually significant. As always, you reap what you sow and effort and cost produces rewards.

Is content management something you should consider? The answer is definitely yes with one proviso; please make sure that the benefits are greater than the cost.

When we were all learning about what used to be called Data Processing we also learned about the hierarchy or transformation of information. That is, “data to information to knowledge to wisdom.”

Unfortunately, as information management is part of what we call the Information Technology industry (IT) we as a group are never satisfied with simple self-explanatory terms. Because of this age-old flaw we continue to invent and hype new terms like Knowledge Management and Enterprise Content Management most of which are so vague and ill-defined as to be virtually meaningless but nevertheless, provide great scope for marketing hype and consultants’ income.

Because of the ongoing creation of new terminology and the accompanying acronyms we have managed to confuse almost everyone. Personally I have always favoured the term ‘information management’ because it tells it like it is and it needs little further explanation. In the parlance of the common man it is an “old un, but a good un.”

The thing I most disliked about the muddy knowledge management term was the claim that computers and software could produce knowledge. That may well come in the age of cyborgs and true artificial intelligence but I haven’t seen it yet. At best, computers and software produce information which human beings can convert to knowledge via a unique human cognitive process.

I am fortunate in that I have been designing and programming information management solutions for a very long time so I have witnessed first-hand the enormous improvements in technology and tools that have occurred over time. Basically this means I am able to design and build an infinitely better information management solution today that I could have twenty-nine years ago when I started this business.  For example, the current product RecFind 6 is a much better, more flexible, more feature rich and more scalable product than the previous K1 product and it in turn was an infinitely better product than the previous one called RecFind 5.

One of the main factors in them being better products than their predecessors is that each time we started afresh with the latest technology; we didn’t build on the old product, we discarded it completely and started anew. As a general rule of thumb I believe that software developers need to do this around a five year cycle. Going past the five year life cycle inevitably means you end up compromising the design because of the need to support old technology. You are carrying ‘baggage’ and it is synonymous with trying to run the marathon with a hundred pound (45 Kg) backpack.

I recently re-read an old 1995 white paper I wrote on the future of information management software which I titled “Document Management, Records Management, Image Management Workflow Management...What? – The I.D.E.A”. I realised after reading this old paper that it is only now that I am getting close to achieving my lofty ambitions as espoused in the early paper. It is only now that I have access to the technology required to achieve my design ambitions. In fact I now believe that despite its 1995 heritage this is a paper every aspiring information management solution creator should reference because we are all still trying to achieve the ideal ‘It Does Everything Application’ (but remember that it was my I.D.E.A. first).

Of course, if you are involved in software development then you realise that your job is never done. There are always new features to add and there are always new releases of products like Windows and SQL server to test and certify against and there are always new releases of development tools like Visual Studio and HTML5 to learn and start using.

You also realise that software development is probably the dumbest business in the world to be part of with the exception of drug development, the only other business I can think of which has a longer timeframe between beginning R&D and earning a dollar. We typically spend millions of dollars and two to three years to bring a brand new product to market. Luckily, we still have the existing product to sell and fund the R&D. Start-ups however, don’t have this option and must rely on mortgaging the house or generous friends and relatives or venture capital companies to fund the initial development cycle.

Whatever the source of funding, from my experience it takes a brave man or woman to enter into a process where the first few years are all cost and no revenue. You have to believe in your vision, your dream and you have to be prepared for hard times and compromises and failed partnerships. Software development is not for the faint hearted.

When I wrote that white paper on the I.D.E.A. (the It Does Every Thing Application or, my ‘idea’ or vision at that time) I really thought that I was going to build it in the next few years, I didn’t think it would take another fifteen years. Of course, I am now working on the next release of RecFind so it is actually more than fifteen years.

Happily, I now market RecFind 6 as an information management solution because information management is definitely back in vogue. Hopefully, everyone understands what it means. If they don’t, I guess that I will just have to write more white papers and Blogs.