Digital Financial Reporting

XBRL has proven itself as a useful tool which regulators can leverage.  But will XBRL ever achieve broad use (i.e. mass adoption) and become something any business user can make use of to exchange information with another business user, independent of IT department assistance?

Exchanging information is tough, but IT departments can pretty much get any business system to talk to any other business system with one hand tied behind it's back.  XBRL provides leverage in this process even to IT departments, implementations of XBRL such as the US Federal Deposit Insurance Corporation (FDIC).  There is ample proof of this. But not every business has the budget of a big government regulator.

So, there are "paths" through XBRL. What I mean by path is that XBRL is a general purpose specification. No one uses all of XBRL and the same thing can be achieve in different ways. These different paths cause interoperability challenges, particularly when you use XBRLs best feature: extensibility. Each implementation today picks their own path which takes highly skilled XBRL experts and can be expensive. But what if one global standard path was created which software vendors could use to implement something workable for business users?

Is XBRL destine to remain only a tool IT departments can leverage? Or will business users ever be able to make use of XBRL: one business user exchanging business information with another business user without the involvement of the IT department?  That is why I am interested in XBRL.

Here are the scenarios which I believe might come to pass. It is hard to say which will play out, only time can answer that question.  But, understanding the scenarios provides clues if one desires to help their desired scenario to unfold.

• IT departments leverage XBRL.  This is where we are today. Why would the status quo magically change on it's own?  It won't. More time will not make XBRL usable by business people. Someone is going to have to do something different for the status quo to be different.
• Proprietary XBRL implementations.  Is it possible to make XBRL easy for business people to use? It is absolutely possible.  The problem is that there are many paths within XBRL to achieve this objective. Also the skills needed to pick the right path through XBRL take too much time to obtain. If individual software vendors implement systems which make use of their selected path through XBRL to achieve business user to business user information exchange, we end up with what the business intelligence (BI) world has which is BI systems which don't talk to each other. BI systems are not interoperable across BI software vendors. You can work around this, but that takes time and costs money. Perhaps this is the destiny of XBRL.
• Proprietary XBRL monopoly. One specific vendor could get lucky or do a good enough job to achieve the status of Microsoft Excel or Microsoft Word as a de facto standard path through XBRL.  One company would certainly like this, the one whose proprietary implementation became the standard; they get to become the monopoly.  This isn't all that bad, but it does have its drawbacks.
• De facto XBRL standard. Maybe one vendor will create their path through XBRL and other vendors will adopt that path.  For example, what if, say, XBRLS caught on and multiple vendors implemented that making some businesssystems able to talk to some other business systems for ad hoc business information exchanges. Some interoperability is better than no interoperability. Perhaps this could happen, it happened to PDF. It is very possible to create products which would yield what business users need from XBRL, there is increasing evidence of this. A de facto XBRL standard would be fine. The control of this whether this happens is really in the hands of the market. Not a bad thing.
• Ad hoc interoperability via global XBRL standard. Having some global standard path through the technical complex XBRL quagmire, rather than each implementation having to figure out their own path, build their own infrastructure; each implementation solving the same problem over, and over, and over. This certainly does not make business sense (i.e. look at all what the SEC system had to do to make XBRL work). But does XBRL International have the funding, the motivation, and the other pieces of the mix to create this path. Time will tell.

Let me be very, very clear about my point here. My point is not that XBRL is complex. It does not have to be complex. All you need to do is give up features. That sort of misses the point of using XBRL though. My point is not that XBRL can't work, in fact XBRL does work, there are more and more business systems which use XBRL but are not interoperable being built all the time.  It is not that every XBRL system has to be interoperable with every other XBRL system. The point is not that XBRL does not have value, it certainly does.

The point is that if a business person picked up XBRL and tried to exchange business information with another person, they could do it but it would be far to expensive.  All they need to do is throw a lot of money and time at the IT department and they could get whatever they might desire, and XBRL certainly contributes to what they could have.

But what if a business person could achieve what they need without the cost.  That is the point. Effeciency.  The more effecient XBRL is, the broader the adoption XBRL will realize. What could be achieved with XBRL is similar to what the personal computer and the electronic spreadsheet provided the business user: freedom from the IT department.

This is not about diluting XBRL and changing its nature as a general purpose tool.  XBRL International consciously chose to solve the problem of business reporting not just financial reporting. I think that was the right choice. But, there would be advantages to an easier to use path. This is not about creating something simplistic or taking features away. It is not about creating "THE" standard path through the XBRL quagmire, it is about creating "A" path which will likely fit 98 percent or more of all system use cases. Then, the business people can pay the technical people to solve 2 percent which does not fit, should they have that need. This means that 98 percent does not need to be reinvented every time XBRL is implemented.

Do you see other possible scenarios?