Subject: Re: Company centric view ends. Data goes to the Net.
Date: 06/15/2000
Author: Bob Haugen <>

>Mike Block said,
>Good, but what is REA?
Todd Boyle replied:
>I'm no expert at REA.  McCarthy is the person to ask.  Or Bob Haugen,
>he is an expert.
I (Bob Haugen) understand REA pretty well from a supply chain viewpoint; not so well from an accounting viewpoint.  But I'll comment on a few points, and leave the rest for Bill McCarthy.
First, an intro:  I am an old MRP/ERP programmer.  My career for several years was dealing with unhappy ERP customers,
whose businesses moved faster than the ERP system.  I mean literally, they could make and deliver products faster than they
could get the orders through the ERP system.  (For example, Ford recently announced that it takes them more than a month
to get a demand signal from the head of their supply chain to the tail.)
So I started looking for something better and found REA.  I use it for multi-company supply chain collaboration, not
>But REA has been around an awful long time, at least 15 years.  It has
>been studied to death by everybody from big 5 CPA firms to OOP developers.
I think "studied to death" is an overstatement.  PriceWaterhouseCoopers uses REA in something they call the Geneva project.  Ralph Johnson and his gang have been working with it for a couple of years.  I've been working with it since 1997.  It is just now coming into its own, for Internet business collaboration through groups like ebXML.
>The problem with REA is it's mis-labeled, as an accounting system.
>It is really a high-level, and very generalized model for business
>objects, or business systems. 
I would almost agree with that, except that it *is* also an accounting system.  But I use it as an object model for multi-company business collaboration.
>It establishes a number of abstractions
>that generalize business events.  It guides developers in the creation
>of software objects.  It was way ahead of its time.  For anybody who
>really wants to know what makes things tick, it's a goldmine.
>REA was created in 1980 apparently in response to circumstances that no
>longer exist, in my opinion. 
Could you justify that statement?  Most of the weaknesses of ERP systems that led me to abandon the ERP model for REA
still exist, although they are being compensated for (at great cost and complexity) in some of the newer developments of
Oracle, SAP, etc.  I see no current solutions for smaller businesses, and even those for big businesses are seldom satisfactory.
>For example, McCarthy identified the
>following weaknesses in the conventional accounting model:
How have those weaknesses been rectified?
Also, I would add:

5. It is difficult to handle multi-company collaboration problems like

  • virtual inventory,
  • vendor-managed inventory,
  • quick-response make-to-order manufacture,
  • event-driven continuous replenishment,
  • joint ventures with a lot of cross-transactions,
  • flexible outsourcing arrangements, etc.

6. It is difficult to integrate acquired companies - either the acquired company needs to change to the acquirer's internal systems, or the acquirer needs to do a major integration project.  If both companies had been using a joint REA system, it could only mean changing a few parameters.  (Utopian idea, I know, but doable.)
7. The logical links between planned events (like dependent demands) are fractured into many different disconnected subaccounts and therefore computing subsystems, e.g. sales orders, multiple levels of work orders, purchase orders, transportation documents, etc., instead of a uniform representation for all business events and connections between them.  (I hope that long sentence communicated ok...)
>Of course it's true, that Controllers hate software that can't deliver
>accurate numbers by the 10th after monthend. 
REA accounting systems are able to deliver accurate numbers immediately, any time, regardless of month end.  Cisco does
something like this, too, although I don't know how close they come to REA techniques - they call it "virtual closing". They say it is critical to their success.

The other aspect of "accurate numbers immediately" is that you can get many other conclusions from REA event records besides the usual accounting reports - because all the conclusions are just views over the basic records of economic events. 
Ralph Johnson and his gang have developed software that continuously updates any views you want, so there is not
even a performance penalty, which used to be a problem for REA implementations.
>So, the real failure of REA is that they didn't cost-justify a whole
>rip-and-replace to CEOs. Microsoft and the ERP vendors got in there with
>their COM story, and their ERP story, and lied about the cost, mostly.
I don't personally advocate rip-and-replace, but rather piecemeal growth, building on whatever systems a company has to develop supply chain connections instead of more internal systems. (That's where I think the payback is these days.)
COM is not really relevant in this context, being a technical software component mechanism.  ERP is something else again.
ERP grew out of MRPII, which was MRP kludged together with "traditional" accounting software, e.g. PO, PR, AR, AP, GL, etc.  I lived through this transition from MRP to MRPII,  and in my view, it represented going backward.  MRP is
a flow-oriented view of the world, similar to CPM and PERT in construction.  PO, AR, AP etc are order-based views
of the world, where every order is disconnected from every other order.  The problem with MRP is that it did not know anything about money.  REA is a  flow-oriented view of the world that unifies all resource flows, including cash flows.  That is why I need it for what I do.
>Resources, Events and Agents are cool. The model has lots of applications
>and there is no conflict with statutory financial reporting at all.
>REA is coming into increasing focus as an SCM solution, a solution
>for logistical and financial information systems which spans enterprises.
I agree 100%.
Thanks for the opportunity to comment,
Bob Haugen