re-inventing the wheel (was: npo-accounting Digest, Vol 15, Issue 8)

[Daniel Pocock]

On 29/08/16 07:28, ssackett at adaxa.com wrote:

> In nearly every system selection process I have been involved with in 12
> years of running an open source ERP business there has been a detailed
> user requirements document and an attempt to compare these specific
> requirements with candidate system's capabilities.  My recollection is
> that there was no such process for this project, just a quick review of
> easily available information.  I have always been puzzled why a detailed
> functional requirements process seems not been followed for the proposed
> npo-accounting system .. perhaps a case of "if you don't know where you
> are going then it does not matter which road you take".
> 
> Before I get flamed, can I point out that I strongly support the idea of
> the "ultimate npoacct" being available to the many not-for-profits doing
> many valuable things within their respective communities, I am just
> disappointed at the slow (and clearly what I see as flawed) process of
> getting there.
>

I also tend to agree that it is more likely to be successful in the long
term if using/partnering with an existing solution, at least for G/L and
probably some of the A/P too.

Looking at two of the main concerns about existing systems:

- reporting: NPO fund reporting, especially for US compliance, is vastly
different from business reporting requirements.  In practice, most ERPs
need to adjust their reports for different countries anyway.  e.g. a US
balance sheet is not the same as a UK balance sheet.  A UK VAT return is
not the same as Australia's GST return (but both are similar).  ERP
vendors already understand they have to be extremely flexible and
provide stable APIs and schemas for people to extract data for
reporting.  If the right ERP is chosen, somebody may still need to make
all the NPO reports from scratch, but that is a lot less work than
making a whole ERP from scratch.

- independent funds within the NPO.  This also looks complicated at
first glance.  It is quite possible that none of the free software
solutions do exactly what is required in their existing schema.  Is that
the end of the road though?  Certainly not.  To give one example (and
I'm not writing this to recommend a particular product, but just to
demonstrate that vendors do listen and change), xTuple had a very
US-oriented sales tax system.  They consulted the community[1] on adding
VAT support in 2006, documented a very highly generalized solution that
should work for many countries and released[2] it in 2009.  When they
did this overhaul, they even included support for multiple taxes layered
on top of each other and provided a wide range of forms for configuring
them.  Could any of the existing ERPs be encourage to undertake a
similar overhaul of their department and cost-center classification so
that it becomes sufficiently general to satisfy NPO accounting?  Even if
it took 3 years for one of them to adopt such changes (the duration of
xTuple's VAT overhaul), would that be better than writing something
completely new?

Regards,

Daniel



1. https://www.xtuple.org/node/663
3.
https://xtuple.com/about/news/xtuple-unveils-powerful-new-erp-features-xtuple-33