« The Vista Experience - Part 1 Look and Feel | Main | Question of TCO leaves Microsoft waiting for take off »

Avoiding the "Evils" of the Spend Analysis Debate

Posted on Wednesday, February 14 by Registered CommenterTim Minahan in | Comments2 Comments

The blogsphere has been ablaze lately with debate over the best approaches and solutions for spend visibility and analysis. (Indeed, ELP's own Editor-in-Chief Mark Perera recent chimed into the debate on these pages.) Solution providers and consultants have been beating their respective soap boxes, each trying to focus the debate on the portion of the challenge for which they have a solution. And the resulting crescendo is a confusing cacophony of advice.

I liken the arguments to the age-old monkey conudrum: hear, see, and speak no evil. (If this analogy doesn’t work for you, feel free to pick your own — possibly three card monte or the shell game.) I’ve even pulled together a simple-to-understand graphic to illustrate this point:

 Spend Analysis Areas Table.jpg

As the above graphic indicates, spend visibility and analysis vendors typically fall into one of three camps:

  • Data-wrangling experts argue that you can’t truly analyze spending without capturing all spend data and classifying it to a common classification schema (e.g., UNSPSC) at a level of detail that’s meaningful for analysis. They are right! Alliant Techsystems Director of Spend Management Greg Shifflett gives a perfect example of the power of detailed classification: “[Prior to using an automated spend analysis solution] we knew what we spend on adhesives and with which suppliers, but we don’t know the detail on what types of adhesives we have bought from whom.” However, data-classification only solves just one-part of the larger challenge — albeit the most challenging and burdensome part.
  • ERP, business intelligence (BI), and even some supply management solution providers promote their reporting and analysis capabilities as the solution to the spend analysis dilemma. Their argument: “We can provide all the reports you need and even let you drill into your spend…Just give us your spend data in a structured format.” And, for those of you thinking “but my ERP captures all my spend data in a structured format,” think again. Important spend data (e.g., ACH, P-card, outsourced spend, etc.) exists outside your ERP system. Even worse, as cited here earlier by Pricewaterhouse Cooper (PwC) consultant Jason Josko: “ERP systems were built for high-level financial analysis. The attribute and classified data procurement managers require to develop informed sourcing strategies is often missing from standard ERP reporting.”
  • Supplier intelligence service vendors focus on the importance of data enrichment, enabling parent-child matching, diversity tracking, and financial, performance, and balance of trade information. And project management solution providers note that data is useless unless you have the ability to take action on it by establishing and monitoring category management plans. Both are right about the importance of these capabilities. (This point is reinforced in Aberdeen Group’s latest Spend Intelligence Benchmark report.) But the value of such features are limited (and can even be dangerous) if built upon incomplete or misclassified spend data (i.e., efficiently getting to the wrong decision can be worse than making no decision at all.)

My point (in case you missed it): effectively managing spend requires all of the above capabilities.

You must have the ability to repeatedly access and classify all your spend data at detailed level. Each  supply management team member and business stakeholder must be able to quickly report on and analyze spend, compliance, and category information that’s most relevant to them. And you need to have the ability to define, share, and track category management plans based on this new-found intelligence. Settling for anything less is an incomplete solution to your spend management challenge.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (2)

You’re right, you can’t fix everything with a hammer – spend analysis is a complicated affair that requires a whole belt of different tools to address the different stages. But a ‘one-size fits all’ approach is not always effective…

Industry standards for categorisation are great within sectors where everyone has the same purchasing needs, budgetary constraints and policies (e.g. public sector), but how can these possibly be valid for every type of organisation within the private sector, where competition and differentiation come into play?

Even if categorisation can be standardised, can we really expect those making purchases at the front end to understand CPV codes and adhere to them? After all, aren’t we supposed to be following the US’ lead in a move towards a polarisation of functions within the P2P cycle, where those on the shop floor are given the power to buy the goods and services needed to do their job, but within the corporate constraints negotiated and controlled by a more strategic procurement department (headed by the CPO, of course). In that case, vendors should perhaps be focusing their attentions on an internal interface that can be customised to categorise items in a way that makes more sense to the individual organisation.
February 19, 2007 | Unregistered CommenterTerry Wilcox
Thanks for the good debate. In my observation there is nothing exists like "automating" spend analysis process. Yes you can automate some of the things but manual reviews, corrections are integral part of the process.

The first thing that you need to do is Data profiling, data analysis (See my blog http://manageyourdata.blogspot.com), then comes the source data quality. You also need to do lot of normalization on that and then classification on whatever taxonomy you choose. All this is integral part of spend analysis process. Once you are done with this, then only you have visibility on what you are spending, on whom, how much etc. This can only be partially automated process. As an engine can guess data up to an extent only.
August 27, 2008 | Unregistered CommenterPrashant

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
All HTML will be escaped. Hyperlinks will be created for URLs automatically.