Supply Excellence
← The Ethanol Debate: A Supply Management View Addressing the Services Spend Visibility Crisis →

Do You Really Need Spend Analysis Automation?

March 19th, 2007 · by Tim Minahan · 4 Comments · best practices, events, spend analysis

Countless analyst reports and enterprise testimonials (and even research conducted by Supply Excellence) point to the benefits of automated spend data management and analysis solutions. Automating data extraction, classification, enrichment, and analysis processes have been proven to improve spend visibility and leverage, enhance compliance rates, and uncover opportunities for cost savings of 2% to 12%.

Despite such positive indicators, many companies still ask the question: “Do we really need spend analysis automation?” The answer, according to Forrester Research, depends on the complexity of your company’s business and technology environments.

In a webinar last week — Beyond Spend Visibility: Turning Data into Action — Forrester Vice President and top Supply Management Analyst Andrew Bartels shared the simple reference below to help enterprises assess whether they should prioritize investments in spend analysis automation:

(Click image to enlarge.)

Slide1.JPG

As the above graphic indicates, Forrester argues that enterprises that will benefit most from automated spend analysis solutions are those that:

  • compete in multiple markets;
  • compete across multiple industries; and
  • have spend and transaction data spread across multiple ERP and legacy systems. (Not to mention all the external systems — e.g., P-card, ACH, 3PLs — that hold important spend data.)

This is a helpful (if simplistic) diagnostic. On the same webinar, Todd Grunert, Purchasing Manager of Supplies and Services at Abbott Laboratories said his company, which has grown through multiple merger and acquisitions, must capture spend data from over 70 sites around the globe. “Doing this manually was not a viable option for us,” said Grunert.

Considering its systems and business complexity, Abbott may be the penultimate validation for Forrester’s spend analysis assessment framework. Yet, Bartels was quick to point out that the framework should only be used as a guide: ”Even in a single ERP, single industry environment, the benefits from getting much more granular detail on your spend data can be significant.”

Indeed, a previous Supply Excellence post details the experience of a global technology company that thought it would solve its spend analysis problem by standardizing on a single ERP system. After the fact, the company’s supply team quickly realized that ERP systems are built for high-level financial analysis and lacked the detailed, line-item attribute data needed for meaningful spend analysis.

For example, ERP standardization reaffirmed what the supply team already knew – it’s company spent a lot on hardware. But the ERP data didn’t help commodity managers determine how much was being spent on high-end SPARC stations versus laptops. Nor could it shed insight into whether these items were being bought for development engineers or office administrators.

The company eventually adopted a solution to automate spend data collection, classification, and analysis, yielding opportunities for 5% cost savings on A and B spend categories (i.e., those they felt they had previously sourced and managed effectively) and 15% savings on categories that had been sourced and managed inconsistently. (Read the full case study here.)

I will be profiling the highlights from Forrester’s and Abbott’s presentations on spend analysis later this week. In the interim, listen to the replay or download their presentations for free here.

Email Forward & Bookmarks
[del.icio.us] [Digg] [Facebook] [Google] [Reddit] [StumbleUpon] [Technorati] [Email] More »

Tags:

4 responses so far ↓

  • 1 Kevin Brooks // Mar 19, 2007 at 12:23 pm

    Tim, the whole question about whether automated analysis is needed seems to me to reflect an ambivalent management attitude about information overload. Of course, more info is better, and of course automating can simplify tedium and make complexity your friend! But if it were that simple, automated analysis would be as ubiquitious as email.

    I was at a lunch the other day with Tom Stewart, editor of HBR, where a variant of this topic came up. Because of the volume and speed of data coming at today’s business execs, decisions are required at ever-faster velocity. We all like to assume that people make logical/rational decisions, but when you max out the input, decisions may revert back to relying on core beliefs and/or psychological reasoning rather than hard data/facts.

    Automated analysis can help bring order to this process, but it doesn’t necessarily cause people to make “better” decisions. This, I think, is the petard for those selling analytical systems.

  • 2 Tim Minahan // Mar 19, 2007 at 4:11 pm

    Thanks for the comment, Kevin. (Very “Tipping Point” of you.)

    I agree that data-overload is a serious issue that can stop a supply management program or business initiative in its tracks. Getting access to more information, more frequently than you can realistically take action on it can have negative connotations on three spectrums:

    1. Quality & Completeness — getting access to incomplete or erroneous data only causes more problems and can lead to misinformed decisions.

    2. Detail and structure — getting access to information that either lacks detail or that is detailed but lacks a logical structure or schema that is useful for comparison or analysis can also be problemmatic, generally resulting in frustrated users and low adoption or support for your program.

    3. Frequency — the nirvana of real-time spend intelligence — although technically feasible — is culturally and pragmatically deficient. Forrester’s Andy Bartels said it best in the webinar that I cite in this post: refreshing your spend data classification and analysis “once a quarter to once a month is optimal.” Any more frequent than that and “you are getting more information than you can act on. And you don’t want to get stuck in the analysis stage.”

    In short, repeatable access to detailed data that is structured for meaningful analysis can provide the insight necessary for companies to make informed decisions, particularly when accompanied by structured reporting and decision support tools. However, to your point, a company must assess its data assimilation threshold and determine how frequently it should conduct data analyses and in what doses. It must also ensure that its supply management team has the right skill set to assess this new information and develop the strategies necessary to take action on it.

  • 3 Supply Excellence » Addressing the Services Spend Visibility Crisis // Mar 20, 2007 at 9:48 am

    [...] Do You Really Need Spend Analysis Automation? March 20, 2007 Addressing the Services Spend Visibility Crisis by Tim Minahan at 9:48am [...]

  • 4 Prashant // Aug 27, 2008 at 8:25 am

    I work on procurement spend analysis for quite some time with the big customers. In real world there is nothing 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 upto an extent only

Leave a Comment