In keeping with what is becoming an Archstone Consulting tradition, I am turning to music to drive home my theme. Forget Rod and Marvin, I'm going to the Gods of Rock 'N Roll themselves, the Rolling Stones, who preach to us that "You can't always get what you want." But more on that later...
In 2006, I was helping a client procure 17 TB of tier 1 data storage (a TB is ~1000 GB) to support their call center operations -- at the time this was a generous amount of storage space for a mid-market company and an expensive (~$1M), complicated acquisition. Five years later, I have a NAS device with 6 TB of data storage in my home office that I purchased for under $500 online.
The moral of the story is not the worn adage about the incredible decline in technology costs, nor is it about extrapolating Moore's Law beyond transistors. This story is about the explosive growth in our ability to retain information. Facebook users now collect, post, and organize 30 billion pieces of content each MONTH. Yet somehow companies seem unable to show similar capabilities when it comes to spend data.
According to a 2009 study, more than 35 percent of global companies will regularly fail to make insightful decisions about significant changes in their business and markets in 2012. This matches my experiences -- even recent ones -- performing spend analyses at clients of various sizes and industries. Here are some typical things I find under the covers:
- The client has difficulty pulling accurate data (requires several iterations and often is returned with the wrong date range, duplicate data, or the wrong type of data)
- Data that should be tracked is not
- Existing data fields aren't populated, are sparsely populated, or are populated in an inconsistent manner
- The data does not cover the right level of detail
- The data resides with the supplier or in .pdf invoices
- The data contains inaccurate information
- Reporting focuses on irrelevant data
- The client has difficulty replicating the spend analysis
I can work with a client to download 250 columns of data that have been populated since 2001, yet it will take three weeks of data manipulation to extract minimally useful information from that data. Improvements have been dramatic at a few companies, but most businesses seem to have no more insight than was available to them back when I bought the 17 TBs of data in 2006. Why, in 2011, with the growing adoption of barcoding, RFIDs, electronic data transfers of invoices, automated BOMs, automated goods receipts, and endless space for standardization to the nth degree, is it so difficult to get complete and reliable information? Better yet, why do we care?
To answer the first question, there are many hindrances to obtaining complete and reliable information. Three factors include:
- Lack of a standard structure to transfer spend information
- EDI is becoming more common for invoice data, but typically it does not include a high-level categorization structure like UNSPSC (my preference is for customized categorization - a viewpoint that probably compounds the issue)
- Deeper level 2 or level 3 categorization, needed for complex categories like MRO or compliance activities, are not even remote considerations.
- Companies are not as automated as we would expect. According to 2011 research by The Hackett Group:
- Only 25% of invoices are electronic for non-top performers
- In most industries, fewer than 50% of POs are electronic
- Data maintenance remains a low priority
- A common example is the same product having two different material codes
- Vendor Master management has been improving but who hasn't seen "IBM", "I.B.M.", and "Int. Bus. Mach." listed as three independent suppliers?
And why do we care? Some people may prefer to make decisions based on their gut, or correctly want to avoid "analysis paralysis." I argue that using common sense about your business can get you 75% of a sourcing program (aligning the right resources to the right spend categories 75% of the time), but that still leaves a mountain of money on the table. Only competent spend data analysis can:
- Identify demand management opportunities
- Track internal compliance with policies
- Track price variance and overcharges and compliance with payment terms
- Enable effective Supplier Relationship Management
- Assist in properly aligning to the remaining 25% of the organization
And even with the appropriate resource allocation, how do you effectively source complex categories like Creative Services and MRO without detailed, organized information? If done correctly, spend analytics will require minimal maintenance and actually help speed decision making rather than slow it down.
So how do you fix the shop? It doesn't take a Herculean effort and gobs of resources -- but it does take commitment. Some considerations include:
- Updating fields in the purchasing system to match data collection needs for the spend categories
- Developing sourcing categorization that is linked but distinct from the Chart of Accounts
- Auto-populating forms where possible to drive consistency (both through EDI and through drop-down menu selections)
- Investing in the resources to analyze the data (e.g., tools, trained internal resources, an outsourcer)
- Having a Center of Excellence dedicated to data integrity; they should focus on root causes of data integrity issues and making the updating process low-touch and repeatable
- Developing training both for data entry personnel and data consumers
In the Stones' immortal classic (clearly written with spend analysis in mind), they said you can't always get what you want. What you want is a comprehensive, accurate, and automated view of your spend that requires no maintenance and costs as much as a Happy Meal, and that vision is not attainable in the next three to five years. But, by making some sensible improvements to your processes and leveraging inexpensive third party technology solutions to help sort, cleanse, and manage your spend data, you can get a 90% view of your spend that will empower good decision making. To finish the Stone's thought -- if you try some small steps and give it a little time, you just might find, you get what you need.
- Ryan Graham, Director, Archstone Consulting