The first step is to define the objectives of the spend analysis exercise. It is important to define the objectives clearly because that will drive data gathering and analysis efforts. Following are some common objectives of spend analysis
- Understand spend at a granular level so that sourcing team can identify saving opportunities.
- Understand key vendors so that procurement can define and execute a strategy for strategic vendors.
- Provide visibility to senior management on key spend areas impacting EBITDA margins. For example COGS(Cost of goods sold) spend, SG&A (Sales, General and Administration) spend.
The next step we create an inventory list of all source systems where your spend data reside. The goal is to ensure that the entire spend is captured for analysis. If your company have multiple business units, it is likely that you have multiple systems. So the scope of analysis will determine what systems you need to capture. A simple inventory table should capture all data points
SCHEMA FOR DATA CAPTURE
Since we are pulling data from multiple systems, they are bound to have a different set of fields. The first and foremost task is to identify what data you want to be captured and put it through a common data schema. A data schema is a simple definition of what fields you want to be captured and what they mean.
Since different systems have different nomenclatures for same fields, it is handy to have a common definition of the information you are trying to gather. It would be much easier to gather data from disparate systems if you have a common data schema/structure.
Now we have the data schema, the next step is to reach out to your IT team and have them check the effort required to pull the data from different spend systems. Few things to keep in mind
DEFINE THE FREQUENCY OF DATA REFRESH
Spend analysis done right is not a one-time activity. We get better value for our efforts if we refresh the data frequently. The frequency of data extraction depends upon the purpose of spend analysis. For example, if you are using spend analysis primarily to identify saving opportunities than a quarterly refresh should be sufficient. That way you can also track how the negotiated savings are being realized.
DATA EXTRACTION FORMAT
The other thing we keep in mind is how data will be provided. We prefer to have the data in the same file format (Excel, CSV) so that it is easy to consolidate the data from multiple systems. If you don’t have multiple systems then this should not be of any concern.
Most of the ERP systems have some way or fashion to categorize the spend transaction into unique buckets. The most common approach is to use General Ledger chart of accounts to categorize data.
In some cases, we might see homegrown or industry-specific nomenclature for classification. If this serves our purpose, we use that, if not, there are multiple options available for classification schema. Data needs to be categorized in unique buckets so that analysis is easy.
Here are some Industry standards we refer in order of their popularity:
- UNSPSC stands for United Nations Standard Product and Services Code. It is an open and global standard for efficient and accurate classification of products and services.
- NAICS stands for North American Industry Classification schema. It is used by federal agencies to classify businesses. It is the primary classification schema used by federal agencies for reporting statistics.
You might have existing GL based spend classification from your ERP system. But it is highly recommended that we reclassify the data into the new classification schema. There are multiple reasons for that
- The same spend might be misclassified and hence needs correction.
- If the category structure is not the same, you might have the same item identified differently across different systems.
- Most of the credit cards companies provide a classification for each transaction. It is generally in grouped into MCC (Merchant category codes).
That is good enough for T&E (Travel and entertainment) spend, however, if your employees are using credit cards for other material purchases, then you need to make sure that spend data is categorized in the same way as the spend from other systems.
Having said that, data classification is the most time-consuming part of the whole exercise. Few things we consider while classification:
GRANULARITY – HOW GRANULAR YOU GO WITH DATA CLASSIFICATION?
You might need granular categorization for certain categories like MRO and direct materials but for products like office supplies, you don’t need to classify it at each individual commodity level. So unless you are planning to present different types of pens your company purchases, it is useless to categorize the data at that level.
Even with fully automated classification systems, it is hard to achieve 100% accuracy so we focus on high ticket items. If we do a simple Pareto on your spend data, you would realize that 80% of your spend is captured by 20% of your transactions. So focussing on important items and ensuring they are correctly classified.
This is probably the most important step in the whole spend analysis exercise – We Slice and dice the data to identify saving opportunities.
The focus of data analysis is to identify trends based on your objectives. If the objective was enhanced visibility, then this exercise will be focussed on identifying spend trend over time. If the focus is on identifying saving opportunities, then the analysis will be on understanding price variance, supplier proliferation etc.
The last step in the spend analysis process is presenting the results to senior management or your stakeholders. Here are some common scenarios on how the output of spend analysis is used
HELPING STAKEHOLDERS UNDERSTAND THE SPEND TRENDS
The scope of this analysis is generally tied to a department. The opportunities presentation would be focused on Spend by the vendor, spend by category, individual category trends, and price trends for high ticket items. For example, IT director might be interested in understanding the trends in contingent labor or annual software maintenance trends.
This is generally requested by senior management. Some of the trends which are helpful are
- Overall spend trend year on year.
- Spend trend broken down by direct/ COGS(Cost of goods sold) vs indirect spend /SG&A (Sales and general administration) spend.
- Cost reduction year over year.