Product Information Management (PIM) Data Governance, by Jackie Roberts

a great article by Jackie Roberts on data governance for Product MDM

One constant truth in the business of data is change. The most critical factor in master data management is agility, within both process and software design. Agility means responding to and managing never-ending changes in the critical data used to support operational decision-making. I firmly believe the ability to respond to changes in the data is a must in the world of Product Information Management, or PIM, as the specialty is sometimes known.

Master Data Management is a broad classification of processes, governance, and software tools used to manage information such as customer and/or product data. CRM (customer relationship management) applications can support the organization’s interactions with employees, members, clients, customers and supply base for marketing, customer services and technical support. Typical data elements in CRM include names, titles, email addresses, physical addresses, phone and fax numbers, etc.

One crucial note is that there is a different expertise required to manage and structure a data governance policy for customer (CRM) data as compared to product (PIM) related data.

Product Data

The definition of PIM is managing information about products, which may also include services. Product data can include equipment, assemblies, spare part components, and commodity type items such as office supplies or hardware items, i.e. bolts, screws, nuts, etc.

The data management and governance challenges escalate with the vast number of variations used to describe a single product. Adding further complexity to product data management is that of sell-side vs. buy-side. Sell-side data is a controlled data set of product information; as a result, the data and the governance is structured and owned internally by the manufacturer or supplier. If the governance is structured intelligently, the structure enables multiple data uses such as exports to web catalogs, print catalogs, engineering libraries, and more.

Buy Side of Production Information Management

Buy side is more complex, depending on the size of the operation. Buy side is the collection of transaction data (product or service) to support the operation of the plants and facilities. A critical aspect of collection and management of spare part information is to support the planning of physical inventory to enable the uptime of the facility. There is a fine balance of inventory cost versus ensuring the maximum uptime, which can be very challenging. It is not uncommon for a large global manufacturer to exceed 15,000 suppliers / OEMs equating over half a million submitted data records a year. These data records need to be reviewed, verified, classified, structured and referenced to ensure there aren’t any duplicate records created in the ERP system.

There are many variations in the data sources to support the operations of a facility. For instance, spare parts data is submitted from the equipment designers via engineering bills of material, or through individual purchasing requests from maintenance staff. The spare parts data includes the manufacturer name, part number, classification name (noun), unit of measure representing how it is sold, and additional information to describe the physical characteristics. However that same OEM part could be submitted as a supplier part with a different supplier part number, a different classification name and no mention of the OEM part number.

The result is the same part data setup with two different master records, each with different contracts for purchases at different prices and stored in inventory multiple times. This results in either excess cost and inflated inventory or a spare part that is not available resulting in the shutdown of the plant line. This is why the data governance and master data management business processes are so critical to an efficient and streamlined operation.

PIM Data Governance

A product information governance project may appear to be a daunting effort when you’re beginning to structure the data rules.

My best advice is to take the time to develop a data roadmap to provide a clear and precise understanding of the data and its use within the organization. The road map should detail how data is required and submitted for use within the enterprise, account for the multiple uses of the data (purchasing, engineering, marketing, and maintenance), plus the required data elements and structure needed to accommodate each software system.

Jackie Roberts Governance Template

As an example, let’s explore how the data is provided to the enterprise. There are multiple sources of data, from engineering drawings created internally or provided by suppliers, maintenance requirements, buyer requests, and more. With an understanding of the source data, a clear data requirement enables an improvement in the quality of the data provided to the organization.

Starting at the initial contract to source a new piece of equipment for the plant, you may include with the equipment specification a spare part data requirement plus a template for the supplier to provide the spare part information. The contract deliverable should include the completed spare parts list that was required per data governance. Now your master data management process has been simplified and data quality has improved.

In the roadmap, the required data elements are defined to support the business requirements. For a large global manufacturer the governance may include a structure for equipment numbers, location structure for the equipment and basic data governance elements specific to the master record.

Commonly required elements will include:

  • Manufacturer Name,
  • Manufacturer’s Part Number,
  • Noun classifications,
  • Technical descriptive attributes,
  • Sequencing of attribute display order,
  • Units of measure (typically a purchasing rather than a use UOM, sometimes referred to as a disbursement UOM),
  • Price
  • Volume purchase prices,
  • Purchasing category,
  • Lead time,
  • Warranty information,
  • Language translation requirements,
  • Other classifications such as ECCN, UNSPSC
  • Any other descriptive elements to ensure smart purchasing decisions and stock strategy for items inventoried.

Benefits as a Result of Data Governance

There are many benefits of implementing an innovative data governance and master data management system. Many of the basic benefits, both in process and cost, are:

  • Reducing inventory through identification of duplicate items,
  • Facilitation of inventory sharing and internal purchasing programs,
  • Reduced employee time spent searching for items,
  • Common spare part usage strategies,
  • Reduced downtime in manufacturing equipment due to lack of information availability,
  • Ability to manage inventory using a just in-time model.

Data Governance supports both indirect and direct cost savings. Businesses can begin to embrace the definition of operational data as an asset of the corporation, ensuring improved data accuracy and confidence of the data users.

About the Author

Jacqueline Roberts is the VP at DATAForge™. She is a data enthusiast and firmly believes a successfully implemented MDM program will improve engineering, maintenance and purchasing data processes, which enables inventory management and cost-saving initiatives. You can follow her on Twitter at or visit

Tags: , , , , , , ,

11 Comments on “Product Information Management (PIM) Data Governance, by Jackie Roberts”

  1. Prashanta C 01/15/2013 at 7:45 am #

    Jackie, this is an excellent post on importance of data governance in PIM solution.

    One other aspect which is profoundly useful is the hierarchical categorization of products for easier access. Typically we create sales, marketing, manufacturing hierarchies and the like so that users from these departments have an efficient way to locate the products which they are creating or modifying. This is of tremendous help and significantly contributes to the governance.

    You probably addressed this when you mentioned PIM solutions allow easier searching (locating) of products.

    Thank you.

    • Jackie Roberts 01/16/2013 at 8:22 am #

      Prashant, thank for the your comment. I absolutely agree on the usefulness of a hierarchical categorization of product data and I firmly believe it is mandatory for a number of reasons including ease of search but also a method of sourcing for a buying team. Hierarchical referential information is also extremely beneficial in the downward stream of data use in both engineering & maintenance. Typically we will have a number of different hierarchical schemas associated to the product data, including the purchasing type categories (proprietary customer developed or UNSPSC) and also an asset (equipment) number and location hierarchical schema, for instance the spare part is associated to an piece of equipment and where that equipment is installed enabling the setup of maintenance tasking and inventory management for spares.

  2. Andrew Yee 01/16/2013 at 3:55 pm #


    Great posting, clearly explains the challenges of MDM and the need to define an MDM process within any large organization. Having a clear MDM strategy and approach to governance and understanding of the process to procure and sell goods makes our job easier when providing a software tool like MDO – MasterDataOnline.

    Would like your opinion on our overview video which tries to summarize the same key points you have mentioned which we can achieve using a software tool.

    • Jackie Roberts 01/22/2013 at 9:25 pm #

      Andrew, thank you for your comments and I would love to hear how MDO alleviates the challenges of MDM in ERP systems. The video is a great representation of the issues with MDM when only an ERP system is implemented for purchasing. Most purchasing systems are not designed with the agility of product item revisions or updates which is standard requirement to address both manufacturer and / or supplier data. One of my beliefs is that a MDM strategy should include a software tool / catalog with the data management intelligence, provenance, governance, processes and procedures addressing the product data used enterprise wide. The emphasis should not only be focused on making the purchase but also how the data is used in engineering (drawing library), manufacturing engineering (tool or equipment bill of material), maintenance (asset structure, equipment referenced to spare parts and to a location of use) and operations (inventory planning and stocking).

      In order to provide effective MDM supporting the spend analysis to our customers that is cost effective and truly meaningful, the data (new & legacy) needs to be verified, structured to a classification system, referenced (managing duplicates & obsoletes), provide the BI analytics to manage inventory and create a new master org / local views or update master org / local views of existing records in the legacy system. For instances a common scenario is I have two of the same item in the purchasing / inventory system, one set up as a supplier part and the second as a manufacturer of the part, both items are inventoried. Once the data is identified as duplicated, the “business” side of the processes needs to start, combining inventory in the store or crib, review for outstanding orders, the contract needs to be updated, etc.

  3. N Venkatesh 01/21/2013 at 6:00 am #

    Jackie, this is well said. Exactly we were facing the same biz challenge of having same OEM name by multiple suppliers and offering this kind of DG solution on MRO engagement really gave great results to our customers. One of the best articles. Thanks!!!

    • Jackie Roberts 01/22/2013 at 8:16 pm #

      Thank you for your kind words. I think it would be great if your shared some of the PIM challenges you experienced, the processes implemented to overcome the challenges and the successful results. As experts in this industry, we can start to add the ROI benefits to the business case to justify the cost of data quality, governance and MDM to support our customers.

  4. Sylvain Gourvil 01/21/2013 at 5:36 pm #

    Nice point of view about the data governance related to PIM solutions.
    I’ll keep it in mind.

    • Jackie Roberts 01/22/2013 at 8:08 pm #

      Thanks Sylvain for your comment. I would love to hear about your experiences of product data management and governance.

  5. Frank Gysberts 01/25/2013 at 11:44 pm #

    Hi Jackie
    great to hear from others who have “suffered” for years as well in the MDM space. We use a quote “The measure of success is not whether you have a tough problem to deal with, but whether it’s the same problem you had last year…” John Foster Dulles, Time Magazines Man of the Year, 1954
    The reason we use this is that Companies do not address their master data issues yet often complain about what is really the same problem year after year. I think this has a lot to do with MDM not being “sexy” and always difficult to put together tangible business cases that support an initiative to address the problems. We have been involved in many data cleansing activities and have found that unless a tool is implemented after the cleansing exercise so that any possible creativity is removed from a master data steward, data cleansing will need to be repeated regularly. We are talking to a customer now who had been told to use cheap labour in other countries to create their master data rather than look at an electronic way to do this. This will not give the discipline and governance required to avoid duplicates etc in the future.

    We also are involved in the MDO tool and have had great results using this with SAP where the 40 character short description is so important for users in the field. SAP has Manufacturer part number functionality and we have used this but also found other ways to have one common item across multiple plants.
    For PIM data governance, we use MDO in the following ways.
    1. Supplier Self Service – Suppliers can log on and update their information, which when approved via workflow, can update SAP automatically.
    2. Via 3rd party connections e.g. GS1. Here suppliers update the GS1 database and this is synchronised with MDO
    3. Direct updates in MDO, then via workflow approval and subsequent automatic updates into SAP.
    The good thing with MDO is we can utlise a “description generator” so that only allowed values can be input into the description. This gives real time governance with human intervention only required for data quality check and approvals. It can also manage supercession chains and update BOM’s which removes much of the admin and complexity from data management in ERP.
    Would love to find out how others have put together the ROI and business case to avoid having the same problem year after year…

    • Dan Power 01/26/2013 at 9:50 am #

      Hi Frank,

      Jackie is on vacation – I’ll make sure she hears about your comment and suggest that she respond to it when she gets back. Thanks!

    • Jackie Roberts 02/04/2013 at 10:27 am #

      Frank, Awesome quote! I agree that the business case is an extremely difficult challenge to create but it is required in business these days to fund a MDM project. I also find that most of the time, the company organization puts the ownership of data projects in either purchasing or engineering. Without the processes or a governance tool to manage the schema of naming conventions, descriptive vales required, etc. it is an exhausted and frustrating experience. Most clients have more than enough work completing their regular jobs, let alone the time or expertise to administer a data quality project and the management of an ongoing data maintenance process.

      I chuckled at your “cheap labor” comment, many employees managing and processing records via excel or access will never get the job done, and will only ensure “NO” standards being followed resulting in “NO” data quality.
      With PIM data, the fact that both suppliers and manufacturers sell the same product increases the risk of duplication of records (product & vendor) resulting in increased inventory, different purchasing cost, etc. This is the ROI we identify at the onset of entering into a new customer MDM project.

      I cannot refer to MDO, but I can say that if there is not a data structure built into a web based data manage tool with processing rules, the data quality battle will not be won. Our tools & process are designed with the understanding of the product data exclusively. Because of our schema structure, agile processes and BI built in, we are able to process the millions of records with the data quality required and provide our customers the ROI to successful support the business case.

%d bloggers like this: