This topic first appeared in a presentation from our Director of PPM, Paul Cegles at our 2018 User Conference.
“We deliver our clients sound best practices when conducting Business Process Re-engineering events. By allowing for flexibility within the BPR framework to optimize processes, and providing a strong understanding of how to couple these processes with our Devex PLM technology, we are able to support our clients in achieving a best-in-class sustainable solution” - Paul Cegles, Director of PPM, Selerant
Implementing a PLM system for product digitization often generates a laundry list of processes that need to be updated across departments.
Yet, designing a system where all data points and workflows work in tandem and meet the needs of multiple teams with competing priorities isn’t for the faint of heart. Changing one area can often seem to impact another, and so on. Creating a comprehensive, multi-stage plan for business process changes using PLM seems impossible when so many systems, processes, departments are connected together.
When working with our clients, our services team uses a systematic three-step process to help unpack this complexity and identify, management and prioritize process changes:
- Map out a future-state NPD processes and each phase gate
- Create templates and workflows for subprocesses, including ingredient data, specifications, costing, specs, and more.
- Set documents and product field parameters, including governance, ownership and data definition standards.
Let’s take a look at how some of these steps work in practice with some real-world examples from brands using Devex PLM.
Map Out a Full New Product Development Process
The very first course of updating business workflows is to map out an ideal new product development (NPD) process and then identify all in-scope processes that can be realistically included in a PLM implementation project.
You don’t have to stick to this master plan as the project develops, but the following exercise makes it easier to break process revisions into clear, prioritized stages.
Perform an “As-is” Analysis
It’s easy to guess what an ideal NPD process should look like and often, many process pain points will be obvious. However, before you jump into creating a new process map, it’s important to step back and evaluate how things currently work across departments. What we find is that most people within an organization don’t agree on what current processes themselves actually look like, let alone how they should be improved.
To start, perform an as-is analysis that gauges how employees involved in NPD complete their work, at what stage, which documents and data they use, and what they require to complete their work more efficiently. This could be in the form of user surveys, pulling reports on document access and use, and other formal or ad-hoc business analysis methods.
The ultimate goal is to identify the most common opportunities for process improvements, which will later be translated into system design requirements.
Create a gap analysis
Once you’ve agreed upon how processes work now and how your teams would like them to function, we recommend conducting a gap analysis between current state processes, to-be processes and PLM functionality.
This should take about 2-4 hours, and we recommend using a pictorial process model to visual changes that will be made to daily work and to product data and formats.
The deliverables from this gap analysis session should be:
- Clear documentation of current processes
- A description of primary disconnects in current process operations, including major bottlenecks, recurring points of data duplication, and error-prone activities.
- An outline of to-be processes.
- Identified stakeholders and planned training events.
Several users emphasized that continually publishing these deliverables to internal users is the key to then taking this analysis into adoption of new business process by employees:
“PLM and Devex wasn’t a popular topic during launch, but fast forward to today we have project managers eager to get in and configure the modules for their use,” said the Innovation Manager at a major cosmetics company, “That’s all because we map out processes pictorially and how we reduced waiting and waste and how we reduced a 50 step process into a 10 step process. We’re always putting out bulletins about how we’ve improved old processes, and it gets people excited about the capabilities and possibilities.”
An alcohol manufacturer also suggested using live data for each group when creating a new process map and conducting these exercises with stakeholders:
“If you’re training with someone else’s data, they may not get it. Until we actually had our products, processes and materials that they could follow through in our new product map, it was hard for users to understand how it would work,” said a brand representative.
Create Templates And Workflows For One or Two Subprocesses
Once you have a full to-be NPD process mapped out for your organization or department, actually implementing all of those theoretical processes can seem overwhelming.
To maintain a viable scope--and generate quick wins with PLM--brands should focus on identifying and optimizing one or two of the major pain points within a subprocess, such as ingredient or formulation management, packaging, vendor communications, or regulatory approval.
For example, a national cereal brand we recently worked with wanted to digitize its full NPD process, which was managed entirely through Excel files. Instead of just overhauling their entire Excel structure, however, they opted to identify just the very top data pain points in each sub-process in order to create rapid improvements.
The brand started by conducted a four-month legacy review process of its ingredients management subprocess. This included:
- Modeling out activities at each stage gate.
- Distributing user surveys and gathering feedback on how work is conducted.
Through this research, they discovered that data was still be hand copied at time and that email communication between collaborators was inconsistent, even in steps that were perceived as efficient. The largest data offender? Their vendor request templates.
Armed with this knowledge, the brand was then able to limit the scope of their initial business process re-engineering project to just their vendor data management tasks. They solved this single challenge by revising their ingredients process in the following steps:
- Developing standard vendor request templates for supplier data submissions and review.
- Creating a pre-workflow for R&D group that collects and validate vendor information in 16 days.
- Creating a primary workflow that allows R&D and Quality groups to verify ingredient specifications in 19 days.
- Standardized SKUs to allow the transactional team to input cost, pack size, branch plant and other manufacturing needs on the specification in 3 days.
- Agreed on and developed one point of data entry from the company’s SPC system used in manufacturing plants into its Devex system.
- Used built-in data analytics to track performance differentials between stages and materials and set a baseline for process efficiency.
By standardizing vendor and manufacturing templates, reviews, and workflow rules, and connecting those data points to food scientists’ internal ingredient specification documents, the entire team is able to get and act on information for ingredient development much more quickly and efficiently. Devex can do the heavy lifting in terms of validating data and maintains a single source of truth.
“You have to have executive support and a steering committee that makes it an important project for the company,” said a leader from the brand. “What has worked over time is that we get the SME involved early. It’s not just an IT or R&D project, it’s a company project and people using the system are part of the project team from the beginning and we’re getting their buy-in over the course of the project.”
Set Documents And Product Field Parameters
The importance of a clear data strategy can’t be overstated when developing new NPI plans and subprocesses. Without an overarching strategy for data management, many brands end up with PLM becomes just another repository for isolated data instead of a cross-functional system.
When overhauling a business process around PLM, consider the following when determining document and data format, usage, and access:
Data duplication/harmonization: Which business system will be used as the master record of data and product data fields? Will data created in PLM data be used to set data fields, or will it be used to intake, validate, store data from other business systems, or both?
User access: Who needs access to which data fields and templates, and how can templates be designed for different levels of access without restricting usability and share-ability of cross-functional data?
Workflows: What order should product data and documents be reviewed and approved? What controls need to put into place to ensure documents don’t
Governance: Where is the system hosted? Where will your employees you submit change requests? Which department(s) should front the cost of the system implementation and maintenance, and reporting if it's a shared resource?
For instance, we recently worked with an international chocolatier that needed to standardize one shared PLM system and harmonize product data fields across three different brands. Balancing data access and standardizing data fields so they could work seamlessly across and for three different companies, product lines, and departments was no easy task.
Here are core choices they made around their product data management from the top down:
- Product fields: Data formats are driven by a shared ERP system, with PLM validating ERP data for usage within the product development templates. Show/hide fields enable each department to navigate only the fields they need without extensive system customization.
- List values: For list values within PLM, decision tables are used to validate values between systems.
- Data access and user permissions: Workflows standardize data access and user permissions. Every user can see only the data that belongs to their brand. At the same time, administrations can permit some items to be shared between employees for collaboration within co-branded product lines.
- Approval workflows: The team standardized workflows across formulation, packaging and other processes across all companies, so data and product changes can be shared between co-manufacturer product lines and shared materials
- System Governance: For ease of maintenance and support across three brands, the companies moved the PLM engine from an on-premise hosting solution to a cloud environment.
By addressing all of these areas of data management at the beginning, the implementation avoided re-work and customization, making it more cost-efficient for each company under the parent brand.
“I’d say it to anybody, try to minimize the amount of customization you do the system because it does complicate the maintenance and support,” said one of the brand’s leaders, emphasizing that the most important critical decisions around data fields are who can access them and how. “User permissions and governances are key decisions. it can take a long time but it is key that you do that.”
Business process re-engineering is a complex activity, but it doesn’t need to be uncoordinated. Creating a process map, identifying core areas for improvement in subprocesses, and standardizing data and documents around these core areas can help save on implementation time and create a structured PLM implementation.