Take these six steps to structure your product data and get ahead of the game.

In this article we set out the first steps a product manufacturer should take on their journey to structure product data. This is a simple process focused on understanding how your company works with information and data* about your products.


This process is usually called a data strategy. It will include a methodology for defining product data consistently across the business, a strategy to roll out appropriate capability across the business, and systems to provide consistent product data, enabling it to be shared externally with your customers and partners.


These steps should be taken before embarking on any investigation of the standards or technology. In order to comply with the UK and European processes emerging from the new ISOs 23386 and 23387 a manufacturer will first have to understand where their data is, who is using it and how it is being managed (if at all), and develop an internal process to bring it together. Do not underestimate the importance of taking these initial steps. Do not avoid taking steps to organize your own data processes simply because a standardised structure for data is not yet in place. Take the first steps so that you are ready to take the next.

What and Where is your Product Data?

Many manufacturers hold their data in several different places (spreadsheets, databases, ERP systems, accountancy programs, manufacturing records). This has the effect that information is added and updated in several disconnected places at the same time, giving the potential for duplication, errors and waste.


The risks of holding several unconnected data sources are many. For example,

1. You may publish information about compliance on your website which is not correctly correlated with the latest compliance information held within your internal systems, leading to incorrect specification decisions being made.

• A member of staff with local knowledge about suitable product applications may go sick or leave the business, and as a result colleagues are left hunting around for information which is only available on their computer.  

The aim of this first stage is to work towards a strategy for interconnected product data in your organisation, so that if anyone in the company is asked about a product, they can access and provide the same, correct answer rather than whatever they have on their PC.


Connecting your data sources essential for compliance, safety and conformance procedures but it also has benefits far beyond this, for example,

1. Launching a product or group of products, perhaps in several different countries, is a much more efficient process, allowing you to get to market quicker;

• Keeping your product information current becomes much more efficient than a manual process;

• As a business you’ll have a much better understanding of your current product offering and be able to improve your customer service relationships by improving consistency of messaging.

What follows is an example procedure for developing a data strategy which you can adapt to suit your own business structures. Whether you are a very small SME, a mid-level company, part of a larger group or a major international corporation, all companies will encounter similar issues, but the solution that suits will vary.


Remember that structuring your data is an ongoing improvement challenge; you may identify problems which will take time and resource to resolve, or the process may be relatively simple. Either way, data management is a process, not a one-off event.

Learning about your product data – a Process

Step 1: Get Executive Buy In

The first step is to ensure that your company leadership recognises that your company has a problem with product data management. 

• Collect together some examples of how poor data management is generating unnecessary risks and waste to the organisation, and examples of positive improvements by data driven companies.

• Speak to your MD and ask for an opportunity to present your initial findings to the Board.

• Ask for resource to support the next stages and a board-level key project sponsor.

The Plain Language Guide we are developing is designed to identify some of the risks your company may be facing and explores some of the arguments for change. When it is published it should assist you in communicating with your company leadership.


Remember that getting leadership recognition and sponsorship is in many ways the simplest task but will help allocate resource to the next stages. More difficult is to help those responsible for generating and managing your data to recognise the problem your current systems produce.

Step 2: Get Management Buy in - Management Workshop

Once you have leadership support, set up a kickoff workshop for second level management. This should be devised to sell to the business the need to change by helping them identify the nature of the problem within your business.


Participants should include second level management across the company – for example IT, Product Data Management, Product Management, Sales, Marketing, Quality, HR, R&D – depending upon your company structure.


Choose a workshop facilitator who will encourage participation by all in the meeting. Throughout the workshop take notes on a flipchart or whiteboard so that people can see the ideas as they are generated; this will prompt them to add their own.


The workshop tasks could follow this type of sequence:

1. Where do we store our data?
   
   
   1. Ask the participants to identify all the places where data about your products is currently stored.
   
   
2. Examples may include ERP software, printed documents and PDFs, websites, local hard drives, CRM software, server files, calculation tools, object identification databases, emails and any other databases the company holds or that are held by third parties on their behalf.

• Treat this as an exercise in generating as many ideas as possible. Remember to focus on the information about your products, existing and legacy.

• Who decides my name?
  
  1. Select a product that you know has different names within or outside your organisation. Ask the participants to identify the product and then identify who decides the name of the product.
  
  
• You may find it useful to begin this step by introducing an everyday item which has several names to explain the principle. Examples we have seen used are often food based – one famous example is the analysis of the common names for a bread roll.

• You’ll find that many people have a say in what the product is called, from the originator to competitors, standards, customers, and more.

• During the discussions you may discover differences in how you define products and systems. For example, your company may produce different products which have different environmental performance but sell them as the same product commercially.    

• How do we start to become Accessible, Transparent and Secure?
  
  1. Ask the participants to suggest ways in which information could be shared within and outside your organisation in a secure way. How do they do it now? How do they think it could be done better?
  
  
• One of the barriers to sharing information is concern about security. By bringing up this concern and asking for ideas you’ll be able to identify ways to improve information security that management feel they can get on board with.

• We’re not talking about increasing transparency by making your information less secure. Rather we are talking about making it possible for those who need to access information, to be able to do it in a secure way.

• Does your security team know where your critical data resides and who has access to it? The key steps in protecting sensitive data include automating visibility, controlling access policies and implementing ongoing monitoring to discover vulnerabilities and risks before they become breaches.?

• Expect a wide variety of answers and share your own. Do you share documents via an email link rather than an attachment? Could you set up levels of authorization? Could you use password protected data viewers? Concentrate on the practical and how it relates to your current working practices. Meanwhile remember to ask for how to become more transparent too – could you combine databases? How could you improve accessibility and promote it?

• What will happen to our business if we don’t change? What are the opportunities?
  
  1. The final workshop exercise is to ask the participants two questions, “What will happen to our business if we make no change to how we create, use, store and structure our data?” and “What are the opportunities if we organise our data better?”
  
  
• Split your whiteboard or flipchart into two areas – Risks and Opportunities - and ask for contributions to each.

• Allow the participants to put their thoughts in their own words. You may be surprised by the list of problems that arise. Some may be specific to your business; some may be more general. All will provide motivation to take the next step.

At the end of the session collate the ideas from all four of the above steps into a short document and share it with the attendees.

Step 3: Employee Buy in - Share with your employees

Task the second level managers who participated in the kickoff workshop to share the outcomes of the workshop with their staff using the document you have produced. This has two purposes:

1. To instill a sense of urgency as well as the sense that something will be done;

• To generate additional information from employees which may have been missed during the workshop, such as a database you forgot you had.

These first three steps in the process create an understanding across your business of the nature of your product data management problem. At this point you should be ready to set up a strategy group to take the work forward.

Step 4: Strategy Group – Conceptual and Logical Data Model

Membership of the Strategy Group should include representatives from as many as possible of the departments who are responsible for product information management. Find out where people go for answers, and make sure that they are in the strategy group.


This could include IT, R&D, Plan, Sales, Marketing, Transportation, Finance, Operation Managers etc


The strategy group will require resource to support its activities, which will include analysing your current data sources and developing a solution.


The first task of the Strategy Group is to create a Conceptual Data Model for the business’ product data, and then create or commission a Logical Data Model.

What is a Data Model?

A Data Model is a diagram or other form of representation that organises elements of data and standardises how they relate to one another and to the properties of real-world entities. There are three main types of data model you need to consider: Conceptual, Logical and Physical.

• A Conceptual Data Model is a high-level description of the information needs of a business, focusing on the main concepts and the key relationships between them. Because the conceptual data model describes the semantics of an organisation without reference to technology, it can be used to discuss the initial requirements of the business with stakeholders.

• A Logical Data Model represents the abstract structure of the information. Based on the conceptual data model, the logical data model seeks to capture things of importance to an organization and how they relate to one another, in the form of data structures such as relational tables and columns, for example. It sets out the processing flow and the relationships between the data.

• Once validated and approved, the logical data model can become the basis of a Physical Data Model and form the design of a database.

• NB An Enterprise Data Model is an integrated view of the data produced and consumed across an entire organisation.

Relationship between Enterprise, logical and conceptual data model

Consumers and Stakeholders

When pulling together a conceptual data model, make sure you identify all the Consumers of your data. These can be Internal or External consumers:



You should also identify all the stakeholders of your data – the internal actors responsible for or have an interest in the data or its use. Map your stakeholders first and understand what their needs and responsibilities are and how they fit into the product lifecycle.


This process of stakeholder mapping will allow you to develop a governance process for your data and develop your operational procedures.


The image demonstrates a simple exercise a manufacturer can carry out in order to map products use cases with stakeholders who use them and different product lifecycle phases/stages of use


Different people have different ideas about the data that they need; because they have a single perspective based upon their own needs. You may find yourself asking, why do we need all this stuff? If it is information that someone in the company needs or uses, it is likely to be necessary. It is important to encourage your strategy group to look at the data more broadly from a company perspective. Make a conscious decision on what is useful or could be useful for the future and focus on the key properties. When you start mapping your data sources you can build on it and add more detail.

Step 5: Data Location Analysis

Once stakeholders have been mapped, the next task of the data strategy group is to locate and interrogate all sources of product data in the company, rationalize these sources and identify and resolve issues.


This process is the first step to producing a logical data model. You can produce this in house, or you may need assistance from a data specialist.


Here is an example procedure: 

1. Map the essential properties within each source of data; 

• Identify the most reliable sources of information;

• Delete common or duplicated elements to rationalise your sources;

• Develop a set of fields (attributes) required for each product according to need;

• Feed the data from the most reliable source of information into these fields (this could be a massive Excel spreadsheet or simple database which allows analysis)

• Use the collated data to identify the issues with the information, and then solve those issues within the data sources themselves.

In the process, remember to identify and focus on every need for data across the company, and be influenced by the users’ search terms when deciding on names. What would a user search for?


Doing this exercise helps you identify areas of waste, duplication and error. You may identify data processes that have been independently developed within the company but are not in fact used elsewhere in the company. You may discover products that no-one is looking after or pricing structures that haven’t been updated for years.

Step 6: Solution Development

Now that the strategy group understands the nature of the information you need and has rationalized where it is held, the next step is to decide how it is connected internally.


The principle is to aim for a Single Source of Truth, but this may not be possible if you have existing internal systems which cannot be changed. If that is the case, make sure that they can be adapted to feed into a SSOT, or between them, provide one.


There are a number of ways to create linked data. One is a Product Information Management system (PIM). This is a useful solution if

• The company has the IT expertise to develop it or can use external providers;

• You need a single system which legacy systems feed into;

• You plan to draw information from the PIM into other systems such as e-commerce, or

• The company is geographically spread and needs a cloud-based solution.

The differences between Master Data Management, product Lifecycle Management, Product Information Management and Digital Asset Management.


You may need to create a Data Administrator role to manage the new data system and identify new requirements.


Where a PIM system is not appropriate, there are alternatives, for example:

• Creating a data hub view within your ERP system and making the data interoperable, or

• For a small company with a few products, a cloud based relational database with templates.

 

Get your Data in Order

We have described the essential steps to set you on the right path to digitising your product data. These are only examples and suggestions; with a diverse industry, there is no ‘one size fits all’ for data management. Each company is different with different structures and products and each company will find a solution that works for their needs.


What we have described is Digitisation: the process of moving from an analogue system to a digital one. Once digitisation has been achieved, you can make the digitised information work for you, for example by implementing ecommerce, setting up a quality management system which included data management operating procedures, rationalize your naming conventions for new products. Making digitised information work for you is called Digitalisation and is the first step to a Digital Transformation – where you create new business applications for your data - from predictive maintenance to remote monitoring, or complete circularity for your products.

Share Your Views

We are sharing a number of articles investigating how construction product manufacturers can solve the problem of product data management. This is part of the process for developing a Plain Language Guide to Product Data, written specifically for the CEOs of manufacturing companies. If you’d like to know more about this project, please subscribe to this blog using the link below and we’ll notify you as new items are published.


In the meantime, we want to encourage as much debate about the challenges as possible.

1. Please comment below with your views and share this article with the #ManufacturersPLG hashtag.

• You can also join the conversation on the 'ask the community' IET forum channel: Plain Language Guide to Product Data for Manufacturers

• We’re also hosting an Open Zoom Call next Friday to discuss this article. If you’d like to join us, please Register to join us on Friday 20 November at 11am-12 noon GMT. The Zoom Call will be recorded and made available afterwards.

We look forward to hearing your views.


*Information is refined, human readable data. Information is data that was processed so a human can read, understand, and use it. 


Watch the call about this article below: