By Col Rae, Services Delivery Manager (Content), Oniqua

Oniqua Content Services is a line of business within Oniqua Pty Ltd which specialises in master data maintenance and enterprise analytics for capital intensive industries worldwide. Oniqua are the custodians and owners of what was previously the Auslang standard. This is now the Oniqua Standards Dictionary or OSD. In addition this business line handles outsourcing of analytics including inventory optimization, spend analysis, supplier performance and failure analysis

The fundamentals of getting the project planning and governance around a master data project is the basis of this paper. The rules by which we do this and the project governance are areas where you can control the output, pace and quality assurance of the data. It is essential that these control measures are in place by you, the client, to set the expectations for the vendor. Listed above are the fundamental areas to a successful master data project and we will be going through these in detail.

Most organizations at some point will question the validity and value of their master data.  The danger is that it will been seen as nothing more than a ’must do’ cleanup. Good master data can and will return numerous benefits to your organization. If you are going to spend the money, make it an investment in information, knowledge and sustainability not a band-aid to a symptom of a larger underlying problem. Be sure that you select a partner to assist and guide you through the many pitfalls, providing value at every step and ensuring you get the maximum potential from the finished product without being too rigid or dictatorial about it. Standards applied need to be flexible enough to fit with your corporate guidelines and your ERP framework. You need to check the credentials of potential vendors to ensure they have a sound understanding of your ERP but also bring to the table experience gained from other ERP systems. Your vendor needs to understand how you utilize your ERP and how to adapt to your requirements without detracting from a standard, partner with you for the long term and be in an approachable position on an ongoing basis.

What I present to you is the key points to ensuring that the problem mentioned previously is treated with the attention it deserves. This is about ensuring you have a successful and sustainable solution. In order to address the problem you must first understand it, break it down and deal with the elements. We already know that the problem is the quality of the data, but have you recognised that the problem will not go away unless it is dealt with effectively. To do this you must understand your vendor and the processes they adopt in rectifying your data issues. The last thing you need is an unwieldy project that is invisible to you and over which you have little to no influence.

The presumption is that you have analysed the issues and determined that data quality is the cause. Do you have a clear picture of all the things you want the cleansed data to deliver to you, for example, do you have enough attributes to describe the item and therefore identify it, do the descriptive patterns contain sufficient information for purchasing or does the purchasing text need to be enhanced? Can you perform effective spend analysis based on category or commodity? This information ensures that you have provided sufficient information and are able to develop a specification for the cleansed data layout. Is this an inventory catalog, vendor listing, BOM, Equipment file or all of the above?

With the changes to the data structure, change management becomes an issue. As the revised data permeates through the organisation questions will arise as individuals discover that things are not as they were. How does this affect end users, purchasing officers, contract managers, inventory controllers and vendors? Will it affect reporting as classifications change? How is this going to be handled?

When spending money on a data project, what are you going to get in return? Have the invited vendors provided you with an ROI statement and are they in a position to assist you past the point of data delivery? In order to deliver an ROI statement the vendor needs to ask questions relating to the data and the issues uncovered leading to a master data project. The vendor needs to understand the issues and the potential cost ramifications of not fixing these issues.

You will have to commit resources to the project. There will be questions, guidance will be required, clarification sought, training undertaken and checks done. The levels of interaction need to be determined and names put in place. Priority needs to be placed on tasks to ensure they are done on time so as to ensure the project is done on time and on budget. Visibility of the project to key stakeholders is extremely important if momentum is to be maintained.

Be sure that your provider is willing and able to provide support both during and after the project. It maybe that we require clarification on the standard you have in place when something new is introduced. Can you call them up and ask them, can you put a face to the name and have you dealt them already?

After you have received your data you should review and audit what you have received against the original charter and specification to ensure it meets the expectations. Because data tends to be looked at as an annoyance rather an asset it tends to overlooked for pre-checks before going into production. What other asset worth possibly hundreds of thousands of dollars would you put into production without first checking to see if it is what your ordered and that is capable of doing the job intended. During the load and validation phase, preferably testing, it is advisable to have a vendor representative present to advise, attend to issues and clarify points of concern. This will go a long way to successful load on the due date therefore alleviating costing delays. Support should be visible to you via a help or support desk at company level not only at the project level to ensure it is still there after the project is completed and that tools provided to assist the upkeep are maintained on a regular basis and supported by documented processes.

A detailed project charter must be developed post sales and before work commences. This provides the opportunity for the operational arms of the organisations to meet and document precisely the expectations of both. This will clearly define the scope and the responsibilities. In contrast the specifications workshop deals with what the data will look like on completion and what format it will be delivered in. This is essential in ensuring that you get what you need and expect. The delivery schedule may be separate or part of the project plan, either way there needs to be one. You should find out as much as possible about the vendors QA processes and guarantees as well as notifying them of your expectations. A risk and issues register needs to be in place and reviewed on a regular basis with results documented in periodical reporting. No project should be closed out until the register is clear. Project completion should take place in the form of a face to face meeting to cover of on QA, issues resolution, handover of documentation and finalisation of commercial transactions. This needs to be a definitive close off of the project and the vendor needs to explain the support process in place post project. Master data projects typically fail because they do not have high level support within the organization and change management is not in place. At the time you go out for quotation ensure management know this work needs to happen. Benefits realization is dependent on internal changes that need to take place to support the outcomes. Resources must be focused on the timely delivery of data and information to the vendor to ensure the project remains on track. Yours is probably not the only project the vendor is handling and scheduling may not be all that flexible. Project governance must be a formal process. Your project manager must detail any internal processes of governance at the project start up meeting. Training is an essential element in realizing the benefits of your project. It is quite common for vendors to see the training scaled down or even cancelled with dire consequences as a result of ineffective change management practices where the value is not seen or the project not supported. Again, all of this needs to be discussed, worked out and planned for at the beginning of the project. What information is the vendor providing you with on a regular basis so you know the status of your project? Good vendors will have as part of their project management reporting processes to keep you up to date. Stay in close contact with the vendors project manager and know who to escalate problems to. Notify the vendor project manager of any changes that may impact on the project well before they happen and work with them to overcome or at least limit the fall-out. Commercial considerations impact on projects. Be aware of invoicing processes and schedules and note the payment terms as non-payment of invoices could slow or halt the project. Discuss any discrepancies with the project manager as soon as they arise. Many projects turn up tasks that were not identified in the original scope of work. Ensure that the proposal given to you covers this and that the prices are not significantly higher than those of the project tasks.

Automation tools are used to undertake tasks such as naming and to look for inconsistencies and errors in the data. Automation tools can use word recognition and differing forms of artificial intelligence along with manufacturers part number databases to consistently name items. The application of a standard is used to ensure a repeatable pattern of data structure and consistent attributes. Quality Assurance auditing relies on automation tolls and human interaction to ensure the best possible outcome.

Labor costs and an extremely competitive marketplace dictates that off-shore resources need to used. How vendors manage this process is paramount to a successful project. The project is managed by the project managers with escalation points to management. Work is scheduled by the workflow coordinator and allocated to either local, off-shore or technical resources. Work is audited for quality during and after each phase. The whole process, timeframe and detail is specified in the overarching documentation. Our project managers and quality assurance officers are our full time Australian based resources as are our lead workflow coordinators and QA officers. These people control the processes. We operate a captive team and have standing preferential contracts in place with several other teams to supply resources in peak periods. We control the processes only the project management interacts with the client.

In terms of communications you should receive from the vendor:

•  Status reports on a regular basis keeping you up to date with progress and outstanding issues.
• Quality assurance reports advising you that the levels of accuracy required are being met.

You in turn should be supplying the vendor with:

• Answers to queries and clarification and rectification actions on issues
• Notice of changes to points of contact
• Any significant issues arising that may impact on the project
• Resolution to any commercial issues

The current status of the change management processes in place

Our weekly reporting keeps you up to date via a simple dashboard style report. This allow you and your management to quickly ascertain the status of your project. These reports are supported by a myriad of information and is available where applicable to the client.

You should consider early how or even if you can sustain the process of maintaining good clean data after your project. Do you have the resources who are focussed, are capable and have the time to do this? Do you have processes in place to able to effectively manage your data? Do you have a contingency plan in place if you lose your key resources either temporarily or permanently and can you ramp up during peak times? These are questions which usually do not arise as part of a project but should be considered as they provide the basis for sustainability. In addition tools specifically designed to address the issue should be considered over tools which overarch multiple like issues without specifically addressing the requirements in detail. You should look to see that you have entrenched processes preferably through workflow to ensure a consistent approach to your data maintenance.

Sustainability of data quality should not be taken lightly as it is quite often overlooked and results in a never ending succession of cleansing projects reducing the time available to analyse the data and make full use of the intelligence that can be gained from it. Do not end the process by having clean data, ensure you follow it up by extracting information from the data and putting it to good use. As in the case of data cleansing ensure you find the right vendor to assist you in this task, your return on investment will be more than worthwhile.

To ensure that you can sustain the improvements in your data you should be looking at vendors who have been around for some time and recognised as leaders in their field. They should be able to demonstrate multiple facets to their operations and a record in differing sizes and complexities of projects. The vendor should provide some background information on the experience and role of the project members along with how long they have been with the organisations role. The vendor should be placed to provide tools to assist you in maintaining the data yourself thus lowering costs. You need to be able to see a history of these tools being maintained and supported in a satisfactory manner. As for reference companies and contacts. Have you considered the standard which you intent to apply? Is it easily decipherable and can you identify the origins and experts in its application? Does the vendor supply the standard? Can their standard be cross-referenced to other standards if required? Is the vendor capable of supplying resources to you during peak periods or as a regular offering? Can the vendor assist you in getting more value out of your data through analysis, interpretation and strategic advice? What support infrastructure and processes are in place for your organization to be able to undertake recourse action if issues arise post project?

We have many resources to assist both you and us to successfully improve and maintain your data. Some of these are used internally by us, others are shared and some are designed for your use. The ones designed for your use are specific about their purpose and matched to suit various problems, scenarios and data sources.

1. Plan for a known outcome via a proven process
2. Ensure you can get the maximum benefit and measurable ROI
3. Ensure the results achieved are sustainable and realistic