IDMP Compliance: Key Challenges
I was interviewed by Marcus Evans ahead of their 3rd Edition IDMP Compliance Challenge: Hit the phased approach deadlines conference held in Frankfurt last week. Below is the transcript that was originally published on the conference website.
What are the key challenges companies are facing today in meeting the compliance requirements?
I think one challenge I am hearing is the lack of certainty. So at a time when there was a fixed IDMP deadline date, although it was challenging, everyone was applying the focus and attention needed to get the project going forward. Now that the goals are shifting, there is a lot of uncertainty. A lot of the teams that were put in place in order to achieve IDMP compliance are now struggling to gain the attention they need to keep the resources and the funding going because of this uncertainty. In my opinion, at the moment, there are a lot of projects on hold which run the risk of being in the same situation in a year’s time: leaving it until less than 12 months to go before the deadline, and compliance projects in the early stage. So that lack of certainty is pulling the focus away, despite the fact that the actual original challenges have not gone away.
One of the key challenges is the broad scope of data and how this is distributed over an organisation. Indeed, there is more time now to pull that data together in the different phases, but you still have to get the data together, source it and identify where it is held in the organisation. The questions rising from this are: “Who can help you understand that data? Will they have incentives to help you understand the data and make time in their day? Can you actually merge data from different sources to build the hierarchies, cleanse the data, and manage the whole cycle so it is not a huge manual burden?”
So in summary, the uncertainty is taking away the focus that is needed to achieve IDMP compliance.
What should be the key focus for companies today for IDMP compliance preparations?
I believe that you need to use this extra time to actually put in something more strategic. Many project teams are looking into practical solutions, what they can get in today or in 12-18 months time, as long as in 12 months time it actually works. Putting your time into something more strategic means that you can design something that is not for a particular regulation in a particular geography, but is rather a means to collect a broad set of product data and manage it for multiple purposes. With the phased approach time is available to design and deliver a flexible, multiple purpose and complete source of product data. In order to do that though, you need to focus on value. If you can highlight the value of high quality data, you should be able to keep some of that focus on your IDMP project. Strategically, the focus should be on the availability of high quality product data which is useful throughout the organisation, with IDMP compliance being a particular deliverable.
How to best deliver business value from data management beyond compliance?
This depends on the organisation. If you go back to the basics, what does it take to be IDMP compliant? It means that you have to be in control of, or have the ability to manage a broad set of product data within your organisation. Data needs to be identified, collected, put in a specific format for everyone to understand, cleansed to get rid of all the data anomalies and discrepancies, and create all the references on the completion. It needs to be done in a way that every time you change the data you can actually understand what, where, and how the data was changed. Either it was changed by a person or a system.
So if you want value from high quality data, you need to look into the organisation to find out who else uses that data. Beyond IDMP, there is probably not going be a single use case for the complete dataset, but there will be subsets of that data that are used elsewhere in the organisation. Informatica has found that in order to understand the value of high quality data, you must talk to the business people. Once you understand departmental goals, what data they need to achieve these goals and where their inefficiencies lie, as well as frustrations on missed opportunities, you can build a more complete picture of the value of data management beyond compliance.
What we generally see if you have poor quality data, it results in inefficiency. For example, in many IT projects, one of the first things that needs to be done is gather product information. This can add months to individual projects, despite being a task repeated across many projects. The delay in gathering data could create a knock on effect that the IT project is not going to deliver value as fast as it could, slowing down innovation and new business opportunities in your organisation.
Examples of how high quality data have delivered value to the business include:
- Increasing efficiency of all employees, as they spend less time looking for data.
- Improved marketing messaging and more informed sales people leading to higher conversion rates and ultimately increasing revenue.
- More informed and faster product substitutions, where appropriate.
- Faster product launches through rapid data distribution internally and with trading partners.
- Reduction of working capital through the ability to better manage inventory levels.
- Improved product data readily available for all relevant websites to improve patient relationships
In summary, the actual value of high quality data will differ by pharmaceutical company because each pharmaceutical company has their own strengths and weaknesses in terms of data availability. At Informatica we have repeatedly seen that every company can benefit from improving data quality. Existing individual capabilities will determine where individual pharmaceutical companies can benefit beyond IDMP compliance.
What is the role of MDM and the MDM department in the IDMP project?
I have spoken to a number of pharmaceutical companies with very strong MDM programmes and strong belief in the value well managed master data delivers. These companies with established MDM departments and capabilities are typically finding that the MDM people are great advisors as MDM and IDMP have similar goals and capability requirements.
In order to achieve Master Data Management there is a need to gather data from a number of data sources for Customer Data or Supplier Data or any of the Master Data domains. Once master data sources are identified, the data is centralised before performing data cleansing and distribution tasks: match, merge, cleanse, complete it and then distribute it out to a number of different target systems and keep a history of that information. Similarly IDMP requires identifying, sourcing, collecting, cleansing and distribution of a broad set of product data. So, at a minimum, the IDMP teams could learn the best practices from the MDM teams. Specifically the IDMP teams can understand how to be successful in their organisations.
If the MDM teams are willing and able, they could be active participants in IDMP compliance. These teams should already have established master data governance programs and procedures. They will also have some MDM technology you can leverage, if it has the capabilities to support the product domain and the complex IDMP data model.
If the MDM team already has product data under management, their systems could also act as a source for at least a portion of the data required for the IDMP scope.
What is your advice for more time and cost effective IDMP project?
For the short term, you have to maximise your budget and that means more automation. For the long term, you need to look at a cost effective, global IDMP programme, which means that you need to design the first phase with further phases in mind. In this respect, further phases should look beyond Europe, and to how IDMP will be implemented globally by other regulators.
First consider the short term and how to reduce cost and increase efficiency of tasks through automation. Re-use of code is also crucial. Examples of re-use include scripts to extract and transform data, as well as data cleansing tasks.
Cleansing of data and applying controlled vocabularies are ripe for automation through data quality tools. Efficiencies will be increased if people who understand the data (i.e. the business people, not the IT folks) can easily contribute to the creation and management of data quality rules. Cost savings in the short and long term, and efficiency increases are going to come from modular automation using appropriate data management technology. Hard coding of data sourcing, extraction and cleansing routines will drive up costs, and reduce flexibility.
For the long term considerations, I would say that you have to plan for change and plan for scale. Change is inevitable, as IDMP is a global standard that is here to stay. This change can come in many formats:
- Different regional interpretations of the ISO standards
- Change in existing guidelines, or of the standards themselves
- Merger and acquisition activity, which is particularly high in the pharmaceutical sector
- New product classes being developed, requiring different sections of IDMP data model to be implemented
Anything you re-work in later phases is going to be expensive. Questions to ask include: Is it easy to adapt the data model for changes in the standard, or alternate regional implementation guidelines? How much are you planning to hard code? Can different regions easily leverage your underlying technology and business processes? If not, you are looking at additional costs down the line.
My advice is to use the extra time the EU has given all of us to plan for a strategic, global IDMP programme. In my opinion this implies planning and delivering product data management capabilities that can easily adapt to changing requirements, and scale to support the entire organisation. If designed and implemented according to best practice for data management, the solution will not only reduce the cost of IDMP compliance, but provide value throughout the organisation by making high quality product data available to multiple business processes.