Data quality is paramount to the success of every business, no matter their industry or offering. From manufacturing to healthcare to banking, data is what allows any business to stay on top of their operations and make sound business decisions. Information that is accurate, complete, and standardized is critical for every business operation, from order entry to delivery to billing. Given its importance, it is surprising how often organizations do not prioritize data quality at the enterprise level.
So, if you are responsible for managing your organization’s data or your position relies on data to be effective, how do you convince your organization to invest the time, resources, and finances necessary to implement a data quality solution? How can you help to ensure that an argument for data quality is justified and approved by senior leadership? While every organization has their own unique concerns and hurdles, we have identified 4 key components that are a part of almost every successful data quality implementation we have been involved with over the past 50+ years.
1. Identify the Problem
The first step in solving a problem is admitting that you have one. Senior leadership may not be aware of the scale and scope of your data quality problems, so begin by compiling a list of all the ways that information is being compromised across your organization. Remember, each element may seem small on its own, but across multiple systems and hundreds of thousands of transactions, they quickly multiply and can negatively affect every aspect of your organization.
Human Errors – Transposed digits, typographical errors, misspellings, and simple mistakes in data entry are common, especially when data is entered by customers, patients, vendors, and others unfamiliar with the system or not beholden to the organization’s data entry rules.
Field Errors – Data can easily be entered into the wrong form field accidentally, but many systems also fall victim to deliberate field errors. For example, when a new piece of data is collected and no field exists in the database, users often enter it in whatever field is available.
Formatting Errors – Poorly structured forms can allow for data to be entered in the wrong format, such as too many digits for a phone number or email addresses without an @ or url.
Style Errors – Data can come from a variety of users, sources, and systems, and not everyone uses the same grammatical style. This results in things like multiple variations in honorifics (MR, Mr, Mr., Mister) and abbreviations (Street, ST, St.).
Extra Data – Well-intentioned users often include extra data in fields in an attempt to be helpful. Common ones include entering “do not call” in the phone field or “deceased” in the last name field. There are better ways to capture this information!
Data Omissions – As data flows through systems and departments, some systems expect to find data in certain fields. The omission of data can lead to incorrect default data or a system assuming that no entry equals “zero”, for example.
Migration Errors – Importing or merging volumes of data can contribute to poor data quality if fields are not properly matched and linked – finding all Last Names in the City field for example, or Social Security numbers where phone numbers should be.
2. Gather the Stakeholders
Poor quality data affects everyone within an organization, so as you are gathering information about the scope of the problem, pay attention to who else within the organization will benefit from improved data quality and get them on your side. Find out what business processes are most important to a business unit or business user and ask them to consider what would happen if the data they use were to be unavailable, inaccurate, or out of date. Consider who has the largest risk if that data is jeopardized.
Accounting & Finance – Are they mailing invoices to bad addresses? Are they wasting time finding the right billing contact person? Are they securing resources based on inaccurate data?
C-Suite – Do they understand the size and makeup of their customer base? Are they able to track trends in leads and opportunities? Are they exposing themselves to undue risk?
Human Resources – Are recruiting efforts proving ineffective because they aren’t reaching the right audience? Are they at risk of discrimination lawsuits from bad data?
Marketing – Are they wasting postage mailing to invalid addresses? Or sending multiple mailings to the same address? Are e-mail campaigns reaching their intended targets?
Shipping/Order Fulfillment –Are they wasting shipping costs on inaccurate addresses? Are returns being handled in a timely manner? Are shipments being returned for lack of paperwork?
Sales – Are they wasting time on invalid contacts? Are they forecasting based on inaccurate information? Do they understand the demographics of their customer base?
These are just a few examples, but your goal should be to develop a network of stakeholders who can help to support your data quality initiative. By understanding their needs and illustrating the benefits of data quality and the risks of poor data to them, they become aligned with your goal and will provide support to your initiative. When senior leadership hears multiple voices from across their organization with a unified message, it helps elevate and prioritize the initiative.
3. Make the Business Case
Achieving acceptable data quality requires investments in time, technology, and labor. Justifying a data quality initiative requires creating a strong business case, so your next priority must be an economic analysis, presented to leadership as a Return on Investment (ROI) calculation. You’ll need to “run the numbers” to determine the cost of your initiative versus the financial benefits that the initiative will deliver to your organization. Start with the “return” half of the ROI equation by taking into consideration three factors that affect positive returns:
1. Increased revenue
2. Decreased costs
3. Decreased risk
This is when the stakeholders on your team really come in to play. They can provide you with financial insights on how bad data is impacting their departments. Sales and marketing should be able to show how they could achieve increased revenue from more effective targeting and communications, while accounting and HR will show decreased costs through automation and a reduction in shipping/mailing errors. And don’t forget the costs associated with risk; better data reduces risk which reduces the chances of financial losses from fines, lawsuits, and lost revenue. This is an area the C-Suite will be especially interested in.
The other half of the ROI calculation is the investment you are going to make in a data quality solution. You have to take into account all costs associated, including the cost of the software platform you choose, installation, training, and any capital expenses involved. Obviously, the lower the cost and the higher the return, the better your ROI calculation will be. So when considering a data quality solution, it is important to consider the entire scope of your initiative. Some questions you should be asking are:
How prepared is my organization for this initiative? Will we need professional consulting to help identify our needs and find the best solution?
How familiar are our people with maintaining data hygiene? Will they need training to ensure widespread adoption of the solution?
What IT resources will be needed? Will we host the platform on-premise, or in the cloud? Will we manage the tool ourselves, or is an as-a-service option better suited for us?
How do we avoid scope creep? If improving bad name and address data nets us the biggest ROI, do we really need a comprehensive MDM platform?
What will our monitoring and reporting needs be? How can we track the success of the initiative and demonstrate the ROI?
What volumes of data are we processing and can our chosen solution handle them without significant lagging or crashing?
Will we need to customize the solution to fit our unique business processes, and is our tool and provider flexible enough to accommodate our workflows?
What security concerns and regulations are our data subject to, and how do we ensure that it remains protected?
By taking all of these considerations into account, you can better identify the right data quality solution to fit your needs. Ideally, you want a solution that can be implemented quickly, so that you can start showing ROI right away, and one that fits your needs without paying for features that you may not use.
4. Find the Funding
You’ve identified the scope of the problem, gathered a team of supporters, and laid out the business case for an investment in improving data quality. The last piece of the puzzle is to figure out how to cover the cost of the initiative. An investment in data quality tools and resources will require budgetary approval, so it’s best to be prepared to address this question before seeking executive approval.
Work with your stakeholder in the finance department to investigate possible funding sources and how they might be allocated to data quality improvement. Work with your other stakeholders, especially those most affected by poor data quality, to find department-level funding opportunities. Many department heads have budgetary authority within their departments that could contribute to an enterprise-wide solution. Your ultimate goal is to go to senior leadership with a well-thought-out argument for the need for data quality, supported by cross-functional stakeholders and the numbers to show the ROI and how the project can be funded.
With these four components in place, you’ll be as prepared as possible to present your case for improving data quality across your organization. By doing the research and crunching the numbers up front, you’ll be ready to overcome objections and answer questions, and you’ll have everything your senior leadership team needs to make a decision close at hand.
コメント