Your B2B data should be an asset on your balance sheet, not a liability. Making your data invaluable to your organization requires the right approach. To start with, improving the quality of your data is not a one-time event. With that approach, you’ll be right back where you started in a year or two.
Four Critical Success Factors to Building Lasting Improvements in B2B Data Quality
To create sustained, material improvements in your B2B data, there are four critical steps. While there are many considerations, these four factors matter the most when building your data quality road map:
- Select the right people
- Set appropriate expectations
- Get the right project leader
- Define and documenting a path forward
Select the Right People
You need to get the most senior leaders to participate from departments or groups that have the greatest impact on the data, operationally. These are the leaders who will set the policies and decide on the business rules for their respective teams. Without executive sponsorship at the most senior levels, the chances for sustained success are greatly diminished. You need the political weight of this group, and you need them to establish policies for their groups.
The specific groups will vary from company to company but usually include the leaders from these departments:
- Post sales
Ideally, this team includes the CEO. Nothing will make everyone take this initiative more seriously.
Set the Appropriate Expectations
These initiatives can get derailed quickly by squabbles and finger pointing. To avoid such scenarios, set proper expectations:
- Making the data better is hard. This point may seem obvious, but many business leaders underestimate the challenge.
- Perfect data is not the goal. No matter how stellar and committed the team and any vendors are, your data will never be perfect. Rather, the goal should be making the data more useful, and the challenge for the team is to figure out how much better the data can be within the economics of the business. Put another way, at what point will further data quality resources reach a point of diminishing returns? Answering this question is the goal.
- Focus on the realistic desired future state of the data. Looking at the past and assigning blame is tempting. It’s also counterproductive. Instead, focus on what will happen moving forward.
- Commit to meeting thirty minutes to one hour each week. Executives are busy. Most of them will not enjoy looking at or thinking about the data or the policies to improve data quality. Still, they must attend and come prepared. The CEO needs to emphasize this point. The only exception to these short meetings is the initial meeting to kick off the project. That meeting will typically take a couple of hours.
Select the Right Project Leader and Give that Person the Right Charter
In addition to the executives, you’ll want to find someone to lead the initiative. The right person in this role will make or break the project and will do a great deal of the work, which will make participation in the project far more palatable for most of the executives.
The typical responsibilities of this individual are as follows:
- Establishes the agenda for each meeting, including preparing everyone for any to-do items after each meeting;
- Guides key micro-projects within the overall initiative, like helping to define and document the appropriate sample data-sets needed at different stages of the project;
- Arbitrates disagreements to find an acceptable compromise;
- Brings an objective, informed perspective to the internal and external discussions;
- Documents requirements and business rules, researches solutions, and otherwise carries most of the work load.
Temperament and experience in this type of initiative are critical. You want someone whom senior leaders will grow to respect because of their in-depth knowledge of data quality best practices and the data vendor landscape, and their ability to navigate enterprise organizational behavior, both at the executive level and with staff. For these reasons, finding an external resource is often the best path. In addition to the experience such a person brings to the project, they also will not carry any internal baggage.
Define and Document a Path Forward
Start by sharing a sample of the current data. This team needs to understand the current state of the existing data. What are the problems? How pervasive are the problems? This reality check will give the team the proper context for making three big decisions about data moving forward:
- Agreeing on definitions. The team must agree to several definitions so that the company has a common language and the data reflects that understanding. An example is the seemingly simple question of what a customer is. Is it a person or an organization? Does it matter how long it has been since the person or company purchased something? How do you want to identify those contacts within an account who were involved in past purchases and those who were not? Working through and documenting these definitions is the first job.
- Data standards. Data gets messy due to a lack of standards. For example, a close look at the data may reveal company names in the address field, phone numbers without area or country codes, different abbreviations for states/provinces, etc. Mobile and direct dial phone numbers may show up in the company phone number field. Numbers such as total employees or annual revenue are listed as an actual number, an abbreviation, or as a range. This issue really comes into focus when you consider what the postal system will allow. For example, the United States Postal Service allows up to eight lines for a business address. The first four can be in any sequence and will typically consist of the name of the person, their title, the name of the company, and (if applicable) a division name. That’s 16 possible combinations. Few internal systems can handle that type of complexity.
- Enforcement policies. Given the data standards and the definitions, what will the team do about enforcement? What are the policies? What happens in the case of non-compliance? How will the company inspect what it expects?
In this process, the team needs to address any areas of the business that are likely to be problematic. A common example occurs in sales. The best salespeople are usually not interested in becoming conscientious data entry clerks. So, what sales people enter into the CRM system is often incomplete and not in compliance.
Another example is marketing-generated leads. Marketing often wants to keep data collection in lead forms as minimal as possible, sometimes capturing just a first and last name and an email address. Marketing wants to do so because there is a direct correlation between higher lead capture rates and lower data collection requirements. However, in the process, they open the floodgates to data duplication.
Finally, you’ll want to document the definitions and policies, a task that should fall to the person facilitating the exercise.
Moving Down the Path to Quality Data
Using the framework we’ve shared makes it possible to achieve rapid improvements or to take a phased approach to achieving better data quality. Developing your data quality requirements and addressing the four factors outlined above are the first steps in the journey. The next step is operationalizing your new approach.
Remember: data quality initiatives are not overnight or one-time projects. They are a strategic investment in your company’s future health and competitiveness. With that in mind, it’s well worth your while to take a proven, methodical approach that paves the way for strong returns.
Download our 36-page eBook and gain insights, business case benchmarks, best practices and a proven roadmap for achieving better B2B data quality.