Are you relying on poor data? What are “red flags” of potential underlying data quality problems? And...how should you respond?
Declining data health can feel overwhelming. Plus, you don't know what to look out for, much less how to fix and avoid common data quality issues going forward.
The following are telltale signs that your organization might have a data dilemma.
The good news is that each of these warning flags is treatable and preventable. You can even set up an ongoing data quality program to identify, monitor, and avoid issues with data integrity, efficiency, and profitability down the road.
Here are key symptoms of data quality problems and what to do about each:
Large piles of returned mail are a visible red flag that not all members received all of the valuable member benefits mailed to them. If your association’s industry has a high turnover rate with members changing jobs every few years, updating those member records is a tedious but necessary task to be performed regularly.
Today, this service may be provided electronically. The returned mail may be just as prevalent, but much less visible. If you still mail a physical magazine to your members, ask your membership director about the process for handling returned mail and updating those members' records.
Most member benefits are also delivered electronically these days, so the email address is the single most important data point. There are several key questions you should also ask about reaching your members via email:
You may need to combine those numbers to get the total picture of your reach with email communications to members.
Having accurate and up-to-date contact information for your key constituents is the most important data issue to monitor. If you send a physical magazine or other mail, then the mailing address is still important.
On the other hand, if you only deliver benefits electronically, the focus can shift to maintaining an accurate email address and mobile phone number.
Our partners at Info-Tech state, “Data quality means tolerance, not perfection.” This quote seems so relevant when talking about our AMS systems because duplicate records are nearly impossible to avoid. For example, you may find multiple records for a company where each record represents a different physical location.
So while the data is accurate, it is hard to know which record contains the key employees. Sponsorship orders for the same corporate entity may be spread across different location records making it difficult to see historical information. Other times, you have multiple records for individual contacts. It is difficult to determine at a glance which record is most up to date.
There are three common scenarios when duplicate records are created.
It feels like a full-time job to continuously monitor and merge duplicate contacts. Your membership staff is too busy with day-to-day processing to make a dent in the number of dupes in the system.
If you struggle with duplicate records in your database, a thoughtful approach can quickly tackle the issue and an ongoing program can be adopted to keep things in check.
So while you may not be able to keep your contact list free of duplicate records, you can still enjoy high-quality data with a disciplined and consistent approach.
If you run a report to see how many registrants are expected for an upcoming meeting, you expect to see the same result across multiple report formats. It is hard to understand why you see different results. Sometimes it seems impossible to get a simple answer to a simple question.
This happens because reports are generated over time for different audiences who are measuring progress on different goals. Let’s look at an event registrant count as an example:
All staff must be working from the same definition of who counts as an event registrant, then all reports need to be consistent with that definition.
Data quality without data governance is equivalent to treating the symptoms but not curing the disease. If data quality is not embedded in the association’s data governance framework, then data quality management will remain a band-aid fix.
Organizations that outline clear definitions for their common business vocabulary and adopt those criteria into the report generation process enjoy clear and consistent answers to commonly asked questions.
Event rosters, name badge reports, committee rosters, and membership mailing lists are all examples of operational reports that are needed to complete specific business processes, yet decision-makers need information presented differently.
They are looking for summary level reports that show member counts by segments, sales levels by product type, or a list of the top engaged members that could be tapped for leadership opportunities.
Data validity issues can be magnified when you display the data in a summary report or data visualization. Typos, misspellings, and data typed into the wrong field are common reasons these validity issues occur. They can be addressed at the data architecture level by using drop-down fields to capture information instead of open text fields.
Data should be at the foundation of your association’s evolution. The transformational insights that executives are constantly seeking can be uncovered with a data quality practice that makes high-quality, trustworthy information readily available to the business users who need it.
In some organizations, not everyone has access to view dashboards or run reports to view basic summary data. Their procedure might be to send an email to key staff in the IT department and wait days to get a response.
These bottlenecks may also be masking the data validation issues mentioned above. By funneling requests through the IT staff, the information can be manually cleaned up or consolidated in a way to hide an underlying data issue.
This delay in access to high-quality data limits the information-gathering process and the ability to use data to make informed decisions.
Insights, knowledge, and information are needed to inform operational, tactical, and strategic decision-making processes. High-quality data and reports need to be accessible to all staff via self-service tools that do not require technical skills to use.
Sending every marketing piece to every member or your entire database is still all too common. If you can't effectively segment your contact lists, then you can't tailor and personalize content to the audience.
When you peek under the hood of AMS systems, you'll often find missing or archaic information in demographics and interest fields.
A lack of a coordinated approach to data collection is to blame. Event registration forms collect tons of useful data that often cannot be easily imported into the AMS for a variety of reasons. These issues may be straightforward to solve but take planning and coordination across several departments.
You can outline when and where you hope to collect key demographic data by mapping a member’s journey from recruitment through the membership onboarding process. Information gleaned about members' interests can be used to segment lists for the promotion of products and to feature the most relevant resources in the delivery of member touchpoints.
How do you fix problems with data quality? Where do you even begin?
Focus efforts where it matters most - at the point of entry. Errors here can cause a domino effect when you share data with other systems.
In other words, if you fix data ingestion, you'll fix the large majority of data quality problems. This requires you to either improve your application and database design or improve your data ingestion procedures.
Cimatri can help you to identify and target the root causes of poor data quality. Prevention is 10x cheaper than remediation. Take our data quality assessment to get a plan in place today.