Many books, articles, papers and lectures are presented about the subject of direction and how to determine where to position your organization in the coming years. The purpose of this paper is to
give you the tools to assess what it will take to get your organization where it needs to be.
Dashboards are magnificent for determining what just happened or what is happening with your current data. Performance reporting dashboards are tactical dashboards. What is needed is a strategic
dashboard that gives an assessment of your organization’s preparedness.
This article presents a series of subjects to be measured to help determine if your organization can follow the course that it needs to take.
Data is the most important thing in today’s enterprises, and the following 10 subjects need to be assessed as applied to an organization’s data. All of an organization’s data should be
considered when going through these subjects, not just the data in the databases.
Assessing these 10 areas will tell you if you are ready for the changing needs of your enterprise. The areas are:
- Metadata
- Infrastructure
- Course
- Human Resources
- Survivability
- Accuracy
- Integrity
- Delivery
- Quality
- Security
1. Metadata: Metadata is defined as data about data.
Metadata provides context for data. Do all of your business units have a common definition of data terms to work from?
Common definitions or at least documented different definitions are the foundation for the subjects of data quality, data delivery and data accuracy.
Is there clear documentation about what technology system is supporting which business unit? How easy is it for new people coming on board, or even outside consultants, to learn this information?
2. Infrastructure: Data infrastructure is the underlying equipment that supports the needs of your environment.
The combined systems of all of the individual elements maintained by your IT staff comprise your core infrastructure.
Are you running on 10 Meg Ethernet? Do you have hardware from vendors that no longer even exist?
Are you using best practices for all of your activities? How is the application of best practices determined?
I have seen all of these scenarios. Each one of them presents its own unique challenges to the organization.
3. Course: Where is your organization going?
Some people refer to a road map for an enterprise. I prefer to think of the direction of our enterprises as more of a course as a ship on the unexplored ocean.
A road map is great for a short-term project. If you need to implement something that other people have already done, you need a road map.
An enterprise cannot follow a road map. An enterprise needs to have a course. This course is a reflection of the mission statement of the organization. Various road maps will need to be followed
for the various projects and tasks that need to be accomplished in support of keeping the enterprise on course.
4. Human Resources: The people that manage your data architecture.
How much time is spent on troubleshooting and support? How much on training (internal – data recovery drills, disaster drills, teaching juniors how to do a better job; external – vendor
training)?If the majority of your staff is simply keeping the ship afloat, who is watching for the icebergs ahead?
Do you have to call in extra people at the eleventh hour to save your systems?
Do you have adequate staff to have an on-call rotation?
Are the same people in on the weekends and nights consistently working on problems?
What are the credentials of your staff?
Does your organization actively support professional training and certification efforts?
5. Survivability: The ability of your data to survive and recover in the event of a catastrophe.
Survivability is the ability to remain alive or continue to exist. The term has more specific meaning in certain contexts.
- Backups: Backing up your data without periodically testing both your restore process as well as your restore media is simply foolish.
- Restores: Does disaster planning take into account that a restore from tape could take potentially more time than the original backup itself?
- Process: Is there a documented process for the foreseeable catastrophic failures possible within your data center?
- Testing and Drills: How often is this process tested and updated?
- Tools: Are you using the best tools, or are you simply using what came with your system out of the box?
6. Accuracy: Data accuracy is the difference in your data between what was meant and what was actually done.
Accuracy is the measurement for both quality and integrity.
Do your analysts have to manipulate data in spreadsheets to make it more “accurate” based on some things that they know that were not systematically captured? Why is this? Do only those
analysts know the magic to make the numbers look right?How much manipulation of data happens between data capture and data reporting/analysis? Are those manipulations/transformations clear?
If you have a clear chain of custody from the people who entered data (customers – our #1 priority, right?) with limited or systematically defined rules for the manipulation of said data
all the way through your systems, then you have a high level of data accuracy.
7. Integrity: Data integrity is the assurance that data is consistent and correct.
Are you relying on tools within the database like foreign keys to ensure your data integrity?
Do you know where the data that is being reported to you is coming from? If some reports in your organization are accepted without question while people spend hours and days double-checking other
reports, the former report has a high level of data integrity, the latter a low level of data integrity.Do your customers trust your data?
Do your users trust your data?
Do you trust your data?
8. Delivery: The delivery mechanism that is used for the consumption of your data by your business users and your customers.
How do you deliver data to your business users? Is it the result of a manually written query and saved into a spreadsheet where it can be manipulated? How many similar queries are run by
different people to report the same data? Is there a common portal that all of your business users can go to get the data they need?Data delivery is as much a conversation on data warehousing (DW) and business intelligence (BI) as it is anything.
If you do not have a strategy for facing the DW/BI questions, then chances are your data delivery could use some improvement.
9. Quality: Data is of high quality if it is “fit for intended uses in operations, decision making and planning.” (J.M. Juran)
I have heard that people think of data quality as little more that address scrubbing of your data. Address scrubbing is the process of verifying that your customers’ addresses are accurate
according to the USPS.Does this take into account all reasonable instances where data quality can be an issue?
For example, if you have someone manually entering descriptive data for products, what is the quality of that data? What if you have more than one system that contains descriptions of your
products? Is the descriptive information on both systems the same?How often do you take a sampling of daily data and validate it to ensure that all of the data is as expected?
10. Security: Data security is the premise that the “right” people have “access” to the “right” data at the “right” time.
Data security is the means of ensuring that data is kept safe from corruption and that access to it is suitably controlled. Thus, data security helps to ensure privacy. It also helps in
protecting personal data.Does your staff know that security is set up incorrectly, yet they are never given the time, resources or approval to fix it?
Are there processes in place for response to a data security breach?
Are you using the right level of encryption for the protection of appropriate data? What is the frequency of your data being transmitted to other organizations? How high is the probability that
other people would want your data? Do you have encryption keys on your organization’s laptops?
Each one of these 10 areas needs to be assessed and assigned a grade for your organization. Higher numbers can show that your organization is prepared for the course ahead. Low numbers in some
areas show you that you need to focus on some growth before tackling significant new projects. Low numbers in all areas should be a red flag to you. If there are low numbers across the board, then
your organization is at a high level of risk for even simply maintaining the ship.
