Common Enterprise Wide Data Governance Issues: #2 The quality of data entered by front-end staff is not as high as desired

This post is one of a series dealing with common Enterprise Wide Data Governance Issues.  Assess the status of this issue in your Enterprise by clicking here:  Data Governance Issue Assessment Process

Causes:

  1. Front-end staff are put under pressure to ‘sell business’ and sort out details later.
  2. Front-end staff have no incentive to enter high-quality data.
  3. Front-end staff are unaware of the downstream impact of poor data.
  4. Little realisation on the part of front-end staff that data entered at account opening may be re-used repeatedly for 30+years.
  5. Front-end systems may lack data validation software.
  6. Front-end systems may not enforce business rules between related fields.
  7. Reluctance to add data validation to legacy Front-end systems

Impact:

  • Incomplete or poor quality ‘informational’ or ‘Master data’ captured at data entry.
  • Risk that the enterprise and/or individual business units will fail to comply with regulatory requirements (e.g. the requirement to ‘Know your customer’).

Solution:
Agree and implement the following policies:

  1. Data owners must define the business rules to be applied to each data item at data entry time.
  2. Data owners must agree Data Quality requirements with the end users of the data, and the owners of the data entry processes.
  3. Data owners must agreee Data Quality Service Level Agreements with the end users of the data, and the owners of the data entry processes.
  4. Data Quality measurement must be implemented at each point of data entry to measure the quality of data being entered, and support the implementation of data quality Service Level Agreements.
  5. Front-end staff must be educated on the importance of data quality
  6. Front-end staff must be supported in taking the time and effort to “get it right first time”

Your experience:
Have you faced the above issue in your organisation, or while working with clients?  What did you do to resolve it?  Please share your experience by posting a comment – Thank you – Ken.

Common Enterprise Wide Data Governance Issues: #1 The quality of informational data is not as high as desired

In most Financial Institutions, transactional data quality is quite good across the Enterprise, in that customers transactions are successfully processed.

However, the quality of informational data (Master Data), such as customer details, is typically not as good as desired.  This is the ‘generic data governance issue’ that faces most organizations.   In this series of posts, I will take this high level data governance issue to a lower level of detail.

This post is one of a series dealing with common Enterprise Wide Data Governance Issues.  You can see the full list, together with the process for assessing the status of the issues in your Enterprise by clicking here:  Data Governance Issue Assessment Process

Your experience:
Have you faced the above issue in your organisation, or while working with clients?  What did you do to resolve it?  Please share your experience by posting a comment – Thank you – Ken.

Common Enterprise Wide Data Governance Issues: #3 No culture of Data as an ‘asset’, or ‘resource’

Some enterprises fail to recognise the true value of their data.  This post is one of a series dealing with common Enterprise Wide Data Governance Issues.  Assess the status of this issue in your Enterprise by clicking here:  Data Governance Issue Assessment Process

Impact:

  • There is little value attributed to capturing and maintaining high quality ‘informational’ or ‘Master data’.
  • The other Enterprise Wide Data Governance Issues in this series are all symptoms of the failure of the enterprise to treat Data as a corporate asset.  An Enterprise that treats Data as a valuable corporate asset understands the value of data, and is likely to have addressed the issues I have identified.

Solution:
Agree and implement the following policies:

  • Data must be treated as a valuable Enterprise asset, that can assist the Enterprise achieve its strategic objectives, and must be invested in proportionally to other Enterprise assets.
  • The CIO is responsible for ensuring that the quality of Master data is measured, target data quality levels are agreed, and measures are implemented to meet the defined targets.

Your experience:
Have you faced the above issue in your organisation, or while working with clients?  What did you do to resolve it?  Please share your experience by posting a comment – Thank you – Ken.

Common Enterprise Wide Data Governance Issues: #4 No clear ownership of data

In many large organisations, it is unclear as to who ‘owns the data’.   There is no one responsible for providing existing data to new projects that require it.   There is no one responsible for measuring, or maintaining data quality.  It is unclear who to ask when looking for information, e.g. to satisfy new regulatory requirements.  This post is one of a series dealing with common Enterprise Wide Data Governance Issues.  Assess the status of this issue in your Enterprise by clicking here:  Data Governance Issue Assessment Process

Impact:
New projects dependent on existing data (especially Master Data) must locate the data from first principles, and face the risk of not finding the data, or identifying inappropriate sources.   New projects dependent on existing data take longer than necessary to complete.  Multiple re-invention of the wheel results in the creation of islands of data, and a maintenance nightmare.

Solution:
Agree and implement the following Data Ownership policies:

  1. Overall ownership for data within the Enterprise lies with the CIO.
  2. Ownership for data within each Business Unit lies with the CIO and the head of the Business Unit.
  3. The CIO and head of each business unit must appoint a person with responsibility for the provision of data from that business unit, to those who require it.  This person is also responsible for the measurement and maintenance of data quality within the business unit.
  4. The CIO and head of each business unit must appoint a single point of contact to handle requests for data / information.

Your experience:
Have you faced the above issue in your organisation, or while working with clients?  What did you do to resolve it?  Please share your experience by posting a comment – Thank you – Ken.

Process for assessing status of common Enterprise-Wide Data Governance Issues

If you work with data in large enterprises, you will be aware that the data, and the ability of the business to access that data is seldom as “good” as it should be.  But just how “good” or “bad” is it?

This post outlines a process for assessing the status of common Enterprise-Wide Data Governance  issues within your enterprise, or that of a client.  I use it as the basis for my “Data Governance Health Check”.

These issues can impact your ability to deliver the underlying data required for meaningful CRM, Business Intelligence, etc.. More seriously, they can impact your ability to satisfy regulatory compliance demands (e.g. GDPR, BCBS 239, Solvency II, Anti Money Laundering, BASEL II etc.) in a timely cost effective manner.

Do issues like these affect your enterprise?  If not, how have you resolved or prevented them?  Please share your experience by posting a comment.

Common Enterprise-wide data governance issues:

1. Quality of informational data is not as high as desired


2. Quality of data entered by front-end staff is not as high as desired


3. No culture of Data as an ‘asset’ or ‘resource’


4. No clear ownership of data


5. Business Management don’t understand what “Data Quality” means


6. No Enterprise Wide Data Quality Measurement of Data Content


7. No SLAs defined for the required quality level of critical data


8. Accessibility of data is poor


9. Data Migration and ETL projects are Metadata driven


10. No Master repository of Business Rules


11. No ownership of Cross Business Unit Business Rules


12. No Enterprise Wide Data Dictionary


13. Islands of Data

14. No Enterprise Wide Data Model

Explanation of the scale and the process for using it:

There are 6 levels on the scale, starting at level 1, and increasing to level 6.  The higher the score, the better prepared the organisation is to deal with the issue.  The worst case scenario is actually a score of ZERO, which means that management in the enterprise is not even aware that the issue exists.  To assess the actual status of an issue, ask for documentary evidence to illustrate that the Enterprise  has actually reached that level:

Figure 1: Status of a (data governance) issue.

1. Aware Senior Management is aware that the issue exists.e.g. Data Quality is not measured, or measured in ad-hoc manner.#Evidence: Captured in Issues Log or Requirements document.
2. Understands Senior Management fully understands the issue; the impact of not addressing it; options available to address it, complete with the pros and cons of each option.e.g. Issue paper explains the impact of no Data Quality Metrics on downstream data dependent projects etc.Evidence: Issue Paper, Rationale paper or Point of View paper(s).
3. Policy defined Senior Management has a clearly stated policy/strategy identifying the selected option.e.g. Data Quality Measurement must be performed by each Business Unit, using a standard Enterprise Wide Data Quality Measurement process….Evidence: Policy document / Design Principles/ Communications/ education material
4. Process defined The organistaion has a clearly defined process detailing exactly how the policy / strategy will be implemented, which common services / utilities must be used, and exactly how to use them.E.g. The standard Enterprise Wide Data Quality Measurement process will use ‘off the shelf tool X’, to produce a standard set of Data Quality metrics….Each BU must train N staff in the use of the tool.  Training will take place……Evidence: End To End Process documentation / Education and Training material.
5. Infrastructure in place Infrastructure (systems / common services / utilities) needed to implement the process is in place.E.g. ‘off the shelf tool X’ has been licenced and installed Enterprise Wide.  Staff have been trained …Pilots have been run…Evidence: Programme Infrastructure document / Utility user manuals.
6. Governance in place Governance is in place to ensure that the defined policy is implemented in accordance with the defined process.E.g.  The stakeholders are…The Data Steering Enterprise includes the CIO and ….The reporting process is….. The following controls are in place….Evidence: Programme Governance document / Education / completed sign-offs

Your experience:
How do you assess Data Governance within your organisation, or that of a client? Please share your experience by posting a comment – Thank you – Ken.

Business is all about data

Technologies may come and go.  At the end of the day, business is all about data.

Take the banking industry:
Hundreds of years ago, banks had Customers, with Accounts, on which Transactions were recorded. Bankers knew their customers personally, and all details were recorded by hand in ledgers, using quills made from feathers. Over time, quills were replaced by fountain pens, and later by biros, to record customer, account and transaction details.

Fast forward to today:
Banks still have Customers, with Accounts, on which Transactions are recorded, only many, many more of them.   Financial Regulators require banks to “know your customer”, but it is physically impossibe for bankers to know their customers personally. Customers can now perform transactions via multiple channels, at the bank branch counter, over the internet, over the phone, using mobile devices.

Customer Relationship Management (CRM):
To provide their customers with the best service, banks have implemented “Customer Relationship Management” or CRM systems. CRM systems analyse data to identify situations when the bank may wish to contact the customer to offer additional services, or otherwise improve the service the bank provides to the customer.

Money Laundering, Fraud, Terrorist Financing:
Banks today face ever increasing risks of Money Laundering, Terrorist Financing and Fraud.  Regulators require banks to implement best practice Anti Money Laundering (AML) and Anti Terrorist Financing solutions.

Best practice solutions:
What is the common thread amongst the best practice AML solutions?  How do Anti Money Laundering solutions enable a bank to “Know your customer”?  How do Anti Money Laundering solutions identify Accounts that require investigation? How do Anti Money Laundering solutions identify “Suspicious Transactions” amongst the millions of transactions the bank processes daily?

The answer:
By analysing the data.  However, the data analysis must be targetted.  The analysis must seek out defined activity patterns, and then alert trained staff to the possibility of wrongdoing.   More sophisticed AML systems can identify transaction activity that is unusual for a given customer type, by performing “Peer Group Analysis”.   For “Peer Group Analysis” to work, a bank must be able to reliably distinguish between different customer types.   Distinguishing between different customer types is often more challenging than one would think…