Data quality: the essential steps to improve and maintain the quality of your data

Ioannis Nerantzakis provides you with vital knowledge to manage data quality and create useful analytics. Ioannis is speaking at the CIPD HR Analytics Conference and Workshop, 27-28 November in London. Book your ticket today.

Data has become an integral part of our daily job and its quality can be critical for effective decision making. KPMG “2016 Global CEO Outlook,” highlighted that 84% of CEOs are concerned about the quality of the data they’re basing their decisions on. Overlooking it can be very time consuming and expensive as organisations (according to Gartner research) believe poor data quality generates on average $15 million per year losses (financial, reputational, missed opportunities). The reasons for that are various, including silos of information, lack of executive buy-in, legacy data and the urgency to deliver tasks based on quick fixes instead of resolving root causes.

Now let’s explore the essential steps to improve and maintain data quality. Both a long-term and a short-term solution are required to respond to this challenge. The first, a Data Quality Project, will initiate a mechanism to look at the big picture and resolve the issues completely, ensuring the improvement and maintenance of data quality in the long term. The second, should respond to the question “What can we do now?” and will cover the time-period until the long-term solution will be implemented.

The building blocks of a Data Quality Project clearly defines the why, what, how and who demonstrated in the below 4 phases:

  1. Define and Analyse:
    i. Determine the business needs and assess business impact by clearly defining the issue, opportunity, approach and goals to drive you through the project
    ii. Analyse the current environment by mapping issues, stakeholders and careful prioritisation
    iii. Assess data quality by utilising data quality dimensions and identify root causes through testing and deep investigations
  2. Plan relevant solutions that firstly prevent future errors by tackling root causes and then arrange appropriate data corrections
  3. Implement solutions and establish ownership while setting controls and reviews to monitor improvements and ensure future maintenance of data quality
  4. Reflect on the results against the goals set initially and communicate the positive impact and the way forward

Driving and maintaining data quality needs to be a visible extension of the project and includes two critical elements:

  1. Systems, Controls and Processes to drive data quality by design
  2. Stakeholders and Ownership to engage people and make them feel part of the solution to sustain improved quality

In the meantime, the urgent need of data will require us to use a short-term solution which may involve:

  1. Poor-data quality classification to clarify which data can be resolved, which should be used carefully and which should not be used
  2. Exploring alternative data, extraction techniques and segmentation can sometimes be surprisingly useful, otherwise, utilisation of automated data manipulation and transformation methods might be the solution

Above all, accurate interpretation and clear stakeholder communication can be critical to maintain credibility and trust.

If you are interested to find out more, a full session on the topic will be presented in the CIPD HR Analytics Conference and Workshop on 27 November 2018.

Ioannis Nerantzakis is presenting ‘Data quality: the essential steps to improve & maintain the quality of your data at the CIPD HR Analytics Conference and Workshop, 27-28 November in London. Book your ticket today.