Data Accuracy – just one aspect of excellent data quality

In this blog, we’re highlighting the most common data quality challenges organisations face, along with some practical techniques that you can use to improve your sustainability/ CSR data quality.

In recent years there has been an increasing pressure from investors, the press, governments and other organisations for independent verification of large companies’ sustainability reports’ contents.

According to the 2017 KPMG Survey of Corporate Responsibility Reporting 93% of the world’s 250 largest companies (Global 250) issued sustainability reports and of these 25% have issued restatements (revisions) of past data in their corporate Reports, highlighting the need for improved data quality.

Many large companies look to external consultants such as PwC or SGS for assessment against key standards, such as the Global Reporting Initiative (GRI) and the AA 1000 Accountability Principles (2018), and independent verification does check that all the social and environmental information published is accurate and correct. Given the level of distrust in the current business and political environment across many contexts, issuing your sustainability report with a commitment to reporting quality, including the assurance provided by independent assurance experts, is a valuable way to enhance the trust from your key stakeholders.

However, this article provides some practical steps you can take to improve the quality and accuracy of your data long before you pull together the data for your annual sustainability report or the auditors turn up.

Six common data quality challenges

Data quality has 6 key aspects and failings and any of these dimensions can compromise the usefulness of your data and ultimately can provide incorrect information that can drive bad decision making. Whether or not to continue with a particular sustainability campaign may ultimately rest on the data you report, so quality data can make a huge difference.  Companies with the most successful programmes are those that invest in the time, tools and processes to make sure their data is complete and accurate; they also will save a great deal of time and money on independent verification.

We will look at each of the six aspects, Accuracy, Completeness, Uniqueness, Validity, Consistency and Timeliness in turn.

  1. Accuracy
  2. Completeness
  3. Uniqueness
  4. Validity
  5. Consistency
  6. Timeliness

Accuracy is the first aspect of data quality we will look at and to quote the ISO definition – data accuracy is its trueness and precision.  The diagram below shows the differences between preciseness and trueness and accuracy.

Therefore, accurate data needs both trueness and precision. An example of this is if you are reporting figures that are consistently over or under inflated; it’s hard to spot but a through and regular review of your data and any calculations within it will uncover issues with data accuracy.

The second dimension to data quality is Completeness i.e. ensuring there is no data missing from your data set.  Ensure the data owners are engaged in the process and care that they report promptly as well as ensuring that the data or reports are useful and informative to them, can really ensure they are engaged.  Add in an automated process of checking and chasing submissions to avoid you having to estimate data will minimise duplication and time waste. Understanding your employee’s need and ensuring they are engaged is the key here.

The third aspect is Uniqueness:  this is avoiding any double entry or repetition of data for the same KPI, market, region or site in your data set.  Like the other aspects of data quality a robust data collection and reporting system and processes will make a huge difference to collection as well as engagement.

The fourth key aspect of data quality is Validity:  this means reporting the data in the right way, such as using the correct units of measure and really understanding what you are collecting and why.

Next is Consistency: when recording data points, it is essential it covers the same thing in the same way every time.; tied to accuracy and trueness, conversely this aspect of data accuracy can fool you into thinking the data is right just because it fits with what you reported previously.

Finally, timeliness.

Timeliness refers to the availability of data for analysis and decision making. If the data is missing as it has not yet been entered it’s obviously impossible to use it, and most often ends up being estimated and potentially never updated. The most common causes of late data are a lack of user engagement, system knowledge or engagement; either the user doesn’t know how to enter it or hasn’t prioritised completion. In the next section we highlight the importance of ensuring your users know “why”, and once they do the timeless of your data reporting will improve.

So, taking all those 6 elements, what practical steps can I take ot improve the overall quality of my sustainability data?

Ensure your users know WHY!

Ensuring that everyone involved in inputting, validating and reporting the data is fully engaged in the process and is shown the data in use.  By reflecting back to the users where the data is used, and how it relates to the organisations’ mission can really add meaning to an otherwise thankless task of data entry or validation. Examples can include live dashboards or info graphics shown to users within the system.  Listen to your Managers and find out what they need and why and build that into your data collection processes.

Whilst you might not get everyone excited to report data every day, everyone is much more involved if the data is visible and engaging across the organisation, and by ensuring the reporting process is not silent and hidden, it will pay dividends for your data quality.

Check the data as you go

If you’re reporting on a KPI each month, then you’ll need a robust process that will review that data monthly; don’t wait until the end of the year.  With tight reporting deadlines at most year ends, checking the data sooner rather than later, gives you more time to investigate possible errors and therefore reduce your stress levels!  Annual data collection is probably the most time intensive way of doing things.  By collecting monthly, quarterly or even six monthly you find that the process is far smoother and efficient, and any problems get ironed out before the year end madness.

Use all the functionality your system offers

Whilst spreadsheets and specifically Excel is seen as the low-cost option for tracking CSR and sustainability data, relying on basic spreadsheet software will actually cost more annually in staff time, inefficient workflows and errors; a major issue with using Excel as is that data is often of inconsistent quality, difficult to audit and nigh on impossible to manage well.

Using a simple software solution to collect your sustainability data, can dramatically improve data quality as well as freeing up significant time and cost in managing the data and driving user engagement in your sustainability programme.  Once you have one, it is essential to keep your software up to date – both in terms of upgrading (if that isn’t automatic), user training and system configuration to ensure you are getting the maximum return on your investment. Below are just some of the many additional functions software provides over a spreadsheet or basic database solution to vastly improve your data quality:

  • Multi step data validation – a two or three step validation process, where the local data entry is checked by a regional and or senior manager
  • Data visualisations – providing dashboards containing graphs and charts to allow users to quickly check for spikes and anomalies
  • Expected range tolerances – limited fixed range of numbers that can be entered
  • Validation rules – predetermined values, often with a feature to limit variation against a previous data point
  • Seasonal validation – pre-set limits based on degree day variances

To recap, the three main ways to improve your CSR data quality are to:

  • Engage with everyone who is responsible for the data
  • Increase the accountability for the data
  • And finally, implement a continuous monitoring process and keep systems and processes up to date.

SustainIt has become the leading expert in sustainability data and data systems for data collection. We have been doing this for over 15 years and have a fantastic team of data analysts who really know and understand sustainability data.  If you need some pointers or advice please do get in touch and we’ll see if we can help you or give you some pointers to make your data management more efficient and more effective.

By Nicola Ainger, Managing Director at SustainIt Solutions

Contact Nicola on 0044 117 325 4168 or

Back to Our Stories