There are two sayings that we will all have heard at some time:

  • Start with the end in mind
  • Start like you mean to go on

A client of ours demonstrates why these two sayings are important from a data perspective.

From a data perspective, the first of these two sayings “Start with the end in mind” is a reminder that you need to have a clear understanding of the purpose that the data is required for and ensure that this data is captured at the correct point.

The second saying “Start like you mean to go on” supports the first saying by reminding us that there is no point starting an activity or a process in a half-hearted way believing that ‘someone else’ will sort out the data later.

It is important to remember that in data activities, similar to a race, if you do not get a good start it becomes ever harder to catch up with the main competitors in the field and virtually impossible to win the race (unless you are extremely lucky or far better than your competitors).

A client of ours demonstrated why starting well is also essential from a data perspective – their transactional business is heavily reliant on data to support different stages in their processes. They do not have any clearly defined data standards, have limited process definitions and do not check that new customers have an agreed minimum of data recorded when they are ‘on boarded’.

Not surprisingly, the sales teams are motivated by ‘the next sale’ and any ‘downstream’ problems are not their worry, so they carry on as they have always done, chasing that next sale. All following ‘downstream’ processes are then hampered by data that is either missing or wrong, therefore, wrong decisions are made or time is spent trying to find data that is good enough to complete that transaction.

Overall, the lack of data standards and checks on the quality of data at the start of the process means that the efficiency and profitability of the organisation as a whole is reduced by poor data. Equally concerning is the impact on customers and customer perceptions – when time has been spent building up a good reputation with a customer, it can quickly be lost by a poor outcome, which is likely if data cannot be relied on.

So, are there any pointers to how to avoid these problems?

  • Ensure that you have data standards defined, particularly for key process steps, such as ‘on boarding’ a new customer
  • Define the minimum data expected from these key process steps – only use mandatory data fields where essential – if a mandatory data field cannot be completed for valid reasons, staff will try to use any value that helps them complete the transaction
  • Monitor the quality of data from these key steps – use comparative reporting, league tables etc. to highlight potential problem areas
  • Review performance targets and incentives to ensure that they do not incentivise creating problems for downstream business processes
  • Ensure that there is sufficient governance over the process to oversee performance and to address emerging issues and performance problems before they get to significant

Such approaches are part of a coherent approach to data quality management, such as that defined in ISO 8000-61 which we have described in more detail here.

So, are you starting well?

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.