6 ‘P’s For A Data Quality Framework

Why do you need data quality?

Back to my house analogies. Try building a wall or even a house, if the bricks all differ in size and shape, some missing and others are just the wrong type, being large stones instead of bricks. Well, you could probably build it, but would it be fit for purpose and pass the inspection criteria…? 

The same goes for data. If we are going to generate revenue from our data, reduce risk, nurture trust for people to reuse the data so they don’t create their own ‘Shadow IT’ with their multiple spreadsheet copies of data. We need to have data that meets the defined standards, is consistent, complete, accurate, valid, etc. 

For example

If your inventory records and product sales data is incorrect, leading to some stores receiving too much stock and others not receiving enough from the ordering system. Customers are being let down in the stores not receiving sufficient stock and therefore losing sales and you have stock being damaged in other stores as there is nowhere to store it, as their stock room is overflowing. Again losing potential revenue as damaged stock is written off and disposed of.

Having the correct data to be able to trust and reuse has limitless advantages and opportunities. If you want to increase revenue opportunities, reduce risk, and develop operational efficiencies, data quality is the first place to start. Even data migration can’t be done well without data quality. Managing your data is the first step on any new system migration.

 I’d go as far as saying Data Quality is the most important activity in any organisation! 

For those saying we operate just fine today without investing in data. Ask your department heads, what are our untapped opportunities or where are our hidden costs for correcting data. Successful business can not operate with blind spots! 

Where do you start?

You have so much data in numerous locations across your entire business. Do you carry out a data audit, a data assessment or implement a data catalog….all remind me of attempting to boil the ocean…do you think that’s possible, you’d need a mighty big kettle! 

How about starting in a small manageable way, a chunk at a time. Like how to eat an elephant (not that you’d want to do that or my favourite analogy is to develop the first ripple of a pond when your throw a pebble in. 

 6 ‘P’s for Data Quality

Let’s start with that first ripple. What is the data that’s critical to the business operations where a lot of noise is being made. Examples from my experience: The reports are not accurate; The logistics team can’t load the lorries effectively due to inaccurate product dimensions; Marketing spend more time cleansing data than contacting customers; We waste money communicating with ‘gone away’ customers; Compliance costs are higher due to premiums paid as we use estimated data to assess premium risk; Customers are complaining about product information being incorrect, risking the relationship and future sales; Procurement are unable to determine the accurate spend with each supplier due to duplicate accounts; Our stock holding costs are inaccurate as we hold stock on obsolete products which should be discounted or written off, due to inconsistent product status across our system landscape. 

Once you have determined where to start try this framework approach for improving your data. 

• 1. Purpose
• 2. Principles
• 3. Process
• 4. Plan
• 5. People
• 6. Perform  

Purpose

With data success your need to invest to generate the results. That’s a combination of investing in: ·       

• People with the right skills, upskilling others
• Technology to assist the productivity of your data stewards to enable consistency
• Time, data success is a journey, nothing is achieved overnight. It takes commitment, leadership and tenacity

Subscribe to continue reading

Become a paid subscriber to get access to the rest of this post and other exclusive content.

Already a paid subscriber?