How much is poor quality data costing you?

Experts think that handling data quality issues typically costs organisations between 10% and 30% of revenue (Experian 2021).  

And that’s purely the direct financial cost. The non-monetary costs of poor-quality data can be even more significant, taking in everything from reputation damage to misguided strategic decisions. According to Experian’s 2021 Global Data Management Research report, 95% of businesses have seen impacts related to poor data quality. 

What counts as poor quality data? 

It’s important to be clear what the term ‘poor quality data’ means in order to see why data quality matters.  

Poor quality data includes data that is one or more of the following: 

  • Inaccurately captured (typing errors, and deliberate mis-types where people don’t want to give the information) 
  • Incomplete (often caused by poor data capture through web forms) 
  • Out-of-date (people die or circumstances change – addresses, jobs, requirements) 
  • Inconsistent between sources (where details on a web enquiry form, for example, differ from those on an order) 
  • Duplicated (where people have completed a form more than once but used slightly different details each time – for example, ‘Rob’ and ‘Robert’.) 
  • Redundant (people are no longer interested; your offer is no longer relevant)  

The costs of poor data quality  

The costs of poor quality data are seen across the organisation and at both strategic and operational levels.   

1. Reduced revenue 

Poor quality customer data could be reducing the effectiveness of sales and marketing campaigns and contributing to declining revenues. 

If data isn’t current, you could be targeting people who are no longer relevant, have moved house, or have even passed away. If you’re in B2B, note that around 10% of people change jobs each year.  

Duplicate records mean customers receive duplicate emails, prompting unsubscribes. Incomplete or incorrect records of customers’ purchase histories or interests make accurate targeting impossible. You could be advertising in the wrong places, or sending irrelevant communications.  

Where contact details are incorrect, a record is useless. And if too many emails bounce, you see higher spam ratings, which in turn means fewer emails landing where you want them to land.  

Poor customer data also makes it harder to monitor what sales activity is working and what isn’t. If you’re not getting the responses you were expecting, how can you know if this is due to the sales and marketing activities or to issues with the data? 

2. Wasted resources 

Databases with poor data quality provide multiple opportunities for wasted time and money.  

A few examples:  

  1. High rates of duplication in a customer database (we’ve found duplication rates of 30% and more). Take a database of 60,000 – what could 20,000 duplicates be costing in database storage and email service subscriptions or direct mail postage costs? 
  1. Incorrect or out of date B2B data. Sales teams can easily spend more time calling the wrong people or trying to call non-existent people than talking to genuine potential leads.  
  1. Incomplete or inaccurate financial data. Finance reconciliation problems when bank statement lines don’t match finance system entries. Staff must search for the relevant transaction and do a manual reconciliation. 

3. Operational problems 

Operations in every department rely on robust data. A duplicate personnel record could mean incorrect pay calculations, for example, whilst missing customer data could prevent order fulfilment. Poor data around risk assessments and accident records put lives at risk. Inaccurate inventory data makes it impossible to manage supply chain challenges effectively. 

4. GDPR breaches 

Poor quality data puts you at a higher risk of accidental GDPR breaches. Where you have records in multiple databases or even duplicate records in the same database, it’s hard to keep track of issues such as email consent.  As a result, you can find yourself sending communications to people who have not given you their permission. This can lead to both fines and reputational damage – as Saga, We Buy Any Car and Sports Direct found out recently to their cost

5. Reputational damage 

GDPR breaches are not the only way poor quality data can harm your reputation. 32% of respondents in the Experian research said poor data quality negatively impacted on their reputation and on customer trust. Customers expect you to get things right: to deliver to the correct address, to provide important information when they need it, to deal with issues taking relevant information into account.

Mistakes can lead to negative media exposure (including unfavourable social media coverage), as well as to critical online customer reviews and a low net promoter score performance (NPS). 

6. Unreliable decision making 

Good decisions rely on sound data. Poor data means statistics are flawed, accurate analysis impossible, and insights unreliable. Consider, for example, trying to plan regional services without being confident in the demographics of service users, or making decisions about future sales promotions if you’re not sure how customers have responded in the past.  

And once you start to mistrust your data, you become more reluctant to act on what it may or may not be telling you. You spend more time questioning the data; decisions get delayed, and new initiatives may be abandoned.  

7. Poor ROI on digital transformation investment 

Digital transformation is key to increasing productivity. Moving away from manual processes should boost efficiency and drive down costs. But if you automate workflows without considering data quality, you won’t see the full benefits. 

Say you’re implementing technology to link social media follower data with customer information held in a marketing database and so streamline marketing communications. How can you do this effectively if the quality of the data you hold in the database is poor? 

8. Poor resilience 

Organisations, where data quality is poor, are less agile and less resilient. As Covid-19 has shown, the ability to rely on and leverage data gives a key advantage when times get tough: the more you know, the quicker you can adapt. 

Retailers with robust data, for example, were able to pivot faster from store to online, monitor customer behaviour, and consider how behaviour might change as stores reopened. Manufacturers with deeper supply chain insights were aware of supply problems sooner and got ahead of competitors in finding solutions. 

How can you fix your data quality issues? 

Poor quality data won’t sort itself out. 

Don’t try to do it manually

As soon as you start to think about the volume of data you process every day, it’s clear there’s no way you can manually fix every issue. Even if you do try to tackle individual issues on an ad hoc basis, you’re not solving the causes of the problems. The same issues will keep coming back. 

Understand your data landscape

But before you can begin to solve your issues, you need to understand them. A data landscape analysis is the best way to find out exactly what you’re dealing with. Using the results of the analysis, it’s then possible to build a view of what you want your data to look like in future and how to make it happen.  

Rules, tools, and stewards

Three elements will be key in delivering this data vision: data quality rules that sit across your whole data set, automated tools that integrate, refine, dedupe and cleanse, and long-term behaviour change that embeds data stewardship as a collaborative effort. 

At Qbase, we take clients through the complete process, from analysis to successful ongoing data operations. We use tools such as Talend to embed data quality as data moves between systems and around the organisation. 

A successful data quality initiative can transform your organisation. Operations become smoother as data frustrations disappear. Customers get happier, because your communications become more relevant, and their experiences of your services improve. Marketing becomes more effective because you’re reaching the right people with the right content. And once you start to trust your data, you can make better decisions, faster. 

Want to explore the bigger picture of data management? Data quality is just one aspect. Our article on Taking the Pain out of Data Management tells you more. 

 

Related Articles

Article
Taking the Pain out of Data Management

Managing your data can seem like a daunting task. Especially when it involves cleaning up a data catastrophe that’s been years in the making. But there’s simply no getting away from it. It is well known that the consequences of bad quality data can end up having a significant impact on your business.

Whitepaper
WHITEPAPER: Preparing for a major data initiative: The importance of a Landscape Analysis

In this whitepaper Rob Jones explains exactly why analysing your data landscape prior to starting your project is so important. Download to find out what steps to take in order to analyse your data landscape in readiness for your data project. 

Article
Five ways our landscape analysis helped Parkinson’s UK improve their data quality

We recently worked with Parkinson’s UK to improve their data quality and help inform the data architecture. The first thing we did was to conduct a landscape analysis – a rigorous process of data analysis and documentation. The landscape analysis played a vital role in helping the charity look at the current state of data quality. As well as looking at the data itself, the landscape analysis looked at data processes, integrations, and platforms used.