More Than a Quarter of Customer Records Contain Errors

Lewis Hamilton shakes off Taylor Swift to become UK’s most-spoofed name

Thursday 25th July 2024, London, UK - Technology-enabled data consultancy Sagacity has revealed the findings from analysis of 140 million customer records which shows that a massive 28% of them contain errors.

Taking a deep dive on a sample of one million retail customer records that had been flagged as having errors, Sagacity found that:

• 90% of the errors related to the customer's address - with 15% having multiple records
• 10% contained errors in the name

"Small errors can add up to big problems and create real headaches when it comes to linking records. For instance, a customer called Thomas Black may go as both Tom and Thomas – they are the same person, but the system thinks they are two different people. This can lead to duplications, unnecessary costs of contacting the same customer multiple times and a poor customer experience,” comments Sagacity CEO, Anita Dougall.

 “Moreover, if an address is incorrect, communications may not even reach the customer – which can create a debt pile that is hard to recover if bills go unpaid. This is a widespread problem: our research found that businesses lose almost £50 billion a year due to inaccurate or incomplete data, poor data reconciliation, and having multiple versions of the truth. It’s vital that companies have the right processes in place to capture data in a consistent way, and that data is routinely checked and cleansed for accuracy.”

Spoofs, profanities and gibberish

Many of the incorrect names in retailer databases are not mistakes – with customers deliberately giving fake names. Sagacity says these tend to fall into three categories: 60% tend to be spoof names (often relating to celebrities or popular culture), 35% are profanities, while 5% are ‘gibberish’ (e.g. First name given as ‘Aaaaa’).

When looking at spoof names, certain names come in and out of vogue based on what is happening in the world. After analysing a separate sample of 100 million customer records and factoring in a small number of genuine cases, Sagacity can reveal the most commonly used suspected ‘spoof names’ are:

1. Lewis Hamilton
2. Harry Potter
3. Gordon Ramsey
4. Taylor Swift
5. Indiana Jones

Yet Sagacity warns companies against making assumptions. “Some names may seem like a joke, but you need to be sure – it is not a black and white issue,” continued Anita Dougall. “For instance, there could well be a Mr Smellie, who lives in Shitterton. Attempting to delete or correct their records could damage the relationship and cut off a stream of income. Rather than having blanket policies, businesses need to dig a little deeper to check third-party sources, to prevent inadvertently deleting or editing customer data that is perfectly valid.”

Learn more about data quality