Returns and fraud – two key issues for e-commerce in 2020
While returns continue to be a fact of life for retailers for a number of reasons, one area they can control is in having accurate customer address data. This will play a key role in reducing the frequency of ‘return to sender’ scenarios.
The issue today is the reliance by retailers on customer provided data. It can often be a weak link in the mobile age, with many contact forms being completed with ‘fat fingers’ on a small screen. One mistype is often all it takes for, at best, a delay to the delivery, at worst, the delivery to be returned to sender. But it’s not only customers making mistakes, call centres do it as well with errors in data entry.
In fact, it’s estimated that 20% of addresses entered into systems contain errors such as spelling mistakes, wrong house numbers, and inaccurate postcodes. These errors not only have cost implications for the business — due to return and re-shipping costs — but in reputation, as a disappointed shopper can ultimately impact customer retention.
Growth in fraud
Then there’s the matter of fraudulent purchases. Growth in e-commerce fraud has escalated every year since 1993 and is set to increase an additional 14% by 2023. This could cost retailers $130bn globally and is caused by the rise in data breaches and criminals using the stolen ID fraudulently to purchase products and services.
With negative monetary and reputational implications for retailers, it is vital that retailers know that a customer is who they say they are.
Both non-deliverable returns and fraudulent purchases can be solved simply by having access to accurate and verified customer data. The best time to ensure data accuracy and to validate those records is at the customer onboarding stage, when they are about to make a purchase.
Autocomplete, data verification and geocoding the answer
A good first step to improve the quality of address data is to leverage autocomplete functionality. This tool automatically presents a suggested correct version of the address as the customer types in their details, allowing them to choose a valid address that’s easily recognised.
In addition to helping prevent mistakes, autocomplete reduces the number of keystrokes required when typing an address by up to 70%, which speeds up checkout and reduces shopping cart abandonment. It should be implemented as standard by all online retailers.
To ensure data quality and prevent fraud, it’s vital to verify a customer’s entire record by cross-validating shipping and billing addresses, phone numbers and email address with trusted reference data in real-time. Trusted reference data includes electoral roll, credit agency and data from utility companies, for example.
The IP geolocation of a mobile phone can also be used to flag a potential fraudulent order for manual review. If the device being used to place an order is in a different territory from the user’s registered address, it’s worth investigating – particularly for those selling high value items.
Additionally, once retailers have an accurate and verified customer address it’s important for them to consider adding geocodes to that address. Geocodes can deliver precise geographic (rooftop) latitude and longitude coordinates to help improve logistics and, in many cases, reduce shipping costs.
By calculating the shopper’s distance from distribution points, a real-time calculation of shipping costs is enabled, presenting several price levels depending on how fast the customer wants delivery. Also, geocoding functionality can power market segmentation and sales clustering to aid the vendor’s sales and marketing efforts.
Retailers should make it their New Year’s resolution to put in place adequate address, ID verification and geocoding tools at the customer ordering stage. This will ensure they have accurate customer data which reduces returns to sender, foils fraudulent orders and, in the process, delivers significant savings in shipping costs and penalties while protecting the company’s reputation.
Barley Laing, UK managing Director at Melissa