Coming up with ways to control false declines
37250
post-template-default,single,single-post,postid-37250,single-format-standard,bridge-core-3.1.3,qode-page-transition-enabled,ajax_fade,page_not_loaded,, vertical_menu_transparency vertical_menu_transparency_on,qode_grid_1300,qode-child-theme-ver-1.0.0,qode-theme-ver-30.2,qode-theme-bridge,wpb-js-composer js-comp-ver-7.3,vc_responsive
 

Coming up with ways to control false declines

Ai Editorial

26th October 2023

Anything that forces a legitimate shopper to leave the cart is an opportunity lost for a merchant.

In this context, it is interesting to assess how travel e-commerce players and fraud prevention specialists are looking at ways to sort this issue. They are keen on overcoming inaccuracies that lead to the rejection of valid transactions.

In this context, it is interesting to assess how #travel e-commerce players and fraud prevention specialists are looking at ways to sort this issue. They are keen on overcoming inaccuracies that lead to the rejection of valid transactions.

According to Chargebacks911, a vertical like travel can experience #payment decline rates of 20% or even 30%. The team recommends that a merchant shouldn’t reject orders based on generic indicators. Deploy smarter, more advanced fraud filters capable of understanding context more clearly.

Another way to cut down on friction or let more legitimate transactions go through is to separate acceptance from stringent checking for fraud. This would mean that transactions are given a green signal and then taking a call from security perspective with a detailed risk assessment. As a specialist in this arena, FUGU – Every Payment Counts states that “KYC verification is initiated only when the transaction risk warrants it” and “the intensive post-payment verification uncovers nuanced and intricate fraud attempts”.

Overall, the industry is looking at capitalizing on data in real-time to act on time in order to understand which transactions are likely to be false declines. Also, review transactions that have been declined. SEON’s recommendation: use your historic data to verify if the new risk rules actually reduce false declines. Swiftness of machine learning models comes into play. The risk scores need to be more accurate. It has to be ensured that the information used for verification during transactions is up-to-date. Checkout.com recommends that one way is to create smarter filters that have a keener understanding of the context of a transaction. Plus, how to count on manual reviews as of today. Another aspect that must be delved into is chargeback guarantee tools.

By Ritesh Gupta, Ai Events

Follow us:
 



Share