Develop a fraud management lifecycle for your e-commerce business
Follow this four-step process to try and increase revenue from your e-commerce payments flows while minimising customer friction in your check-out experience
Use managed machine learning as much as possible through all stages of your fraud management lifecycle
A robust system to detect, analyse and mitigate fraud is essential for any e-commerce business, but particularly important for merchants expanding into new markets or new regions. If you’re concerned about the level of work required to analyse fraud reports every day while your workforce are also busy growing market share, start thinking about increasing the level of automation and machine learning in your fraud software. But remember that machines only learn as much as you tell them, so look for a managed learning solution that allows you to set some baseline rules and then customise those to suit your business.
1. DETECTION: Identify fraud trends and variance in authorisations
You need to make sure that you can see trends, in order to develop rules to identify them. The most important details to understand from your fraud trend reports are peaks in fraud across countries, regions, times of day, week or month, cart totals, devices, IP addresses and billing addresses. Remember that fraud is instant. You don’t know when it will strike or where it will come from and it can move on somewhere else just as quickly.
2. STRATEGY DEVELOPMENT: Develop rules based on trends to mitigate losses and prevent fraudulent transactions
Lean on your acquirer for support and consultancy here. It is in their interest to work together to protect the e-commerce ecosystem, so they should be helping you manage the rules, creating test rules and configuring the solution console to give the right insights for your business
3. DEPLOYMENT: Set baseline rules and start the managed machine-learning process
To ensure that your software has the ability to detect, don’t just take the software out of a box and tell it to run free. Set baseline rules to start the minimum amount of detection until the product gets to know your customers.
4. MONITORING AND MAINTENANCE: Monitor and review rule performance and remove or adjust accordingly
Review the process constantly. Even if you are not seeing a lot of fraud-related chargebacks, it’s advisable to still invest in a robust fraud management strategy, because technology is expanding across many markets so threats have the potential to increase exponentially.
Complement your fraud management lifecycle with a proactive acquiring strategy. Your acquirer can help you:
1. Manage wallets and use tokenisation
2. Optimise your transaction messages
3. Develop a balanced re-try strategy
4. Build advocacy and outreach with issuers
Learn more about the revenue-driving tactics you can implement in your authorisation optimisation strategy.
Contact your J.P. Morgan representative or call our merchant support team on:
Europe: +353 1 726 2909
UK: +44 845 399 1130