Notifications
Clear all
Topic starter
12/07/2026 12:20 pm
TL;DR: Travel booking platforms processing high volumes can no longer rely on CVV and AVS alone, with Tripster saying it blocked $14.4 million in fraud while protecting 100% of transaction volume, according to Riskified. The practitioner takeaway is that fraud prevention now has to preserve checkout speed, not simply raise more alerts.
NHIMG editorial — based on content published by Riskified: a Q&A with Tripster CEO John Johnson about fraud prevention and booking growth
Questions worth separating out
Q: How should travel merchants balance fraud prevention with checkout conversion?
A: They should treat fraud prevention and conversion as a single decision problem.
Q: Why do basic card checks fail in high-volume booking environments?
A: Basic card checks fail because they only validate limited payment data at a single moment.
Q: What do fraud teams get wrong about automated checkout decisions?
A: They often focus only on blocking fraud and ignore the governance needed to explain, tune, and audit decisions.
Practitioner guidance
- Map fraud controls to fulfilment timing Place the strongest risk decision before instant ticket issuance or other irreversible fulfilment steps, because post-purchase review cannot recover value once the buyer has consumed the product.
- Build a layered decision model Combine CVV and AVS with behavioural signals, velocity checks, device intelligence, and prior transaction outcomes so the review process can distinguish low-risk customers from coordinated abuse.
- Track approval rate and loss together Set operating targets that balance conversion and fraud loss instead of treating them as competing dashboards, because one-sided optimisation usually hides risk transfer.
What's in the full article
Riskified's full blog covers the operational detail this post intentionally leaves for the source:
- How Tripster operationalised automated fraud checks while keeping checkout friction low for legitimate buyers.
- The specific backend decision details available to analysts after a capture or decline, which matter for tuning and dispute handling.
- How Riskified and Tripster approached approval-rate fluctuations and model re-analysis over time.
- The business-facing framing of why the guarantee mattered in a reseller model where fraud losses exceed net revenue.
👉 Read Riskified's case study on Tripster's automated fraud prevention →
Travel checkout fraud: how should platforms balance approval and risk?
Explore further
12/07/2026 12:33 pm
Checkout fraud at travel scale is a decision-quality problem, not a manual review problem. Once volume rises into six figures of annual transactions, human review becomes a bottleneck that attackers can route around and legitimate customers can feel. The control question is whether the business can make reliable, low-friction decisions in real time. Practitioners should treat risk scoring, approval logic, and dispute handling as one operating model, not separate functions.
A question worth separating out:
Q: Who is accountable when fraud controls block legitimate customers in real time?
A: Accountability should sit with the team that owns the end-to-end decision path, not only the fraud model. If checkout, identity, and risk signals are not orchestrated into one control, then the business is responsible for the conversion loss as well as the fraud loss. Governance needs shared ownership across fraud, product, and security leaders.
👉 Read our full editorial: Fraud at checkout in travel platforms demands automated review
