How VALID
CheckDetect Works
The deposit enters the workflow.
CheckDetect evaluates the deposit as it moves through the institution’s existing teller, ATM, or mobile workflow.
Behavioral context is added.
The model evaluates transaction and account behavior, including payer-related signals, rather than relying on the check image alone.
Risk is scored.
CheckDetect scores the deposit so the institution can apply its configured risk thresholds and route higher-risk deposits for review.
Image review sees the check. CheckDetect sees the behavior.
Traditional review often evaluates the item in front of it: the image, the amount, or a rule triggered at that moment. Fraud is more contextual. CheckDetect evaluates the deposit against behavioral and transaction patterns so risk teams can see what the image alone cannot show.
See the behavior behind the item, using payer, account, and transaction history to identify risk the image alone can’t reveal.
Stop high-risk items before the loss, with automated holds applied while the deposit decision can still change the outcome.
Add intelligence without replacing what works, layering behavioral scoring onto the OCR and image review already in place.
CheckDetect
Score every deposit for fraud.
Focus your team on the fraction that matters.
CheckDetect scores deposited checks in real time across teller, ATM, and mobile channels, using behavioral and transaction signals to identify more fraud while keeping the review queue focused. In one deployment, VALID identified 95% of fraud losses while flagging only 0.5% of items.
Tune the threshold to your operation
There is no single “right” alert volume for every institution. CheckDetect lets fraud teams set the cutoff that fits their risk tolerance, channel mix, and capacity for review, and adjust it as those conditions change.
| Items Flagged | Teller Fraud Identified (USD) | ATM Fraud Identified (USD) |
|---|---|---|
| Top 0.5% | 76.5% | 92.9% |
| Top 2.5% | 90.2% | 98.2% |
Results from a Top-10 North American bank, test period January to December 2025.
Results are specific to that deployment and threshold.
Catch more fraud with the team you have
Identify more fraud exposure
Prioritize the deposits most strongly associated with loss so analysts can act while the decision still matters.
Reduce false-positive friction
Apply holds and reviews more selectively instead of sending broad populations of good deposits into exception handling.
Increase automation
Drive automated holds by using high-accuracy fraud alerts that eliminate manual review.
Fit the institution
Adjust thresholds by operational capacity and risk appetite rather than accepting a fixed alert volume.
Proven across operating strategies
Each result reflects a specific client, scope, channel mix, threshold, and study period; results will vary.
Super-regional bank
58% reduction in mobile review rate and 51% improvement in loss rate, with roughly 300 analyst hours saved annually.
A top-10 North American bank
46% fewer false-positive holds.while identifying 55% more fraud exposure.
Top U.S. Bank
A mobile-deposit client saw an 87% decrease in check deposit losses within months of going live with CheckDetect.
Add precision to the strategy you already run
CheckDetect is designed to strengthen the fraud strategy and operations you already have. VALID works with each institution to connect the relevant deposit data, define thresholds and workflows, test performance, and move the scoring process into production.
Launch in 3–4 months
Depending on integration scope.
Score every channel
Across teller, ATM, and mobile.
Set your risk thresholds
Based on your risk strategy.
Timing varies with data readiness, integration scope, testing, and the institution’s review and approval process.
FAQ
It doesn't have to, but some FIs rely only on CheckDetect as their sole risk review process.
Can we control how many items are flagged?
Yes. Set and adjust the threshold to balance fraud exposure with your team’s review capacity. Fewer alerts mean a smaller queue, but potentially more exposure.
Which deposit channels are supported?
Teller, ATM, and mobile check deposits, with results measured across all three in current deployments.
What happens after CheckDetect identifies an item?
CheckDetect identifies and scores risk. The institution determines the action, such as review or hold, within its own policy and workflow.
How quickly can we launch?
See CheckDetect on your own deposit data
Bring us a recent sample of deposited checks and the outcomes your current process produced. We’ll show you where CheckDetect could improve fraud capture, reduce review volume, or both.