
An analyst at a credit union is looking at a card that was just used three times in a row somewhere it has never been used before. The alert and evidence are in Unit21. The card needs to be blocked before a fourth charge goes through.
So the analyst opens a second system, finds the member and the card, and blocks it by hand. That gap between seeing fraud and stopping it is where money gets lost, and mistakes happen.
The Unit21 Corelation integration closes that gap. It brings member data into Unit21 for fraud and BSA/AML monitoring and lets analysts act back in KeyStone without leaving the case. Here is how it works, what it changes for your team, and how to get it set up.
The integration has two halves: data coming in from KeyStone, and actions going back out.
Corelation builds and supports a dedicated extract for Unit21, called the Unit21 AML Extract. Your own KeyStone administrator or data team configures it inside your environment, so there are no custom extracts for anyone to build or maintain.
KeyStone data lands in the same Unit21 workspace as your other sources, linked to one member record. Joint ownership comes with it, so an alert on one member shows every account and card that member touches.
This then saves the analyst from opening a second system. From inside a Unit21 alert or case, next to the evidence that justified the decision, an analyst can:
Unit21 authenticates to KeyStone through Corelation's KeyBridge API, sends the request, and records the result in the case: what was sent, what came back, who clicked, and when. If KeyStone rejects a request, the analyst sees it in the case instead of finding out later.
The point of the integration is that a case never has to leave Unit21 to be resolved. A typical one runs like this:
Most credit unions miss catching fraud due to the lag and friction of manually jumping between multiple systems. When blocking a card or placing a hold takes one click instead of a second login and a manual lookup, teams eliminate significant friction and delays, accelerating response times when catching fraudulent activity.
It also changes what you can show an examiner. The record of what you did starts the moment the analyst acts, tied to the alert and evidence behind it, instead of whenever someone remembers to note it somewhere else.
Directions Credit Union, a $1 billion credit union with 23 locations across Ohio and Michigan, runs Unit21 on Corelation. It replaced nightly batch detection and a homegrown system with rules its own team builds and tunes.
"I have true control of our fraud strategy and systems. Unit21's customizability allows me to create and test a rule in one day and deploy it that very same day."
If your credit union runs KeyStone, the integration is ready to set up:
Data from KeyStone arrives about every five minutes, which is near real time and fast enough for most fraud and BSA/AML monitoring. For decisions inside a member's digital banking session, such as a login or a payment, Unit21 also connects to digital banking platforms and returns a decision in under 150 milliseconds.
Your KeyStone administrator or data team configures it in your environment, with Corelation's support and Unit21's implementation team. You don't need to build or host a separate data pipeline.
Today, analysts can block a card, activate a card, place a stop payment, and place a check hold. Each action is sent through the KeyBridge API and logged in the Unit21 case with the request, the response, and the user who took it.
Yes. The same KeyStone data powers fraud rules, AML transaction monitoring, case management, and SAR and CTR filing, so one team or two can work from one platform.
Talk to your Unit21 or Corelation representative to get started, book a demo, or meet us at the Corelation Executive Forum, October 21 to 23. You can also read the full announcement in our press release.
Read the full announcement in our press release to see the depth of the Unit21 + Corelation integration.

Kunal Datta is the Chief Product Officer at Unit21. Prior to Unit21, he led the Product team for Checkout at Fast, and prior to that, led the Product teams responsible for automating aerial wildfire safety inspections at Pacific Gas & Electric.
He has a background leading Product teams using AI to automate processes at regulated entities, as well as financial products, machine learning products, web applications, mobile applications, hardware products, and data products. Kunal is a Fulbright Scholar and studied Civil and Environmental Engineering and Music Science Technology at Stanford University.