Know who you’re onboarding, before you say yes
Merchant Risk Queue
Live
1,284
Assessed
37
In review
3 min
Median decision
Ranked by risk score
Northgate Supplies
RISK 84
Bluepeak Retail
RISK 62
Vantage Foods
RISK 41
Orchard Labs
RISK 22
Zoven builds a full risk picture of every merchant, scores it, and tells your team what to do next. Clean cases clear fast, and the reasoning stays on file.
Trusted by risk and compliance teams
Underwrite more merchants, with fewer surprises
The point of a risk check is a confident decision. Zoven does the gathering and the scoring, so your team spends its judgement where it counts.
Clear the easy ones fast
Low-risk merchants come back with a clean score and a ready recommendation, so approvals don’t pile up.
Minutes
from application to a first decision
Catch the risky ones early
Adult content, prohibited goods, sanctioned regions, and shell-like setups get flagged and pushed to the top of the queue.
120+
risk signals read per merchant
Keep watching after go-live
Approved merchants are re-scanned on a schedule. When something changes, it opens a case, not a blind spot.
Always on
re-scans on your cadence
One profile, one score, one next step
Zoven pulls a merchant’s whole story into a single profile, turns it into a risk score from 0 to 100, and pairs that score with a clear onboarding recommendation. The reasoning stays attached, so a decision made today still makes sense when someone reads it back next quarter.
Northgate Supplies
RISK 84
84
Critical · EDD required
Domain & security
72
Business identity
88
Compliance
54
Contact & social
41
Adverse media
63
120+ signals per merchant
Domain, identity, compliance, contact, and adverse media, gathered across five areas.
No manual data pulls
Point Zoven at a URL or a name and the profile assembles automatically, every time.
A recommendation, not just a number
Each score comes with a plain call: approve, review, run EDD, or decline.
The reasoning stays visible
Every factor behind a score is on the record, so any decision is easy to defend.
From a URL to a decision, in three steps
Point Zoven at a merchant and the assessment runs on its own. Your team only steps in where judgement is needed.
Submit the merchant
Send a URL or business name through the API, the dashboard, or a bulk upload. No manual data gathering.
01
API, dashboard, or bulk
Zoven scores it
The engine gathers 120+ signals across five areas and turns them into a risk score with the reasoning attached.
02
scored automatically, end to end
Your team decides
A clear recommendation lands the case in a queue. Approve, review, run EDD, or decline, with a full audit trail.
03
review and sign off with maker-checker
The checks that back the decision
When a merchant needs a closer look, run the identity and business checks right from the case, without switching tools.
Tax ID verification
Validate PAN / EIN format and holder details.
Sales tax registration
Retrieve GST / sales-tax registration and business data.
Bank account verification
Confirm account validity and metadata.
Corporate registry lookup
Look up corporate identification numbers.
Personal ID verification
Lightweight identity verification.
Small business registration
Verify MSME / SBA registration details.
Mule account screening
Screen accounts, phones, emails, and UPI / ACH IDs.
Litigation records
Search NYSCEF records by company or owner.
Available checks vary by market and plan. KYB and vKYC onboarding flows are part of the same toolkit.
Connect it to how you already work
Submit merchants and pull results through the channels your team and systems already use.
REST API
Submit an assessment and fetch the result by ID.
Bulk upload
Score a whole portfolio from a single spreadsheet.
Dashboard
Submit and review merchants without writing code.
API keys
Secret keys for your servers, publishable keys for the browser.
Identity & business data
Identity, tax, registry, and bank records under the hood.
IP & device intelligence
Geolocation and proxy or VPN signals on every check.
Court & watchlist records
Adverse media and court-record lookups built in.
Government document fetch
Government-backed document retrieval.
Available data and channels vary by market and plan.
Re-scan · Bluepeak Retail
LOW → HIGH
MCC changed to high-risk
Insight
Prohibited keyword appeared
Difference
SSL still valid
Unchanged
Onboarding is day one, not the whole job
Approved merchants don’t stay the same. Zoven re-scans them on a schedule you set, compares each scan against the last, and turns any meaningful change into a case for your team.
Set the cadence
Re-scan weekly, monthly, or quarterly.
Scope the scan
Re-scan only the modules you care about.
Sorted changes
Every change split into insights, differences, and unchanged.
Alerts to cases
Each alert opens a case: dismiss, escalate, or offboard.
A platform built for a risk desk
The parts that make Merchant Risk fit how your team already works, not the other way round.
AI risk agent
The agent reads the signals Zoven collected, writes up the assessment, and cites what it used. Reviewers open a case with the reasoning already laid out.
Drafts the written assessment automatically
Cites the exact signals behind each point
Leaves the score and final call to your team
Assessment draft
Northgate Supplies
HIGH
Sanctioned-region exposure
Registered address and payment routing trace to a jurisdiction on the sanctions blacklist. Recommend EDD before onboarding.
Business identity · Compliance
MEDIUM
Young domain, thin contact
Domain registered 41 days ago with no verifiable phone. Common in shell setups, though not conclusive on its own.
Domain & security
Draft ready for review
6 signals cited
Rules & thresholds
Set the bands and weights that match your risk appetite, so scores and recommendations reflect your policy.
Score bands
Low
0–25
Medium
26–50
High
51–75
Critical
76–100
Recommendation per band
Low
Approve for standard onboarding
Medium
Send for analyst review
High
Enhanced due diligence required
Critical
Decline and flag for compliance
Case workflow
A scored merchant opens a case. A maker reviews it at L1, a checker signs off at L2, and every action lands in the audit trail. Anything unclear returns to L1.
Risk assessment run
Northgate Supplies · scored 84
L1 review · maker
Risk Assessment L1
Approve for review
Reject
L2 review · checker
Risk Assessment L2 · awaiting sign-off
Approve
Reject
Return to L1
Case closed
Decision recorded to the audit trail
Catches what a checklist would miss
The depth behind the score: the checks that separate a real merchant from one dressed up to look like one.
Judges where a URL really lands, so a silent switch to another domain gets caught
AI vision checks premise photos for a movable setup, a faked visit, or a mismatch with the map listing
Inspects the actual TLS certificate for expiry, self-signing, and a broken trust chain, not just for HTTPS
What a missed merchant really costs
A risk check is not paperwork. The merchant you wave through is the chargeback, the fine, or the name in the news you did not see coming.
The risky ones look fine on paper
A clean-looking site can hide a movable shop, a silent redirect, or a category nothing like the one that was declared. The tell is in the details, not the application form.
Risk does not stop at sign-up
A merchant that scored clean today can drift into something else next quarter. Onboarding is one moment; the exposure lasts the whole relationship.
Manual review cannot keep pace
Every merchant deserves the same rigour, but a person reading sites one by one cannot give it to a thousand of them. Something always gets skipped.
Good to know
How is the risk score calculated?
Zoven reads dozens of signals across five areas, then rolls them into a graduated score from 0 to 100. Any critical factor, such as adult content or a sanctioned region, moves a merchant straight to the top. Lesser factors add up within their band. The result maps to Low, Medium, High, or Critical, each with its own onboarding recommendation.
What kinds of merchants can it assess?
Does it keep watching merchants after they’re approved?
Can our team review before a decision is final?
How does the AI risk agent fit in?
Bring a handful of real cases to a walkthrough and watch them score, sort, and land in a queue.