Services / Impact Analysis
We map every new rule to your own policies.
Upload your compliance policies once; every new rule from then on gets scored against them automatically.
Why this matters.
Knowing a rule changed is half the battle — knowing whether your policies are affected is the other half. That second step is the one that takes time. A compliance officer reads the CFPB bulletin, understands it, then has to hold it against every relevant policy in their library and judge: does section 3.2 of our loan origination policy still cover this? Does section 5.1 of the servicing manual need an update?
That judgment call happens against a backdrop of thirty other items that came in the same week. Most compliance teams get to it eventually. Some don't get to all of it. And because the miss is invisible — you don't know what you didn't read — it tends to surface at exam time rather than at the desk.
What this looks like.
Overdraft fee safe harbor change affects your Fee Disclosure Policy
Source rule: CFPB · Final rule · 12 CFR 1026.4(b)(11)
Matched policy: Fee Disclosure Policy §3.2 (Overdraft fees)
Why this matched
Your §3.2 specifies a $5 cap on overdraft NSF fees consistent with the prior safe harbor. The amended rule raises the cap to $5.50 effective 2026-01-01. Update the cap and the effective-date references.
How we do it.
Closed compliance taxonomy
We maintain 25 canonical compliance tags — TILA, RESPA, fair lending, servicing, flood insurance, and so on. When you upload a policy, we tag it against this taxonomy. When a new regulatory item lands, we tag it the same way. Matching happens at the tag level first, so the AI impact scoring is focused on genuinely relevant pairs rather than checking every policy against every item.
Automatic fan-out on both sides
Impact runs in two directions. When you upload a new policy, we score it against the last 90 days of matching regulatory items. When a new regulatory item lands in the feed, we score it against every policy in your workspace that shares at least one tag. You don't trigger anything — it runs in the background.
Severity scoring with rationale
Each finding is scored CRITICAL, MATERIAL, or MINOR — and includes a rationale explaining why. CRITICAL means the rule directly contradicts or supersedes existing policy language. MATERIAL means a policy update is likely warranted. MINOR means awareness is enough. NONE results are not persisted — only findings that register impact show up in your queue.
What's in the product today.
Per-finding severity score: CRITICAL, MATERIAL, or MINOR — based on how directly the rule overlaps your policy text
Workspace-wide triage queue showing all open findings across all policies
Dismiss or resolve findings in one click; status survives prompt-version refreshes
Fan-out runs automatically when you upload a new policy or when a new regulatory item lands
Impact is scored against the last 90 days of relevant items on upload — no retroactive blind spots
Hard cap of 25 findings per fan-out run to keep wall time predictable; highest-severity findings are scored first
Each finding records the policy section, the regulatory item, the AI rationale, and the full prompt snapshot for auditability