Lenders think strategically about business continuity for their core banking systems. Core, origination, servicing and payments all have recovery time objectives (RTO), recovery point objectives (RPO), backup environments and annual tests. Yet business continuity for recovery case management is often overlooked entirely. Why?
In recent months, we’ve seen recovery agents taking to industry social media groups to ask some version of the same question: "Is anyone else down?" Slow screens, integrations that stop passing data, and platforms that go offline for hours at a time.
Every technology platform has bad days. But when case management fails, the consequences land on the lender:
- Wrongful repossession risk. If a borrower pays, files bankruptcy or is placed on hold, and that update cannot reach the agent, a vehicle that should have been left alone can end up on the hook.
- Recoveries that go unrecorded. If an agent cannot mark a unit on hook or stored, the lender does not know it has the collateral. Notices, storage, condition reports and remarketing all start late.
- Lost time and money. Agents sit idle, storage fees accrue, and staff spend days reconciling what happened while the platform was down.
- Reputational risk. A wrongful repossession or a missed notice becomes a consumer complaint, a regulator inquiry, or a post that follows your brand online.
For many lenders, recovery case management has quietly become a single point of failure: one platform, one network, no fallback. The question is not whether that platform will ever go down. It is what your operation does when it does.
The blind spot in the disaster recovery program
Lenders treat recovery case management as the vendor's problem, but the risk lands on the lender's balance sheet and in their compliance file.
Recovery case management rarely makes the disaster recovery inventory, usually for one of three reasons:
- It feels outsourced. The platform belongs to a network or forwarder, so continuity feels like their job.
- It is "low volume." Repossessions are a small share of the portfolio, so the function gets tiered low.
- Outages are short. Most disruptions last hours, so no one has felt the full cost yet.
But consider what actually stops when case management stops:
- Custody of collateral. Vehicles are recovered but not recorded, so storage, condition reports and remarketing all start late.
- Legal clocks. State notice and redemption timelines run from the date of repossession, not the date the system catches up.
- Personal property and consumer complaints. Inventory and release requests have nowhere to go.
- Vendor payments and audit trail. Invoices, fees and the record your examiners will ask for later.
- Depreciation and remarketing losses. Every day a vehicle waits to reach auction, it loses value. That cost is often what makes continuity a priority.
That is not a low-tier function. That is regulated, customer-facing work with deadlines attached, and at most lenders it runs through one system with no backup. Anywhere else in the bank, we would call that a single point of failure.
The single point of failure you can't see
At least one of this fall's disruptions can be traced back to the platform's cloud hosting provider. This also illustrates something every vendor management team should sit with:
- Your recovery vendor has vendors. Hosting, mapping, LPR data, payments, messaging. Any of them can take your recovery pipeline offline.
- You cannot see that chain unless you ask. Most lenders do not know where their recovery platform is hosted or whether it has a failover region.
- Concentration compounds it. Industry estimates suggest a single network receives roughly 90% of U.S. repossession assignments, and several of the most widely used agent tools share one corporate parent. When that much of the industry runs through so few systems, one upstream incident becomes an industry-wide event.
Regulators already expect banks to plan for this. The Interagency Guidance on Third-Party Relationships calls for contingency planning around critical vendors, and the FFIEC Business Continuity Management booklet asks institutions to understand how third-party disruptions affect critical operations. Recovery belongs in that conversation.
What a disaster recovery plan for case management looks like
A recovery disaster recovery plan uses the same building blocks as any other system plan; it just has to be written down.
- Set an RTO. How many hours can recoveries go unrecorded before it creates compliance or customer harm? For most lenders, the honest answer is "less than a business day."
- Set an RPO. How much case history can you afford to lose or re-key? A nightly spreadsheet export is a one-day RPO, not a backup.
- Remove the single point of failure. Your open assignments, statuses and documents should also live in a system you control that stays in sync, not only in one external network.
- Define an alternate path. If the primary platform is down, how do assignments, updates and invoices move? Who declares the failover, and who tells the forwarders and agents?
- Plan the comms. Agents found out about these outages before most lenders did. Decide who on your team hears first and what your vendors are told.
- Test it once a year. Ask your vendors for their DR test results, and walk through your own failover on paper at a minimum.
Questions for your next vendor review
These are fair questions to ask any recovery technology provider, including us.
- Where is your platform hosted? Can the host potentially be a single point of failure, or do you have a failover region or provider?
- What are your contractual RTO and RPO, and what happened to those numbers during your last incident?
- How do you notify lender clients during an outage, and how quickly?
- Can I get a current copy of my case data on demand, in a usable format?
- When was your last disaster recovery test, and can I see the results?
- Do you hold a current SOC 2 Type II report, and does it cover availability?
- When was your last independent penetration test, and were all critical and high findings remediated?
- If you are down for two days, how do my forwarders and agents keep working?
The plan you write before you need it
Recovery has moved from a back-office afterthought to a regulated, data-driven operation. Its continuity planning should catch up. No lender would accept a single point of failure in its core system. It should not accept one in recovery either.
The next outage will not be announced in a press release. It will show up as a Facebook post from an agent asking, "Anyone else down?" The lenders who handle it well will be the ones who already decided what happens next.
At MBSi, business continuity for recovery case management is something we think about every day. If you would like to compare notes on building a recovery DR plan, reach out at mbsicorp.com.
