Latest / B2B SaaS Talks with Fexingo: Enterprise Software, Sales Cycles, and Procurement / Why Enterprise Software Buyers Now Demand a Business Continuity Audit
Transcript
- Lucas: We've talked a lot on this show about the new paperwork buyers are piling onto software deals — AI audits, data portability checks, vendor exit plans. But there's one request I'm hearing more and more that doesn't get as much attention: the business continuity audit. Luna: You mean, like, 'prove your system won't go down'? That's been a thing for decades. Lucas: It used to be a checkbox — 'do you have a disaster recovery plan? Yes or no.' Now buyers are asking for evidence. Specifically, they want to see your recovery time objective, your failover testing logs, and your contractual commitment to data availability. I have a concrete example from a few weeks ago. Luna: I'm listening. Lucas: A financial services firm — about $50 billion in assets — was evaluating three CRM platforms. One was a well-known legacy vendor, one was a fast-growing cloud-native player, and one was a smaller SaaS company I'll call 'CloudCRM' for now. The legacy vendor had a generic SLA that promised 99.9 percent uptime, but when the buyer asked for the actual business continuity plan, the vendor sent a PDF that was two years old. Luna: Ouch. That's an instant red flag for a regulated buyer. Lucas: Exactly. The financial firm's procurement team had a checklist with four specific requirements. First, a documented recovery time objective of four hours or less — meaning if the system goes down, it's back within four hours. Second, evidence of a live failover test within the past six months. Third, a data residency guarantee — they needed to know their data wouldn't leave a specific geographic region during a failover event. And fourth, contractual penalties if the vendor missed those targets. Luna: That's a lot. How did CloudCRM respond? Lucas: CloudCRM was actually the most prepared. They had built their entire infrastructure on AWS using a multi-region architecture — primary in us-east-1, failover in us-west-2. They had test logs from a quarter-end simulation that showed a 47-second RTO for read workloads and under three minutes for write workloads. They also had a clause in their contract that gave the customer a 10 percent service credit for any outage exceeding two hours. Luna: Forty-seven seconds is impressive. But that kind of architecture is expensive — can a mid-market SaaS company really afford to build that for every deal? Lucas: That's the tension. CloudCRM's CEO told me they spent about 18 percent of their engineering budget on resilience that year. But it won them that deal — worth about $2.5 million annually over three years. They beat the legacy vendor, which couldn't provide a current test report, and the cloud-native competitor, which only had single-region deployment and couldn't meet the data residency requirement. Luna: So the buyer walked away from the legacy vendor because of an outdated PDF? Lucas: Not just the PDF. The legacy vendor's sales rep actually said, 'Our uptime speaks for itself — we don't do scheduled failover tests because they risk customer disruption.' The financial firm's risk team interpreted that as: they don't have a tested recovery process. That killed the deal. Luna: Right. So from the buyer's perspective, it's not about punishing vendors — it's about forcing transparency. They want to know: if something goes wrong, are you actually going to restore my data? Lucas: And that's what makes this a procurement trend rather than just another IT checkbox. I've now seen similar requirements from healthcare companies, from e-commerce platforms processing payments, and even from one government contractor who asked for a 'chaos engineering' report — meaning they wanted proof that the vendor intentionally breaks things in production to test resilience. Luna: Chaos engineering as a sales requirement — that's wild. So what should a SaaS vendor do if they're getting these requests and they don't have a multi-region setup and a full-time SRE team? Lucas: First, don't try to fake it. Buyers are getting sophisticated — they'll ask for cloud provider console screenshots, third-party audit reports like SOC 2 Type II with the availability criteria, and they'll even schedule a call with your VP of Engineering. But you can start small. One vendor I know uses a managed database service with automated failover and a single secondary region in the same cloud provider. They documented a six-hour RTO and were transparent about it. The buyer accepted it because the vendor was honest and had a roadmap to improve. Luna: So honesty is actually a competitive advantage here. Lucas: Absolutely. And it's not just about the contract — it's about the sales process itself. I'm seeing procurement teams add a whole new meeting to their evaluation: the 'resilience review.' That's typically a 60-minute session where the vendor presents their architecture, their incident response playbook, and their post-mortem process for past outages. If you can't articulate that clearly, you're out. Luna: What about the small SaaS company that just can't afford an AWS multi-region deployment? Are they locked out of enterprise deals? Lucas: Not necessarily. Some buyers are willing to accept a 'shared responsibility' model — the vendor handles application-level resilience, and the customer handles their own backup and recovery at the data layer. But that requires a very clear agreement about who owns what. And the vendor still has to prove that their application can recover in a reasonable time. There's no free pass. Luna: So the bar is rising. What's the one thing a SaaS startup should do this quarter if they want to be ready for these audits? Lucas: Run a failover test. Seriously. Take your staging environment, simulate a regional outage, and time how long it takes to come back online. Document it. Then do it again next quarter. That alone — having a test log — puts you ahead of maybe 60 percent of vendors. And if you can do it in a production-like environment, even better. Luna: And for buyers listening, what should they be asking for that they might not be asking for now? Lucas: Ask for the vendor's 'maximum acceptable downtime' in writing. Not just uptime percentage — actual minutes or hours. Then ask for the last three incidents where that threshold was tested. If the vendor hesitates, or says 'we don't track that,' you've learned something. Also ask for their data recovery point objective — how much data they'd lose in a worst-case scenario. That's often more revealing than the uptime SLA. Luna: So it's not just 'will it be up?' but 'when it goes down, how bad is it?' Lucas: Exactly. And that mindset shift — from uptime to recovery — is what's driving this whole trend. The old world was about preventing failure. The new world is about surviving failure gracefully. Luna: Yeah, absolutely. If these episodes are helping you navigate procurement changes or build better sales processes, it really does make a difference. We're at buy me a coffee dot com slash fexingo — that's buy me a coffee dot com slash f-e-x-i-n-g-o. Lucas: And to be clear, no pressure ever. We love making the show regardless. But if you find value and want to keep it accessible to everyone, that's where you can help. Now, back to resilience reviews — one more thing I think is worth flagging. Luna: What's that? Lucas: The business continuity audit isn't just a one-time document. Some buyers are now asking for annual recertification. So if you passed this year, you still need to maintain that testing cadence. I know one vendor who lost a renewal because their failover test in year two used outdated infrastructure — they'd migrated to a new cloud region and forgotten to update the recovery plan. Luna: That's a painful lesson. So the audit is really a continuous process, not a checkbox. Lucas: Exactly. And I think within the next 18 months, we'll see industry-specific standards emerge — like a 'finra style' resilience requirement for fintech SaaS, or a hipaa aligned continuity standard for health tech. The smart vendors are already building their testing cadence now. Luna: Makes you wonder: is there any enterprise software deal where this just doesn't apply? Lucas: Maybe for a pure internal tool with no customer-facing data. But even then, if the tool goes down and blocks your sales team from quoting, that's revenue impact. I think eventually it'll be table stakes for every B2B SaaS deal above a certain size. Luna: Well, that's a good lens to end on. Thanks for walking us through it. Lucas: Thanks, Luna. And thanks to everyone listening. See you next time.