Between Product and Partnerships

Integrations look deceptively simple until they become the backbone of your business. In this episode of Between Product and Partnerships, Pandium CEO Cristina Flaschen sits down with Scott Lavery, Senior Product Manager at Arkestro. They unpack what really happens when integrations shift from a "nice to have" feature to something the company can't function without. Scott shares hard-earned lessons from a decade in B2B SaaS, covering sectors from martech to procurement. He discusses the headache of inheriting messy stacks and why iPaaS tools often hide long-term costs. The conversation also explores how integration work fundamentally changes what it means to be a product manager. Together, they dig into common failure modes and the tough tradeoffs junior PMs face when they’re "volun-told" to own integrations. Who we sat down with Scott Lavery is a Senior Product Manager at Arkestro. With over ten years of experience in B2B SaaS, he has repeatedly found himself responsible for integrations, often without ever intending to specialize in them. Scott brings expertise in: Unwinding complex iPaaS-driven environments.Designing integrations built to be "set and forget."Managing third-party dependencies alongside specific scale constraints.Advocating for pragmatic, cost-aware strategies. Key Topics Why integration PM work is fundamentally different  Integration success is defined by invisibility. Unlike standard features, value is found in reliability and trust rather than how often a user clicks a button. The hidden costs of low-code and iPaaS tools  Teams often end up writing code blocks inside "no-code" tools. We discuss how pricing models can distort architectural decisions and where velocity eventually hits a wall. What to do when you inherit a messy integration stack  Practical advice for PMs walking into undocumented systems filled with inherited workflows and vendor dependencies they can’t control. Episode Highlights 01:48 - How most PMs “fall into” owning integrations03:58 - Why integration metrics flip traditional product thinking on its head06:31 - Contextual success metrics: Why volume is not the same as value08:21 - Navigating ecosystems without becoming a domain admin11:18 - Why API docs lie and customers ignore your design intent15:37 - Warning signs of an unhealthy iPaaS environment19:05 - Silent failures and the pain of hearing about outages from customers23:45 - The code-block paradox in low-code platforms31:52 - Scott’s playbook for PMs inheriting integrations Key Takeaways Great integrations are designed to disappear  Successful integrations are rarely touched after the initial setup. In this space, reliability is a far more important metric than user engagement. Metrics are contextual, not universal A monthly sync can be just as vital as one that runs every five minutes. Frequency alone does not signal success. You can’t abstract away real-world usage  API contracts rarely reflect reality. No tool removes the need to understand how customers actually use systems like NetSuite or Salesforce. Low-code tools often trade speed for long-term pain  Teams save time early but spend years optimizing around pricing models and managing fragile logic. Inherited workflows is a scalability risk  If only one person understands the system, it is already brittle. This is a massive liability once customers are live. Silent failures erode trust fastest  Learning about outages from customers is a major failure. Proactive monitoring and clear communication are bas