Fragment deployments require custom context

This reveals that Fragment is closer to a services heavy automation company than a pure software seat business. Each deployment has to learn a customer’s vendor rules, approval paths, ERP setup, and exception patterns, so the real unit of work is not just processing an invoice, it is building the operating context that lets the model act safely. That makes gross margin depend on how much of onboarding, integration, and exception handling can become repeatable across customers.1 2 3 4 5

• In AP automation, integrations are a large part of the product, not a one time wrapper. Tipalti sells pre built ERP connectors and also supports custom integrations, while Rossum offers both self serve connectors and solution architect support, which shows that even mature vendors still spend real effort connecting each customer’s stack.2 3
• Customer differences are operational, not cosmetic. Rossum case studies show invoice queues, approval flows, master data checks, duplicate detection, and ERP mappings being configured around each customer’s process, and one enterprise rollout took about five months from October 2024 to March 2025 after scope refinement.4 5
• That is why savings for buyers do not automatically mean software like Fragment scales with near zero delivery cost. Finance automation competitors from Zip to Tipalti and Rossum all win by fitting into messy existing systems, and that fit work is exactly what can absorb margin until workflows become standardized enough to reuse broadly.1 2 3 6 7 8

The next phase of the market favors companies that turn custom deployment work into packaged playbooks for specific ERP, procurement, and approval environments. If Fragment can shrink implementation from custom engineering into reusable modules, it can keep outcome based pricing while making each new customer look more like software revenue and less like a consulting project.2 3 4 5