Every vendor pitch sounds the same at the start: get up and running in a day, no engineering required, focus on your business instead of your infrastructure. That pitch is often true, and for a long stretch of a company's growth, building on someone else's platform is genuinely the right call. The problem shows up later, quietly, when the business has grown around the platform's constraints, the pricing has scaled past what anyone budgeted for, and leaving would mean rebuilding years of accumulated workflow rather than just switching tools.
Why Lock-In Is Easy to Miss While It Is Happening
Vendor lock-in rarely arrives as a single decision. It accumulates through dozens of small, individually reasonable choices: a workflow built around a platform's specific automation rules, a data model shaped by what the platform's fields support, an integration that only exists because the vendor built it, a team that has spent two years learning the platform's specific interface instead of transferable skills. None of these choices looks like a risk in isolation. Together, they describe a business that cannot leave without a project large enough to require its own budget line.
Where the Cost Actually Shows Up
Pricing That Scales With Your Success, Not Your Usage
Per-seat and per-record pricing models that felt trivial at ten users and a thousand records become a material line item at two hundred users and a million records, and the vendor knows this growth curve better than the customer does when they sign the initial contract. The platform did not get more expensive to run at that scale. The pricing model was built to capture more value as the customer succeeds, and by the time it is expensive, switching costs have grown right alongside it.
Feature Requests That Depend on Someone Else's Roadmap
A capability the business genuinely needs, a specific integration, a workflow rule, a reporting view, either exists on the vendor's roadmap or it does not, and a single growing customer rarely has the leverage to change that roadmap's priorities. Businesses running on a platform they do not control end up designing their operations around what the vendor has decided to build, rather than what the business actually needs.
Data That Is Technically Exportable and Practically Unusable
Most platforms allow data export, satisfying the letter of "your data is yours," while the actual structure, relationships, and business logic embedded in how that platform organizes information do not export cleanly, or at all. A CSV of records is not the same as the workflow, permissions, and automation logic built around them, and rebuilding that logic on a new system is where the real migration cost lives.
Outages and Deprecations Outside Your Control
When core business operations run on a third-party platform, that platform's downtime, pricing changes, feature deprecations, and even acquisition by another company become the business's operational risk, with no ability to fix, delay, or opt out of the change. A vendor's decision to sunset a feature the business depends on is not negotiable from the customer side.
Making the Build vs Buy Decision Honestly
Platforms Win for Undifferentiated, Fast-Moving Needs
For functions that are not core to the business's competitive advantage, and where speed to launch matters more than long-term control, a mature platform is usually the right call. Email infrastructure, payment processing, and generic CRM functionality rarely benefit from being custom-built, because the platform's economies of scale and specialization outweigh the loss of control.
Custom Build Wins When the Workflow Is the Business
When the specific way a business operates, its client workflow, its pricing logic, its operational process, is actually the source of its competitive advantage, building that on someone else's platform means competing with one hand tied to a vendor's constraints. The businesses that grow past their platform limitations are usually the ones whose core differentiation was never something a generic tool was built to support in the first place.
Evaluate Exit Cost Before Signing, Not After
Before committing to a platform for anything operationally important, ask what a genuine migration away from it would look like in a year, in three years, and at ten times the current scale. A platform that looks cheap today and prohibitively expensive to leave in three years is not actually cheap, the cost has just been deferred to a point where the business has less leverage to negotiate or walk away.
Finding the Middle Ground
The choice is rarely a clean binary between fully custom and fully platform-dependent. The businesses that navigate this well use platforms for genuinely undifferentiated infrastructure while building custom systems around the specific workflows that define how they actually compete, and they make that split deliberately rather than by accumulated accident.
MAPL TECH helps growing businesses build the custom systems that protect their competitive advantage, without over-building where a platform genuinely makes sense. Get in touch to talk through where your business should build versus buy.
