What an integration scope should cover before you agree
Learn what an integration scope document must cover before you hire—systems, data flow, error handling, inclusions and deliverables explained.
The most common mistake buyers make is agreeing to start work without a clear scope—then discovering halfway through that the developer has a different picture of what "done" looks like. One person thinks they're connecting two systems end-to-end; the other thinks they're writing one authentication module. Scope creep, miscommunication, and unexpected costs follow. Before you book anyone, you need a scope document that both you and the developer can point to and say "yes, this is what we're building."
A scope worth signing isn't fancy. It's specific. It names the systems being connected, the exact data flowing between them, who triggers what, and what happens when things go wrong. It lists what the developer will and won't do. It sets boundaries on your expectations and theirs.
The systems and the flow
Start by naming the two (or more) applications you're connecting. Not "our accounting system"—the actual software: Xero, SAP, a custom database, whatever it is. Then describe the data journey. Which information moves from where to where? If you're connecting your e-commerce store to your accounting software, the scope should say: "Customer orders from the shop automatically create invoices in the accounting system. The developer will map product SKU, quantity, price, and customer email. Tax and shipping calculations will be handled by the shop platform and passed as totals."
Be concrete about triggers. Is the sync real-time or nightly? Does it run on a schedule, or does a person click a button? If something fails—a timeout, a duplicate record, a missing field—what happens? Does it retry automatically, email you, or pause and wait for manual intervention? A good scope covers the happy path and the error paths.
What's included and what isn't
This is where misunderstandings live. Spell out:
- The developer will write the integration code and test it in a staging environment.
- The developer will not redesign your database, migrate historical data, or train your team.
- The developer will handle API authentication but will not manage your API keys or credentials.
- The developer will not modify the third-party application (e.g., they won't customize Xero itself).
- Testing will cover normal scenarios and documented error cases, but not load testing at scale.
You're not being pedantic. You're protecting both of you. A developer who agrees to "integrate our CRM" with no boundaries might end up spending weeks troubleshooting your data quality, which wasn't their job. You end up paying for work outside the scope. Ask the developer to write these boundaries down before you hire them.
Deliverables and handover
What do you actually receive when the work is done? The scope should list: the integration code (in your repository or theirs), documentation of how it works, a list of API endpoints used and their authentication method, and instructions for monitoring or restarting it if it breaks. Some developers hand over a running system; others hand over code and say "deploy it yourself." Make sure you know which.
Also clarify what happens after launch. Is the developer available for a week of fixes if bugs emerge? Or are they done the moment you sign off? If the third-party API updates and breaks the integration, is that the developer's problem or yours? The scope can't predict every scenario, but it can set expectations about support.
Before you agree
Read the scope carefully. Look for vague words like "optimize," "streamline," or "ensure compatibility." Ask the developer to rephrase anything that isn't testable. A good scope lets you check off each item when it's done.
Then ask: "If we need something outside this scope, how do we handle it?" The answer should be a change request process, not an argument about what was always implied.
Once the scope is solid, you're ready to hire with confidence. Look for a developer who's clear about boundaries, willing to write things down, and experienced with the specific systems you're connecting. Strove's verified freelancers can show you examples of past integration work and explain their scoping process upfront.
Common questions
- What happens if we discover we need something after we've agreed the scope?
- That's normal. The scope should include a change request process—how you ask for additions, how the developer estimates the cost and timeline, and how you both agree before work starts. This protects everyone from unexpected bills and delays.
- Do I need a lawyer to review the scope?
- Not necessarily. A clear, straightforward scope written in plain language is often better than legal jargon. What matters is that both you and the developer understand it the same way. If your organisation requires a contract template, your lawyer can review it, but the scope itself should remain simple and specific.
- Should the scope include testing and quality assurance?
- Yes. Specify what testing the developer will do (functional testing in a staging environment, for example) and what you'll do on your end. Be clear about acceptable error rates or performance thresholds. This prevents disputes about whether the work is finished.
- Can the scope change once work has started?
- Yes, but changes should go through a change request process. The developer estimates the extra time and cost, you agree, and the timeline adjusts. Avoiding this process is how projects spiral over budget.
Find a verified provider on Strove
Compare vetted api integration providers, check their credentials, and book or request a quote — all in one place.
Find a Business