A documented API
REST or GraphQL with authentication and reasonable rate limits. Straightforward, and the case where a fixed price is easy to give.
Services · integration
Every company has one: the industry-specific application that runs a core process, has four thousand customers worldwide, and appears on nobody's integration marketplace. It is usually the system holding the data that would make everything else useful.
Tell us what you need connected and what it exposes. We will tell you if it is feasible.
The situation
In descending order of preference. We will tell you which applies to yours before quoting.
REST or GraphQL with authentication and reasonable rate limits. Straightforward, and the case where a fixed price is easy to give.
Many industry applications have an API their vendor does not publish. Asking usually works, and it is worth a phone call before assuming otherwise.
A read replica or a read-only user against the application database. Effective, faster than an API, and requires care because the schema is not a contract.
A nightly CSV or fixed-width drop to SFTP. Unglamorous, entirely reliable, and the right answer more often than people expect.
Last resort, and we will say so. It works, it breaks when the vendor redesigns a page, and we price the maintenance honestly rather than pretending it is stable.
Whichever path is used, a daily check compares counts and control totals between source and graph. The extraction method changes; the verification does not.
The failure mode in custom integration work is a fixed price given before anyone has looked at what the system actually exposes. The engagement then discovers in week six that the API paginates at fifty records with a two-request-per-second limit, and the conversation becomes a change order.
So we scope first. A short paid discovery — usually $1,500 to $3,000 — establishes what is reachable, at what grain, at what refresh rate, and how it will break. The fixed price comes after that, and the discovery findings are yours whether or not you proceed.
Not age. Some twenty-year-old systems have excellent, stable database schemas and are easier to work with than a modern SaaS product with an aggressive rate limit.
What makes it hard is a lack of stable identifiers, no way to detect what changed since last time, no bulk access, and a vendor who treats the data as theirs. That last one is a commercial obstacle rather than a technical one, and it is the only category we sometimes cannot solve.
If the data is used monthly by two people and the integration would cost $18,000 to build and $400 a month to maintain, we will say that a scheduled export into a spreadsheet is the better answer.
We turn away this work reasonably often. An integration that cannot justify its maintenance gets neglected, and a neglected integration silently produces wrong numbers, which is worse than the manual process it replaced.
Building it is a one-off. Keeping it working through API versions, schema changes, credential rotations, and vendor redesigns is ongoing, and it is the line most integration proposals leave out.
Ours is $150 to $500 a month depending on the extraction method, stated up front. Browser automation sits at the top of that range for the obvious reason.
Where to start
$1,500 to $3,000, one to two weeks. What is reachable, at what grain, how it will break. Findings are yours regardless of what you decide.
Build price and monthly maintenance, both stated. If the answer is that it should not be built, that is in the report too.
Extraction, mapping into the graph, and a daily reconciliation that is independent of the extraction code rather than sharing its assumptions.
Failures alert us before you notice. Version changes and vendor redesigns are covered by the maintenance fee rather than billed as change requests.
Questions
We will tell you what is reachable, what it costs, and whether it is worth doing.