Import.io platform
Fully managed service
Also called: Managed service, managed web data
An arrangement in which the provider builds, runs, quality-checks and maintains data collection and delivers the data under an agreement.
In practice
At Import.io it starts with a schema contract and runs with daily QA and an SLA.
Agree who owns each task
In a managed data engagement, the useful distinction is responsibility, not the label on the proposal. Specify who onboards sources, defines the schema, approves samples, operates collection, investigates failures and maintains extraction when pages change. Identify what the customer still owns, including approved uses, destination access and downstream business decisions.
Example: accepting a delivery
For an illustrative daily catalogue feed, acceptance criteria might require a product identifier, valid currency, a recent observation timestamp and a documented treatment of unavailable products. Agree whether a failed batch is withheld, partially delivered or sent with a warning. A provider running the scraper is not, by itself, an agreement about complete and usable records arriving at the destination.
What belongs in the scope
Document coverage, cadence, delivery format, support contacts, escalation and change-control procedures. Define how new sources, schema changes and historical backfills are priced and approved. Service availability, data freshness and record correctness are different commitments and should be measured separately. Compare the actual proposal and exclusions rather than treating every service described as managed as equivalent.
At Import.io it starts with a schema contract and runs with daily QA and an SLA.
Import.io products →