Do we have to replace our legacy system to modernize it?
Not necessarily. If the system still does its job reliably, the more practical route is often to leave it running, put a modern interface in front of it and move functions out gradually. Replacement makes sense when the system can no longer be maintained, cannot meet a real requirement or carries risk that wrapping does not remove. We assess which applies before recommending either.
What is the difference between wrapping, migrating and replacing?
Wrapping puts a new API or integration layer in front of an existing system without changing it. Migrating moves specific functions or data to new components in stages, while the old system keeps running. Replacing retires the old system entirely. Many projects combine them: wrap first, migrate the parts that cause the most trouble, and replace only what has to go.
Can you connect a system that has no API?
Sometimes. Options include reading from or writing to its database, exchanging files, using an export or import feature, or, as a last resort, automating its user interface. Each carries risks around supportability, data integrity and the vendor's licence terms. Whether any is suitable depends on the system and your requirements, and we confirm what is feasible during discovery rather than promising it upfront.
Can you work with SOAP, XML or older interfaces?
Yes, these are part of the work. Older SOAP, XML and file-based interfaces can often be exposed through a simpler REST-style or event-based interface, with the translation handled in one maintained layer instead of in every consuming application. We review the existing interface definitions and sample traffic first, since poorly documented interfaces are common.
How do you keep a legacy system secure when it is exposed through an API?
By not exposing it directly. A controlled layer, such as an API gateway, sits in front and applies authentication, authorisation, request validation, rate limiting and logging, so the legacy system only receives traffic that has passed those checks. The specific controls depend on your environment, the data involved and any obligations you have, and we agree them with your team.
What happens when a connected system is unavailable or returns errors?
Failure handling is designed in from the start, not added later. That typically means safe retries that do not create duplicate records, queues so work is not lost while a system is down, and alerts to a named owner. Items that cannot be processed go to a visible place where someone can review and resolve them.
Can the change be made without interrupting the business?
That is the aim of an incremental approach. Consumers move to the new interface one at a time while the old one stays available, and each step can be checked and, where practical, reversed. How much parallel running is needed depends on how critical the system is and how it is used, and we plan it with you.
What do you need from us to get started?
A description of the systems involved, how they exchange data today, who and what depends on them, and any documentation, interface definitions or sample data you have. Access to someone who knows the system's behaviour is often more valuable than documentation. We start with an inventory of interfaces and consumers, then agree a small first increment.
Which technologies do you use?
That depends on the systems involved, your environment and the team who will operate the result. We choose after reviewing the existing interfaces and constraints, and prefer widely supported tools your team can maintain without us.
How long will it take and what will it cost?
It depends on the number of systems, how well they are documented and how much testing the change needs, so we do not quote timelines or prices before reviewing the situation. Starting with a small first increment lets you see progress and cost early and adjust before committing further.