Integration maintenance
When a regulation or an API changes, we want someone to take care of it.
- What we do
- We maintain the integrations already in use between your systems: we update them, fix what stops working and adapt them when a regulation changes.
- How we do it
- First we list with you the integrations in use: which systems each one connects, which data passes through it and what it depends on, such as a marketplace's API or the e-invoice format a country requires. Before including an integration another supplier built, we read its code and its documentation and tell you whether we can maintain it. In the contract we write down which integrations are included and how quickly we respond when something stops working. From then on we follow the changes each system on the list announces and what the tax authority publishes in each country your integrations invoice in: when an announced change affects an integration, we tell you what changes, adapt it, test it and put it into use on the date we agree with the person in charge of IT, or for the date the change sets, when the notice leaves time. When you report that something has stopped working, we respond within the written time and look for the cause: if it is in the integration, we fix it; if it is in one of the systems it connects, we tell you which one and what whoever runs it needs to do. Every change is recorded in the integration's documentation.
- What you receive
- Maintenance of the integrations on the list, with the response times written into the contract.
- What changes for you
- A change to a marketplace's API or to a country's e-invoicing is no longer an emergency that nobody owns: an announced change is followed before it arrives, and one that arrives without notice is handled as something that has stopped working, within the written time. Each integration's documentation says what was changed and when.
- What we need from you
- The list of the integrations you use and of the systems they connect; access to those systems, to each integration's code and to its documentation, if there is any; a test environment for each system, where there is one, to test changes without them reaching customers, marketplaces or the tax authority (where there is none, we tell you before we start how changes are tested); the change notices your suppliers send to your company's accounts, or access to them; and the person in charge of IT, who reports to us what stops working and agrees with us when each change is put into use.
- Who it is for
- IT and management
- How you buy it
- For a recurring fee: the fee is written down before we start and we give it to you when you ask.
Example use cases
These are examples: they do not describe a real client.
A car parts distributor in Italy that sells on two marketplaces in Spain and Portugal
The situation
The integrations that pass the orders from the two marketplaces into the ERP were built years ago by a supplier that no longer works with the company, and no one there knows how they are built. One of the marketplaces announced well in advance that it was changing the part of its API that provides orders, and the notice reached the online sales mailbox, where no one knew what to do with it. When the change took effect, the integration stopped receiving that marketplace's orders, and someone had to be found in a hurry to rebuild it.
What we do
We read the code of the two integrations and what is left of their documentation, and write the list: which marketplace, which data, which part of its API each one depends on. From the start of the maintenance, the marketplaces' notices reach us too. When one announces a change, we tell the person in charge of IT what changes and by when, adapt the integration, test it in the marketplace's test environment and put it into use before the date of the change. Each integration's documentation is kept up to date with every change.
A food group with companies in Italy and Spain, with an invoice bridge between the two countries built by another supplier
The situation
The e-invoices of both companies come out of the ERP, the Italian company's through the SdI and the Spanish company's in the format its customers ask for, and the invoices between the two companies go through the same bridge. When the tax authority of one of the two countries changes the data an invoice must carry, someone has to change the bridge, and no one in the group follows those changes: last time, the accounts team found out from the rejected invoices.
What we do
We read the bridge's code and it goes on the maintenance list. We follow what the tax authority of each of the two countries publishes about e-invoicing, and when a change affects the bridge we tell the accounts team what changes and from when. We adapt the bridge, test it with test invoices in both directions, without them reaching the tax authority, and put it into use on the agreed date. Which taxes to apply is still decided by the group's tax adviser, as it is today.
Frequently asked questions
Can an integration built by another supplier be included?
Yes, if we have access to its code and to the systems it connects. Before including it we read it and tell you whether we can maintain it and what it would take. If the integration is another supplier's closed service, with no code we can access, we tell you before we start: that one we cannot maintain.
How is it different from integration monitoring?
Monitoring checks the flows automatically, alerts you when one stops and gives a first response within the hours set in the contract. Maintenance changes the integrations themselves: it updates them, fixes what stops working and adapts them when a regulation changes. They are two separate services.
What does the fee cover?
Updates, fixes and regulatory adaptations of the integrations on the list. A new integration, or a function an integration did not have, is not maintenance: it has its own price, written down before we start. Adding an integration to the list is agreed in writing beforehand, like any change of scope.
What does it mean that the response times are written down?
That the contract says, before we start, how quickly we respond when you report that something has stopped working, and within which hours. We do not promise that someone will be on watch at all hours.
What about the solutions bought as a project plus a fee?
Their recurring fee is this maintenance, for what we built in that project, with the response times written into the contract.