What needs to be in place
If you run NAV 2009, it has to be on SQL Server, as we don't support the native database. The client has to run on a version of Windows that Microsoft still supports.
NAV 2009 to 2018
The integration covers NAV from 2009 to 2018. Which data we fetch, and how prices are fetched, varies from version to version, so we go through it with you and your NAV partner before we start.
Why we mirror data
The webshop could look things up directly in NAV via web services, but using a data mirror gives faster response times and better stability. We keep the dependency on NAV being available to a minimum and reduce the load on your back office.
It's about efficiency and robustness, and the experience is better on both sides: for your customers in the webshop and for your staff in NAV.
You decide when data is sent
The client is started by Windows Task Scheduler, and when it runs, and what's included when, is something we agree with you. Stock levels can be sent more often than invoices, for example, and individual tables can run on their own schedules - the more data there is to move, the more thought we put into what goes when.
If the jobs are set up automatically, the client sends invoices and credit notes twice a day and the other data every hour between 6 am and 6 pm. In the evening and at night it usually pauses, so NAV's own overnight jobs can run in peace - they can lock tables if they're being read at the same time.
The order
Web orders are loaded into a journal. NAV doesn't have standard modules for e-commerce, so the journal has to be built, and here we have components from other customers that your NAV partner can take inspiration from - what's possible depends on your licences and which object ranges are free. Orders can also be dropped as CSV or XML files in a folder we agree with you.
Together with your NAV partner we go through the mapping, and sales orders are created in NAV from the journal or the file. From there, sales order processing, release, picking, packing and shipping happen exactly as if you'd entered the order yourselves in the order screen in NAV. Our responsibility ends at the journal, and after that your normal sales order flow applies.
Web orders are fetched from the webshop by a job that Windows Task Scheduler starts on your side, every 5 minutes by default. Until they're fetched, they wait in the webshop.
Prices
Every combination of prices and discounts in standard NAV is supported. If you've customised the price calculation, we go through it with you before we start.
Which items and customers are included
At start-up we fetch all data. If there's a pattern for what shouldn't be included, such as blocked items and customers, we put a filter on the replication, and we agree that with you once we've seen your data.
Sales history goes back 2 years as standard.
Related data
Items and customers come with their related data, such as customer addresses and contacts and item groups. If you have your own tables, fields or lookup tables in NAV that the webshop needs, they're easy to add to the replication.
If the server is switched off
Then we don't get new data, but the webshop keeps running and takes orders, which wait until the connection is back.
We monitor the replication and have fixed playbooks for outages and known errors.