Re-write integration or leave it alone?

We have a well-functioning existing API integration with our customer facing website. This was written last year. Some of the calls are BAQ based services but many use the BO’s directly.

I’m trying to decide if the juice is worth the squeeze on re-doing this integration with only using BAQ services.
The advantage of BAQ services is that the client is then mostly insulated from changes in Epicor. Additionally, the complex business logic can be handled internally by the BAQ service rather than the client.
The disadvantage is spending time and effort re-doing a working product.

Any words of wisdom from people who have been faced with a similar decision?

Don’t fix what ain’t broke lol. If a specific area of the integration requires updating maybe do a conversion of that piece since you’re touching it, but I wouldn’t go out of your way unless you’re really bored.

4 Likes

Haha fair enough. I think it’s a matter of principle for me and trying to prevent any future issues.

I hear you there it would be nice to live in a perfect world but sometimes it’s best to leave a sleeping dog lay. You would hate to kick up dust and potentially any new problems and now be in a scramble to fix them. Maybe otherwise come up with a schedule to migrate a small piece here and there so it has ample testing time between moves and just get it upgraded in small bites?

1 Like

I usually look at two aspects.
Customer satisfaction and maintenance costs.
Are internal or external consumers struggling with some aspect?
How is the cost to run the service in human time or hardware costs (or rental if you live in a public cloud world)?

Basically what’s the biz value?

The next consideration is the boy scout approach. Have your roadmap and when in the area, incrementally improve the service. Like a boy scout, leave it better than you found it.

3 Likes