I’m using Zapier and built-in tools like Shopify and Klaviyo Flows. We were using Boomi for some things but are moving those over to Zapier. This concept of things breaking over time rings true as my team loves to change processes on their end at the drop of a hat and not tell anyone which of course breaks automations. It is painful when a) things break and it’s noticeable or b) they silently break. So, I’ve put together a routine maintenance schedule on Notion to check in with them all regularly. This catches the silent ones pretty effectively. All it takes is a quick spit check to make sure everything is working right and adjust when it’s not. I’m curious if you think this is viable because it sounds like you’re saying we’d need to eventually graduate to a more complex system (maybe one like Boomi)?
Nice one Lawson. The key message is less "buy enterprise grade tech like Boomi, Jitterbit and get rid of no-code automation tools" it's more "whatever you're using, build in the checks and balances to spot errors and to mitigate from errors occurring through proper ownership/governance". Aware that's one of the most boring sentences anyone has ever written but it doesn't make it not true.
Reality is, most people solving things in no-code tools are doing it for a really good reason but the barrier to getting something functional has come down faster than the experience and knowledge of how to make something truly production ready and stable (inside of a commerce business) has risen. So what you see is a ton of apps that look good in the demo but not in the production environment.
Sounds like you've got a pretty decent solution. Presumably your maintenance schedule checks source and destination end points for all records of a particular type (e.g. orders) generated in a particular timeframe (e.g. past 24 hours) to ensure that everything you expected to move from system A to system B has made it?
And yeah something like that although it’s not as detailed of a check as I’d like it to be. Any thoughts on whether an AI agent can/should do these maintenance checks? I have had a small amount of success in having it audit Klaviyo flows as so far Klaviyo is the only software that hasn’t blocked my Chat GPT agent
I’m using Zapier and built-in tools like Shopify and Klaviyo Flows. We were using Boomi for some things but are moving those over to Zapier. This concept of things breaking over time rings true as my team loves to change processes on their end at the drop of a hat and not tell anyone which of course breaks automations. It is painful when a) things break and it’s noticeable or b) they silently break. So, I’ve put together a routine maintenance schedule on Notion to check in with them all regularly. This catches the silent ones pretty effectively. All it takes is a quick spit check to make sure everything is working right and adjust when it’s not. I’m curious if you think this is viable because it sounds like you’re saying we’d need to eventually graduate to a more complex system (maybe one like Boomi)?
Nice one Lawson. The key message is less "buy enterprise grade tech like Boomi, Jitterbit and get rid of no-code automation tools" it's more "whatever you're using, build in the checks and balances to spot errors and to mitigate from errors occurring through proper ownership/governance". Aware that's one of the most boring sentences anyone has ever written but it doesn't make it not true.
Reality is, most people solving things in no-code tools are doing it for a really good reason but the barrier to getting something functional has come down faster than the experience and knowledge of how to make something truly production ready and stable (inside of a commerce business) has risen. So what you see is a ton of apps that look good in the demo but not in the production environment.
Sounds like you've got a pretty decent solution. Presumably your maintenance schedule checks source and destination end points for all records of a particular type (e.g. orders) generated in a particular timeframe (e.g. past 24 hours) to ensure that everything you expected to move from system A to system B has made it?
Ah that makes sense.
And yeah something like that although it’s not as detailed of a check as I’d like it to be. Any thoughts on whether an AI agent can/should do these maintenance checks? I have had a small amount of success in having it audit Klaviyo flows as so far Klaviyo is the only software that hasn’t blocked my Chat GPT agent