I've just tried a Lookup Cache for the first time and thought I'd share a bit of feedback and why I'm not using them. I read all the help documents so I feel like I have a good sense of what is possible, but I could have missed something.
They are account wide, all Integrations. This can be good or bad, depends on your use case. For me, this isn't a good thing. I had a need for an Integration specific map that contains data that makes no sense for anything else in the account. I think we should be able to scope these things, I don't agree with the global to everyone and everything approach.
Can't easily create a new Lookup Cache by defining the values in an IIO editor, you need to use a CSV file or create one with a Flow. Again, this can be great for some uses cases, but for mine it felt like a miss. I have two entries used in 5 flows. Quickly creating a Lookup Cache directly in the UI is what I was expecting.
Anyway, seems like it could be cool for some uses cases. Really need to be able to scope them.
Thank you for trying out Lookup Cache and for sharing your feedback—it’s really valuable. Here are some updates based on your points:
Scoping Lookup Caches: We recognise the need for scoping Lookup Caches to specific integrations rather than making them account-wide. This capability is on our roadmap, but we had to defer it from the current release to ensure we delivered MVP sooner rather than later. Additionally, we are planning more enhancements, including access via Handlebars and scripts. If you have specific use cases that could benefit from these improvements, we’d love to hear them.
UI-Based Creation: This will be addressed in our upcoming March 6th release. You’ll be able to create Lookup Caches directly in the UI without needing a CSV file first, making it much easier to set up smaller datasets quickly.
Let us know if you have any other feedback or if you’d like to discuss your use case further - We really appreciate your inputs.
@priyankakoundal - I am still getting familiar with the lookup cache, but curious what the advantages are to using the lookup cache vs static mapping when there are only a "few" mappings? I guess I always thought the "bang for the buck" for lookup cache was when I had a multitude of mappings that needed to be done (for example taking all the State Names and Mapping to their 2 letter abbreviation) that would have otherwise been cumbersome doing the static mapping.
That would for sure be one use case. What I would also use it for is to save API calls for certain data. For example, I would create a lookup cache of all my items so that I can reduce API calls by avoiding always having to do a dynamic mapping for items. I probably would do the same thing for customers.
I'm still learning this as well. Are we now able to basically replicate the lookup step within the lookup cache feature as if we were building the lookup step within the flow like normal or is that what is coming in March?
@chrishuovinen305 - Its available now, but as Steve mentioned it requires you to load a CSV. I think the enhancement that is coming is where you can do the mappings in the UI (for example if you only have a few mappings) rather than always needing to load the CSV.
It depends on what use case you're trying to solve. In my example of using it for items, what I would do is make a flow that pulls items from your master item system (NetSuite or PIM or something else), then upsert those items into a lookup cache you've made. I would then schedule that flow to run once a day or so, so that the cache is always up-to-date. From there, I would then reference that lookup cache in my mappings by choosing lookup cache instead of static or dynamic mapping. You could also have an individual lookup step to fetch the value of a particular item stored in your cache.
In my case it's reduction of duplicate static maps. We map the same few values in 4 or 5 flows (NetSuite Location to 3PL Warehouse). defining in this in one place is a perfect use of a Lookup Cache.
@tylerlamparter are there any plans to natively support fail-over to a full dynamic lookup when the cache fails to yield a result? It would be powerful to hit the cache for a lookup and in the event there isn't a match, automatically continue with a dynamic lookup and automatically add the result of the loop to the cache. Like a lazy-loaded cache or lazy-appended
That would definitely be nice, and I'm not sure if @priyankakoundal has it on her radar yet. You could do this by having individual lookup steps and then branching the flow off to make a different lookup when the first lookup against the cache yields no results. Individual lookup steps can get cumbersome, though, if there are many fields you need to do lookups for.
@steveklett ... we have discussed using this lookup cache as a front end to an inline application lookup, but it's further down the stack for priorities. I couldn't suggest a timeline for that currently.
Appreciate the vote though! Agree it would be a very great add to this.
I have an additional feature request. I would find it useful to have a comment or memo field for the mapping. For example, mapping UPS facility codes to NetSuite internal IDs looks like a bunch of gibberish and a simple note "Bourbon, KY BOND > US 3PL Bond" saves a few minutes of research.
This is what I'm doing now, which is OK, but an actual memo would be better IMO
Very interesting idea, @steveklett I assume you’d use this label/memo while configuring Lookup Cache mappings and would prefer to see it displayed on the UI for quick reference during the mapping process. Is that correct?
Yes, basically. Use it will setting it up and also when updating it later as things change. It's useful to be able to record a little context or intent with things like this. Similar to the Description on Mapper 2.0 rules, those are great when you have a setup that isn't obvious or has a special consideration.
