Is Processing Faster if you specifiy the form in a "Fills Out Form" Trigger?

[Anonymous]

I am in the habit of using:

Fills Out Form->Form is Any

Web Page is <Name of LP>

Because if the form changes, your campaign still works - and you don't need to remember which form is used.

However, someone recently told me the processing was faster if you specify the form, as in:

Fills Out Form->Form is <Name of Form>

Web Page is <Name of LP>

Can anyone verify if this makes a significant difference in processing time?

Thank you!

Denise

[Devraj_Grewal]

I would imagine the smart list would be faster to process since it is filtering down even more by only looking at that specific form being filled and then looking at if it was on that specific web page. Because then if the form being filled was not that specific one, the constraints wouldn't even bother being checked.

[Anonymous]

That makes sense but wondering if anyone can verify if it's a significant difference. Just found a whole trove of these campaigns in a client instance that were referencing the wrong form and wondered about recommending leaving out the form name to avoid this sort of error.

[SanfordWhiteman]

I would imagine the smart list would be faster to process since it is filtering down even more by only looking at that specific form being filled and then looking at if it was on that specific web page. Because then if the form being filled was not that specific one, the constraints wouldn't even bother being checked.

Yes, if you look at it from a relational-type query perspective, where if there's an index on a Form column but not on Web Page, then there would need to be a table scan on Web Page.

But this is a trigger. So if you're inspecting a single object that fires the trigger then it is faster to ignore one aspect of the object (the Form) and just match on Web Page. In this case the difference could be infinitesimal, like microseconds, but deep down it would still exist.

Of course it's always possible to introduce a weakness into a design such that it unexpectedly bottlenecks, despite theory to the contrary!

I don't have real-world evidence either way, and it would be really hard to benchmark since you can only send one form post every couple of seconds from each IP (so if the difference on an unloaded system is <1s you'd never see it).

[Anonymous]

Sandy - Are you psychic? I swear I was on the verge of replying to my original post with an @ mention to you - specifically to get your take on this!!

[SanfordWhiteman]

Psychic or just "always watching this browser tab."