You should have used different tracking parameters to track the source of form fills, as that’s the main reason we include the UTMs; if UTMs are the same across different assets, then it’s as good as not including them (at least from the perspective of not being able to distinguish interactions distinctly from the emails).
Assuming that all the CTAs in your 3 emails went to the same page (i.e., to the same URL) and the UTMs were also the same, I’m afraid you would have to go through the route of looking at the Click activities just before the form fill activity (export activities using Bulk extract APIs). You might find multiple click activities near to the form fill activity, but that shouldn’t be too many, so you can at least get an holistic view of form fills from each email. Also, I’d also like to highlight the potential of a click activity from a bot (and possibly a subsequent visit webpage activity too, as a sophisticated bot can follow through the CTA and log a munchkin hit too); however, a click, visit, and a form fill activity in proximity would most likely be from a real person.
You might miss attributing form fill activities where the person filled out form a little later after clicking the email, e.g., they might have navigated away for a while, came back, and then filled out the form. However, if you don’t have your other emails (other than these 3) pointing to this webpage (or at least with the same UTMs), you can still look for the last email that had a click activity given that the UTMs in form fill activity and the ones that you have in your emails’ CTA match.
Edit- Sandy was quicker than me to post an answer, but I’m glad we think alike overall regarding the activities’ proximity solution. 😊
.png)