A game changer in the research process but it still freezes

We use Dovetail to both analyze and parse through our user testing results and also as a research repository. We have to handwrite our notes and so we transcribe the items, then tag them by topic (we create for our team), group them by type (usually starting with the AI then manually updating and refining), then create insights. The insights AI feature is an ok start but doesn't work as well on our data set because of the handwritten notes. We use that as reference but then make our own. The AI ones are usually too general. We do things like Task completion rate and likert scales outside of Dovetail (in Excel) and create the final report in word. There is not enough functionality in Dovetail for either of those use cases and when working with a teammate it tends to freeze.

They only work so-so for our data sets. I think the AI is geared towards users with videos and we don't have that because of our customer base. I think the AI also makes assumptions that sometimes I don't trust and that's because they are pulling it from data/notes from testing sessions, where they don't have the whole context (we are doing user testing using prod applications or interactive prototypes so verbal feedback is given but no image is included in Dovetail). The only thing that it has helped us/where my team users the AI feature is the sorting of the tags on the highlight board. However, we still have to go back and sort more because again it's missing information and is not good at grouping by theme (the tags with ok as a start).

My team is small (for my project not company) and so it has helped for us on that smaller scale but only to the design team level.