Product Sense: Why Note-Taking AI (Currently) Disappoints
Earlier this week, I wrote about how easy it is to test products now to hone your product sense. In this article, I focus specifically into note-taking apps.
At a surface level, note-taking seems like a killer case for AI. However, I have discovered that until it is closer to perfected, it causes more work than it is worth, because you still have to take your own notes and you now have to read through and edit the AI output. When they get it right, it will be a killer app. But right now, it can only give you no-brainer notes. It is missing the subtle things you catch on your own, or so is my initial read.
Right now, AI seems to work well in generating article ideas - your muse or focused influencer. With article or document ideas we are open to rough ideas, we know the AI outline is just the start of something we need to dive in and edit. But with meeting notes, we want to plug and play solution, it is meant to save us time. We don’t want to take time to fact-check, merge with our own notes, and then send it out. This takes 2x or 3x the time it would to just do the job ourselves to start with.
Here are some initial observations on the various platforms I tried in the last few weeks. I use note-taking apps to solve the following job: Providing clients with summaries of what we have discussed and key takeaways so they can practice on their own.
Still an ongoing effort but here goes:
First impressions: Wow! So much information. Trying to give me notes + videos + feedback on speaking.
Using it Day-To-Day: Too much info to process. The results seem more helpful on the surface but when I dive in, there is too much to wade through to make it work.
Biggest Detractor: Really annoying that they take up space on the Zoom screen and are always obviously there. I started turning it off because it was not helpful and too loud and the info was too much to process for all my calls.
First Impression: Wow! so much info. Loved the ability to tag important parts of the video.
Using it Day-to-Day: Practically, I couldn't find a way to tag anything - I think they cut off my free version because I signed up weeks before I could use it because of computer problems.
Biggest Detractors:
Failed Freemium/Free Trial >> Too little time to see if marking the video would work for me. I never reached the Aha moment, but I might have given more time.
Given the want $20 a month and I have soo many other things I am testing, it isn't worth pushing to see if the one little thing works. MAYBE if customers told me they would pay a premium for outputs, I would consider it.
BUT it would require me to mark concepts AFTER they happen and so even if you use it, it just means you tag that you need to roll back X minutes to get to the tag. It has now added more work than just taking the notes myself.
Annoying UX: No matter what I do the trigger a recording button always floats in my way and I can't minimize it. And it keeps treating me like a first-time user after 7 days of regular use. And 99% of the time it does nothing because the recording is already running. I don't understand the value it should drive.
I got so frustrated I removed it today and even that was a pain, I had to go past the standard force quit methodology to be able to delete it. Seems sloppy and not respecting my privacy so seemed a red flag to me.
Zoom's AI
First Impression: Right place-Right Time.
Using it Day-to-Day: I seem to need to trigger it. So it really is only good for catching up if I came in late or spaced out. I need to play with it more.
Biggest Detractor: It came with soo many other add-ons, I am confused about where everything is. To trigger it for all meetings, I had to log into the web interface, I can't adjust settings from the app, really annoying.
Related
Note: I am willing to pay for both of these services. I think it boils down to simple is better. They are solving my problem better than the other services that are trying to do too much. They both offer much more than I use, but they let me use what I need as I need it without overwhelming me. Yes, I would love one interface but I will take a few interfaces to get my jobs done quickly.
First Impression: What I have always needed to monitor my speaking cadence.
Using it Day-to-Day: I have to get back into the swing of things.
Biggest Detractor: It wanted me to fix a setting, and so gave me a warning message without a link on how to fix it at the top of ALL my meetings. So I just minimized the product to concentrate on the call, causing me to stop using it. I had to ask for help to figure it out. And thus, I was distracted for the first 5-10 minutes of most calls, causing me to be annoyed with the product. I wish they had a smart email when they realized I didn't have things set up properly. Tell me how to fix it when I am NOT on a call.
First Impression: Clean note-taking app that can help me manage notes and clients as well as to do lists.
Using it Day-to-Day: For me the backlinks took a minute to grok. I just started playing with audio notes and linking process. Loving that I can link to people and meeting notes and pull up templates as I take notes, so as I type in feedback, a simple backslash will input valuable info into notes for me in seconds.
Biggest Detractor: Not recording and summarizing, everything is on me. (Although I am testing the recorded note feature).
In Summary
In general, curent AI note-taking apps create more work than just doing it myself because they give me too much to weed through, so I might as well just quickly type my own summary. Because I have a lot of mock interviewing in my sessions, the note-taking apps get confused and summarize the content of the mock as if it was a real meeting.
In their defense, most of what I am sharing is my reaction to the content, which is in my brain. What would be nice is if I could trigger it to only record when I impart advice/wisdom. Maybe trigger words that would tell it to record or just weight what I am saying more highly.
