June 30 Cycle Retro: What could have gone better?
9
760Ṁ215
resolved Jul 7
ResolvedN/A
20%
Moving on to new features too quickly
12%
Not engaging with paid creators/their community.m enough
10%Other
8%
Getting feedback from new users
7%
Nobody really owned manalinks
6%
Not integrating work between eng/biz/design
6%
Under-documentation of new features.
6%
Stalled user growth
6%
Not prioritizing core product enough
5%
I got sick like 3 times this month >:(
4%
Limit orders slow to be implemented
4%
I learned a lot of important stuff but I didn't teach anyone else it
3%
Standup/Discussion dragging on for too long
2%
Comment tips missing a notification and/or a way to see who gave the tip
1.4%
Roadmap and process still disorganized
Where did we drop the ball? The point of this isn't to point fingers or assign blame, but rather to identify where our processes and strategy are not serving us well. Mostly, this market is for the Manifold team; don't expect the probabilities, etc to mean much other than like a subjective vote on how much you agree with a particular statement, or how much you want to discuss it. https://manifold.markets/Austin/june-17-cycle-retro-what-could-have-e58b8dbf7390
Manifold Features
Get
Ṁ1,000
to start trading!
Sort by:
@LivInTheLookingGlass the email with the mana link for the survey went to my spam folder, check yours.
I think this is one of the more egregious ways we neglect new features. Like, we build manalinks and then didn't use it very much. I've personally been procrastinating on using manalinks to reward the winners of Argumate's markets on tumblr. I also need to post markets to LW more :/
Implementing a new feature should probably also trigger a new entry in the FAQ.
@mqp Wait, so did the survey get paid out, and I missed it somehow?
[J] Vision here - growing to new users
like, spending more time on working only gives like 2-4x productivity, whereas other things on this list are potentially 10x
@mqp Austin thought Manalinks would be cool. Implemented half of Manalinks. But then Austin got busy Then Marshall picked up the other half. Not sure the feature was important, or did a good job emphasizing how it'd be important Decided how Manalinks would be the way we pay out to the survey - maybe didn't make sense. Not an important use case for Manalinks Someone with a clear vision needed to make Manalinks to do their vision
[S] Sometimes you want to merge half a feature. [R] Would only announce on Discord if it's a new market type. Just fixing something small (duplicate a market) - could be more lowkey. [D] Create a new channel for this!
[D] Proposal: Literally times that a feature is launched, and David doesn't realize until a day or two later. Good, easy place to start: Have a channel on Discord where you post "I have launched X feature", and would be appropriate to write down things for engagement. E.g. Ian created referral links, put it into Discord chat; better if it's in a dedicated channel. New features, if they have ideas -- "good if David posted a manalink"? Post that to discord
[S] Core betting interface, different types of markets. If we could go back and do things over again, should just focus on binary markets
[mqp] Comparing learning from eg manalinks, compared to utility from eg making betting on markets better, would take "betting on markets" any day. More important to make the core experience much better than to build features on the edge to see what's good and what's not
[J] Reason to pursue strategy: we're exploring the space of things we could implement. We'll remove a lot of things eventually. We learn a lot from building something that's not that great
We should be easily able to pay or incentivize their community users to do interviews, but because creators' time is valuable, just tossing them $100 won't get us their thoughts. I'd rather add it as a contract condition.
[D] Power users will adapt. But new users come in, see a bunch of unpolished features, and that could deter them
[J] Missing a layer of polish on features where we do move on; best way to make a great product is to solve specific issues they have (requires listening to users)
[S] Makes sense to do the 80/20 of fast experimentation, but oftentimes the next best thing is to tweak something, then to do an entirely new feature
@SirSalty I'm happy to help with user research as well. It's important for us to interview creators, because their usage patterns may differ from that of average users. I'm formalizing the creator contract, so do we want to add a 15-minute call where they evaluate their Manifold experience into the contract?
[D] some good examples: comment tipping got a bunch of improvements after launch
[D] Seems like hard work of adding is done, but not making users want to use it, because of a few minor things that are wrong
I 500% agree with this, and I wish we spent more time iterating on our existing stuff, but it mega accords with my biases, so I recuse myself from saying whether we are actually doing it wrong.
[A] Discord self-selects for very dedicated users [J] Could also pay for hotjar to see how users use the site [i] Would rather just pay our users
[i] Still interested, even if it's not perfect
[D] Decided this wasn't worth setting up [A] As a team, we've been dropping the ball on getting info from new users
[D] Not just about quantitative data, but also qualitative. A lot of new eyes, should determine: is our site friendly to use? did it make sense? is this something we want to do with friends? is this interesting to them?
@SirSalty [mqp] What was the purpose in asking Aella to make questions? Was there something we were trying to learn? [S] To drive traffic; to test Twitter promotion for the use case, see how well that pipeline worked [A] So... did we drive traffic? [S] Numbers: Viewers - thousands of unique visitors Conversions to signups - low hundreds (1-200) signups (David thinks is an overestimate) Retention - harder to say
© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules