Manifold Team April Retro: What could have gone better?
18
246
resolved Jun 15
24%11%
new search isn't great
21%6%
Free response still a trap
20%7%
No growth this month
9%3%
Private groups taking a long time
7%2%
Our tracking of bugs, features, and projects is a mess
7%2%
Standups are very long and can feel nitpicky/get bogged down in details
7%1.9%
Supabase dev process undocumented, confusing, fragile
5%1.5%
less high value markets than usual/trading on interesting things
0.0%
Lots of new user testing!
0.7%
Manifund: Not much activity post ACX grant
0.7%
Manifund: Ops taking a lot of time
0.6%
site outage right when isaac was going full degen
1.8%
Manifold gets shared on each Twitter and Reddit 5-10 times per day. Majority gets no engagement.
0.6%
Failed to get streamer subreddit on boad
5%
@Wobbles still hasn't gotten a badge
9%
Many new features have been introduced, though none have seemed to have a large impacted
1.6%
Seems bad to let one person control API keys for hundreds of other users for coordinated activity, one API key per user/use case IMO.
18%
Need to implement guild system for Manifold
26%
Separate leaderboard for predictive and non-predictive markets

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.

Last time: https://manifold.markets/JamesGrugett/october-retro-what-could-have-gone

Get Ṁ200 play money

🏅 Top traders

#NameTotal profit
1Ṁ56
2Ṁ53
3Ṁ40
4Ṁ37
5Ṁ33
Sort by:
answered
new search isn't great

[i] Made a bunch of changes to try and make it better
[c] Do we track search usage? How do we tell if search has gotten better...
[J] It's declined a bit
[S] But so has overall site usage
[i] Asking if people are frustrated or not; "does it work for you"

answered
Manifund: Ops taking a lot of time

hard for me to tell how bad this compares to the other issues on this list, just because I don't really get Manifund day to day looks like (nor do i need to, just pointing out in context of this ranked list)

answered
less high value markets than usual/trading on interesting things

[D] If there's no super juicy ongoing news, there's less excitement to trade about it.
Even during a dry news cycle, we should be able to highlight/surface "this is what's going on"
Feels like we have this automatic feed, which just shows you some things. Even non-interesting things we could make become interesting.
News websites do this (click-baity titles, overexaggerate) but we should figure out how to do this while remaining true to Manifold

answered
Manifold gets shared on each Twitter and Reddit 5-10 times per day. Majority gets no engagement.

[J] Might just be the pattern
[D] Kind of out of our control?
The preview especially sucks -- only good for people who knows about Manifold already. Could try to be more call to action. RN just a weird squiggly line with some numbers, leading you to a strange crypto-looking app

[S] Can create some mockups

answered
Many new features have been introduced, though none have seemed to have a large impacted
bought Ṁ40

@Wobbles VS markets, ads, stocks, etc

answered
Supabase dev process undocumented, confusing, fragile

[c] Types don't work; seed.sql sucks. When trying to build a simple feature "money goes into a thing and goes somewhere else"

Most frustrating part was working with Supabase. (Perhaps too used to Vercel?) there has to be a better way

[J] We should strive for a better dev experience; should tackle these problems and come up with a better overall path for adding new features and adding tables and migrating stuff

Agreements with the rant?

[S] Absolutely
[J] Reminder that SQL is super powerful -- so pros and cons
[C] Worse experience for simple stuff, it feels

answered
@Wobbles still hasn't gotten a badge

@Wobbles u have to wait one month for each secret alt u made ;)

bought Ṁ30 of @Wobbles still hasn'...

@DavidChee that would take approx 12 years

answered
Standups are very long and can feel nitpicky/get bogged down in details
bought Ṁ10

@Austin agreed, I feel like my most productive conversations are when talking with 2-3 ppl for 10-20 minutes randomly outside of standup

bought Ṁ10 of Our tracking of bugs...

[D] Standup sometimes feels like fluff; but afterwards has a productive conversation in the room. Not captured in standup because too many people; too many discussed in detail (others don't need to have a say or be involved)

Perhaps be more picky over what we gloss over vs discussing

[A] Will own this as an area to improve

answered
No growth this month
bought Ṁ15

most important. This itself indicates a lack of use of markets in news, answering cool new questions that a lot of people care about, or people having a lot more fun/interest than before.

It is a reasonable proxy for how much total value we are actually delivering to people

answered
Free response still a trap

@ian Multi-choice / Free response still DPM and can't buy NO

answered
Lots of new user testing!

@JamesGrugett we coulda done a lot more! no one came with me to the bar :(