Detailed content plan
Plan : 100x our poasting
- We need to increase the total volume of content we’re producing substantially to ensure that we’re reaching as many people as we can
- Quantity is its own form of quality
-
[d] agreed, tweet spamming is quite a successful strategy. Bc even if you have a lot thatmiss, Twitter doesn’t rly punish you as long as you have some that occasionally do well which they will then push to all your followers and additional ppl. (altho algos have changed recently with the new for you feed so I will need to research how true this is).
-
Creators generally can’t predict which of their tweets will go viral
The more types of content we produce, the better we’ll be able to see which things work
- [d] true, but I can be sure of some tweets that definitely won’t go viral - eg. posting top level tweets that are just links to trending markets will never do well. (altho posting them in the right spot as a reply to get retweeted from shock factor could go viral like we saw with polymarket).
- [d] Is still important to find a balance. Don’t want people to perceive our twitter as a place we just dump all our markets to self-advertise. Want people to perceive our twitter as entertaining and informative.
- Optimizing for volume will improve our processes
- If we make it frictionless and low-cost to produce more content, than we will end up producing a lot more
- Our current newsletters and TikTok videos take too long to produce, so we don’t put out that many of them. If we can figure out ways to simplify production, we can do a lot more.
- [d] +1, plus motivation is also another factor beyond time constraints.
- Twitter should be our primary focus
- But our Substack newsletter + Reddit are worth engaging with too
- We should strive to be genuinely informative and focus on big, newsworthy topics
- Shitpoasting, memes, and internal Manifold drama are all good content, but it’s not enough
- Much of this stuff is inside baseball that’s totally opaque to people outside the rat/EA bubble
- [d] I don't agree with this subpoint - you can post extremely generic stuff (not even related to prediction markets and get a lot of impression and followers). But I do agree with the top level point being made here that we should strive to be genuinely informative.
- News has universal appeal and is one of the top reasons people log on to Manifold
- [d] we rly need to make it a habit to check Manifold every single time something news worthy happens. We need to really make that subconscious connection in people’s brain that new news = check Manifold (right now it might be check X news website or check twitter etc).
- [d] push notification + tweet every time a news topic is added
- even if ppl don't click it (altho I think they will) - it still re-enforces that subconscious connection btwn new news = check Manifold.
- [d] push notification + tweet everytime news triggers a big movement in a market/a new market
- Making a Manifold a more consistent place for people to find out about news (both internally within the platform, and externally on social media) gives people a reason to return to our platform
Goals
By the end of August
-
Twitter followers: 1,400 ⇒ 10,000
- [J] 🎉 Nice goal!
- [d] lol very unrealistic goal. The only way this is achievable is by posting a mega ton of replies to really huge, generic creators. Eg. spam the replies of all of Elon Musk’s posts with generic memes that appeal to a broad audience. And since that doesn’t seem to be our main strategy, it seems sub 2% chance IMO. I made a more realistic one.
- For example mealreplacer has been growing very well (6k to 12k over the past 4 months), but it’s because in addition to EA memes he also posts a bunch of generic ones that appeals to a broader audience.
https://manifold.markets/DavidChee/will-manifolds-twitter-account-have-a8e811b0f2fe
https://manifold.markets/SG/will-manifolds-twitter-account-have
https://manifold.markets/SG/will-manifold-have-1-million-visito
Strategy
-
Manifold news dashboards
- We should keep the news content on our site up-to-date so users can rely on us as their go-to source of information
- Organizing news content will also make it easier to source content to post on social media, etc.
- Where to look
- Trending markets
- Our users notice and start trading on big events before anyone else
- If there are several Trending markets on a given topic, that’s a good sign
- Google News (or any other news aggregator)
- Many of the top stories will be things we want to include
- (I personally have found GNews to be great)
- Twitter
- When creating a dashboard
- Include a mix of markets, news articles, other relevant content (graphs, images, memes)
- Make sure we include the largest and best markets our users have created
- If we’re missing a big newsworthy market, we should create it ourselves
-
Twitter
- Once we’ve added news, it’s easy to source content for Twitter
- See Twitter strategy: Prediction market news through tweets (as a product!)
- [d] This seems to focus on posting lots of top level content. This is fine for engaging our existing audience, but the overall number of impressions for top-level posts tends to be magnitudes lower compared to being fast to post replies to popular tweets. You rely really heavily on big influencers commenting or retweeting to get new users. And I’m not sure general newsy stuff is the sort of content people feel inspired to share.
- [d] I think our core goal should be thinking how we can make it easier to flood other large twitter account’s replies with memes, markets, valuable info (similar to polymarket, even if our strategy is replying with more informative content and not just generic memes). This way we skip that step of reliance on engagement from their side to get impressions.
- [d] Top level posts are easier to mass produce (can schedule them, don’t have to constantly be on twitter/checking alerts to find appropriate tweets to be an early replied to). So we can and should do both.
- [s] Reply-guy strategy
- I think it’s worth thinking more about which specific areas of Twitter / accounts we want to break into (”It’s time for some game theory”)
- Some possibilities
- Politics/wonk Twitter
- Current Manifolders: James Medlock, Jeremiah (neolibs podcast),
- New targets : Yglesias, Vox people, neolibs
- Econ Twitter
- Noah Smith, GMU economists (Cowen, Hanson, Tabarrok)
- Biz Twitter
- Current: Byrne hobart, patio11
- Targets: Matt levine
- Tech twitter
- Crypto twitter
- EA / rat sphere
- current users: Yud, Aella, etc.
- Eric’s feedback (51k twitter followers):
- I really agree with "we rly need to make it a habit to check Manifold every single time something news worthy happens. We need to really make that subconscious connection in people’s brain that new news = check Manifold (right now it might be check X news website or check twitter etc)."
- i think from that perspective, the "reply guy" strategy doesn't work well because no news organization or platform really does that. it makes it come across more like a corporate PR social media manager behavior rather than an alt platform for news. reply guys are also very cringe and undermine broad trust
- generally, accounts that share new and breaking information tend to attract a lot of followers. i think sticking to that principle will get the most followers, and trying to go for the roon / thought leader / humorous -style poasting is more of a luck-of-the-draw kind of thing
- i personally find the Manifold twitter very unfunny, but that is also my bias against EA / rat content
- for instance, for a given breaking news event (e.g. russian coup), you can take advantage of that by signal boosting a relevant market (e.g. odds of Pirgozhin survival) on the twitter, since that broadcasts new info to the community. Add hashtags so people browisng russia news on twitter see the relevant markets
- broadcasting broad sentiment and beliefs around highly volatile situations will prob yield very useful info to ppl very interested in newsworthy events
-
Newsletter
- Instead of one giant newsletter update that goes out around once a month and is mainly focused on new Manifold features and community drama,
- We need weekly update emails focusing on current events.
- Should be relatively easy to write since we’ll have already identified the big newsworthy topics that week
- A short paragraph on 3-5 different topics discussing market movement and maybe a link to an external news article, etc. would suffice
-
Video
- Everyone loves video clips, but unfortunately, they’re expensive to produce (timewise)
- It’s worth experimenting to see if we can figure out a formula to churn out shortform videos to post on Twitter, TikTok, YT, Reels, etc.
- I think a very simple format where we create a short clip per news item, read off the market probability changes, and include a few animations, graphs, images, etc. could work.
- The important thing is making them really easy to produce.
- (As an aside, I spoke with an entrepreneur last week who believed that the right pace for a growing startup is more like 100 videos a day, rather than per month) ****
-
Ads
- Much less of a priority than producing organic content
- Might be the only way for us to make inroads on Reddit, FB
- Still in an experimental, meta phase
- We’re testing to see if there’s a process we can use to reliably create cost effective ads, rather than trying to optimize for object level results
- I’ve created a few highly targeted ads with excellent conversion rates, low CPCs—but were too small to scale up. We want a path to scalability.
- General approach: only create ads to scale up content that we know works
- Ad content should be at least semi-viral on social media before we spend any additional ad $$$.
- Don’t bother with novel, untested ad ideas.
- Twitter, Reddit, FB are top tier ad candidates for us
- CPC <<< $1, otherwise it’s not worth it