📔 week 5 of @ycombinator's @startupschool kicked off today new format today: a live discussion between @mwseibel & @daltonc 🚧 Recognizing and Avoiding Tarpit Ideas 📝 sharing my notes:

in the first startup school talk, we discussed tarpit ideas this talk will go deeper into the topic and how to get out of these tarpit ideas

it's the first time they're doing it live - usually they have time to edit!

today they're giving advice on the startup ideas that founders most often pivot into and out of they are TARPIT ideas!

michael did some research on a (non startup) tarpit: a place on the earth where tar and petroleum is seeping up from the ground - forming a tarpit

to animals, it resembles a freshwater pond - so they go in and get stuck their screams attract other animals, who also get stuck everyone dies ☠️☠️☠️

you'll find many fossilized remains of startups in the startup version of tarpits just don't go in!!

consumer internet startup ideas - they start here because the most common ideas on YC applications are consumer most tarpit ideas are in consumer most ideas that they reject are in consumer

what is a consumer business? simple definition: free or paid service and random people will consume it. not businesses but individual e.g. @netflix, @facebook / @Meta if there are hundreds of millions of potential customers, then you're probably in a consumer business

why do founders tend to choose consumer ideas? because they are consumers. we all have opinions about how goods and services that we consume can be better for people not in tech, when you think of startups, every startup you can think of is typically a consumer idea

here's the issue - for most pre product market fit companies that they see at YC, the founders are miles away from something working they're not calibrated on what a good consumer idea is or what success looks like

they'll mistake a few early users as successful traction for example @reddit has been around for nearly 15 years. when you at how it started small and grew and grew - now 10s of millions of people use it everyday for the longest time, when it was small, no one cared

people only cared much later when it was huge consumer founders want money up front to drive huge growth - but the reddit story was of an under-capitalized company that grew slowly reddit spend zero on user acquisition, it grew organically burning basically zero money

@mwseiebel is on the board of @reddit so he knows what he's talking about! it's still early in the @reddit monetization strategy - which is crazy given that it's 17 years old!

another example - @twitch interesting case because within 2 years, they got to 30m monthly active users, approx $8m annual revenue. they got a $24m valuation. they spent zero on user acquisition. all their time was focused on making streaming cheaper so they wouldn't die

they only survived by finding a way to make their streaming 10x cheaper than competitors investors didn't believe that they could monetize via video ads. so they had to wait for that entire market to form

the calibration here: there was infinite demand for these products and user acquisition wasn't the issue both products got millions of users for free but they were still nearly dead if you don't have millions of users then it's definitely not a good sign

looking at @google now - they crawl the web, they don't pay anyone for that. they pay for servers. people go to google .com voluntarily and use it 20-30 times a day. each search is worth money. costs basically nothing

they get $500m a day from users to run the website. that's a great consumer business. they could fire everyone and continue to make that money @google spent literally zero on user acquisition. they built something amazing that everyone wanted

the MVP was hosted on a stanford university website and people found it organically that's the benchmark

@facebook also a great business their retention and sign up numbers: 75% of their harvard users signed up within 48 hours. they used it for 2 hours a day. that was pre-funding, when it was mark in a dorm room zero on CAC. zero on engineering hires. zero money raised

the same thing happened at yale and other early schools all pre-funding that is the bar - if you build a new social network and you can do that, then maybe they will notice

the message - if you're doing a consumer idea, don't lie about what business you're in and the (very high) bar for success there aren't many facebook type businesses left to build because there is a cap on the number of hours someone can spend on social media

the current stats suggest that people are already on social media for nearly as long as people are awake so it's totally zero sum. you will compete against giants like facebook and tiktok

there are consumer services like @airbnb and @uber - where you increase the market with superior services there is a play there but you can't ignore the scale element you need MASSIVE scale for these things to work

people often ignore tech adoption cycles, where a new platform comes in and there is a race for that platform e.g. with @apple's iphone, now who will take advantage of that without new platforms coming out, the incumbents get really powerful

consider if you're in the start or end of one of these cycles front running the cycle is almost never worth it e.g. VR 5 years ago

a lot of ideas are now about adding tokens or decentralization that might work but don't lie to yourself that there isn't a new problem: incentive structure is different if the value of the token doesn't go up then people get mad

you will have tens of thousands of anonymous strangers criticizing you - a lot of people who take this path end up quitting michael compares it to going public before launching - sounds very stressful to have all these shareholders watching your business

you have to do something more than just add a token it's a bit like adding 'social features' to random products in a previous tech cycle

what is a tarpit idea? ground rules: these are not ideas that can NEVER work or that they would never fund them at YC

they fund tarpit ideas all the time BUT the founders who make it work are deep experts, are aware of that risk, that many people have failed and understand the problem that all helps them execute better

during a YC interview, they will know who has failed and why they will do it differently. they'll leave a 10 minute interview learning something new from these founders

another note: the reason these are tarpits is because something is appealing or sexy. people may give positive feedback about it the worst tarpits make founders emotionally attached and super defensive about criticism about the idea

that's why they will see the same ideas over and over again - even across 7 years it's a bit weird

a non tarpit idea - which is comically bad - a personal air balloon for commuting. it's just obviously wrong

consumer businesses have a higher chance of being tarpit ideas than non consumer businesses it's hard to have expertise in non consumer businesses. a larger group can fall into the consumer traps

it's compounded by people sharing these consumer ideas with other non tech startup friends

faux world of startups like Shark Tank contributes to this if you believe that world then you'll fall for things like consumer packaged goods they are targeting views for their show and it's not related to actual big successful businesses

you have to put things into the context of the time you're in some ideas that worked in the past: TV channel for people who like music (MTV) news channel that plays 24 hours a day (CNN) these days there are not as many ideas for new TV channels

you'll see low hanging fruit get taken and the bar gets higher and higher with time

here are some tarpit ideas 1 - the app to discover new things to do: events, concerts, etc everyone wants to believe there is something great out there and they could learn more about it "discovery is broken"

Michael realizes that that mythical restaurant doesn't exist out there no amount of discouragement will stop founders Dalton thinks it's a phase that founders just have to go through

so many good people waste years of their lives on this psychologically, humans are susceptible to FOMO and this hits on the human fear

2 - something to do with creators a lot of people on social media are exposed to creators because they are celebrities you come up with tons of ideas around helping them connect, monetize, etc because it's available

why don't these work? creators are proto-celebrities. they are on the path to becoming celebrities why don't billion dollar companies exist to serve A list celebrities? because those users can consume very tailor-made solutions that are perfectly calibrated to them

it's also very hard to separate the creator from the platform as the platforms change then they may not keep up just like the celebrity who can't control which movie they are casted in very tricky group of people to build for

because this is a tarpit, people are really defensive about this space it makes them irrationally hold on to the idea

3rd tarpit idea - something to help brands michael's best guess on why these all fail: trying to sell software to people who don't value software

people love luxury brands but they don't value software

4th tarpit idea - anything about ads problem: the platforms where ads run control the ecosystem it's very scary because you can get revenue, so you can think it's working but watch out. massive platform risk

when you need them and they don't need you, it's very problematic

5th tarpit idea - consumer packaged goods & consumer hardware the reality is that these businesses do free R&D for large brands producing physical products at scale is extremely hard and extremely expensive

they have shitty cashflow dynamics huge upfront R&D and product costs, which is better for large companies they typically sell to big brands like @Nest selling to @Google

it's tricky because there is a real demand from consumers but you cannot fulfill that demand at scale

6th tarpit idea - crowdfunding it's been popular since 2008 so it's hard to find more crowdfunding niches very low success rate crowdfunding betrays the problem because money is only one of many inputs for success

7th tarpit idea - gambling software people LOVE this idea because they gamble Michael's explanation is: sharks vs newbs. Sharks prey on newbs. Newbs don't want that. Sharks will always find more newbs. News will run away when faced by sharks. You can't balance these two

8th - options/crypto trading again people love this personally but it doesn't necessarily mean that it's a good idea a lot of people get caught in it

9th tarpit idea - vague web3 startup idea with tokens etc when you dig in, it's basically just tokens attached to some physical thing like real estate @opensea wasn't theoretical, they had customers and revenue. not just adding tokens

Michael didn't think opensea was a good idea but he could use it. and there were people using it that loved it it's hard to get honest feedback in web3 because people can make money off speculating on it the feedback loop is broken in web3

10th tarpit idea - another social network people are VERY defensive about these ideas they really believe they can make it Michael's advice is do it - don't pitch when it's not yet launched

non consumer ideas: there are less tarpit ideas in non consumer land because many just got stuck and die in consumer land this topic will become a part 2 because we've run out of time...

Q&A with @KatManalac cliche college student ideas? Dalton says just experiment on fun ideas in college. Have fun. Don't worry too much about VC funding. Michael sees many people trying to be at the next stage instead of enjoying the college experience

when should I pivot if I'm in a tarpit idea? if you can launch fast and fail fast - do that the problem with tarpit ideas is that you can get stuck in ideation phase for months

Like this tweet? Follow me for more @afscott You may also like this other @startupschool thread that I made about finding startup ideas: https://twitter.com/afscott/status/1544364463159066625