- Eric Mijakovsky, the Startup School course facilitator, and Dalton Caldwell, a partner at Y Combinator, hold a live Q&A session.
- They discuss the topic of pivoting and the YC admissions process.
- Dalton advises that founders should consider pivoting instead of shutting down their company completely, as it is easier to change the idea while keeping the same structure.
- He emphasizes that founders should focus on product-market fit and not waste time on unnecessary corporate changes.
- Dalton also mentions that founders should be aware of their strengths and weaknesses, as some ideas may require sales expertise that they may not possess.
- They discuss the importance of metrics-based decision making and getting customers to validate the product before making any major changes.
- Dalton suggests running experiments to test different user segments and determine if the product is resonating with the right audience.
- They caution against getting caught up in the excitement of an idea without considering its feasibility and practicality.

- The importance of turning a vague startup idea into a concrete product and asking for money is emphasized.
- Validating customer interest and willingness to pay for a product is crucial.
- Financing risk is a challenge for hard tech companies, and a plan is needed from the beginning.
- The example of a robotics company that successfully hacked together a product and built a financial model to fit customer budgets is given.
- The need to validate customer demand and create an MVP before scaling up is highlighted using the example of Tesla.
- The importance of focusing on specific market subsets rather than a total addressable market slide is discussed.
- Investors asking about TAM may not be genuinely interested and it's important to make the overall company seem exciting in multiple ways.
- Founders being specific about their target market can lead to more engaging conversations with investors.
- The question of where to find case studies showing before and after pivots is raised.

- The video discusses the best sources for case studies on before and after pivots in startups.
- The speaker mentions that Startup School is a good resource for detailed information on this topic.
- Triggers for considering a pivot include working on an idea with no market feedback or having no customers.
- If users consistently suggest additional features before becoming regular users, it may be a sign to pivot.
- The null hypothesis for consumer products is that 80% of signups never come back, so if a product performs better than that, it is interesting.
- Founders should talk to each other about their progress towards product-market fit.
- In YC applications, founders can mention if they are considering a pivot or have other ideas.
- The speaker advises against submitting two applications and recommends putting the best foot forward.
- A case study of a successful pivot is mentioned, where a startup changed their idea mid-batch in YC and received funding.
- Social apps may face challenges in standing out due to crowded markets, while B2B apps may have an advantage in getting first customers.
- YC has a search feature to find applications in specific domains, and there is likely someone at YC interested in each domain.

- Building successful social apps often requires experience in creating and growing previous projects with millions of users.
- When pivoting a startup, it is context-dependent on how to explain it to current investors, but sophisticated investors will usually be supportive.
- YC accepts applications from startups targeting markets outside of the US, as they focus on connecting founders to Silicon Valley resources and networks.
- For B2C apps, it is important to consider the cost of acquiring and retaining users, as negative unit economics can be detrimental to the business.
- When applying to YC, a clear and easily understandable one-liner about the company's purpose is crucial.

- The speaker discourages people from trying to hack the system or manipulate the application process for Y Combinator (YC) funding.
- They emphasize that YC is interested in funding companies that are genuinely doing something good, rather than those who focus on gaming the system.
- The speaker mentions that YC partners read every application they receive, and if applicants put effort into their application, YC will put effort into reading it.
- They highlight the importance of focusing on building a good startup rather than trying to raise money through manipulative tactics.
- The speaker also mentions that YC is still interested in innovative non-profit startups, especially those with a business model that generates income.
- In terms of interviews, the speaker advises applicants to behave like normal human beings and engage in a natural conversation rather than reciting memorized scripts.
- The speaker acknowledges that the coronavirus may affect startups, but advises being flexible and making smart strategic decisions to adapt to changing circumstances.

- Startups that can quickly and flexibly react to market changes tend to succeed, while those that don't have a harder time.
- Companies usually die due to running out of money or the ability to continue working.
- Pivoting a service business to become a product business is difficult.
- A common case is founders finding customers who will pay them to manually do what the eventual product will do.
- The acceptance rate for international applications to YC is similar to that of US applications.
- YC encourages companies to come to the Bay Area for networking, but it is not required.
- Having a provisional patent for a mobile app may not provide much value when applying to YC or pitching investors.
- Software patents are generally not seen as valuable in the industry.
- Focusing on an MVP and getting initial customers is a better use of time than creating an elaborate IP strategy.

The Startup School course facilitator and a partner at Y Combinator discuss the importance of pivoting and the YC admissions process. They emphasize the need for founders to focus on product-market fit, validate customer interest, and create a concrete product before seeking funding. They also advise against getting caught up in the excitement of an idea without considering its feasibility. The importance of finding case studies on before and after pivots is mentioned, with Startup School being a recommended resource. The speakers also provide insights on YC's application process, the challenges faced by different types of startups, and the importance of genuine intentions and effort in applying for funding. They mention that YC partners read every application and encourage applicants to behave naturally during interviews. They also discuss the impact of the coronavirus on startups and the need for flexibility and smart decision-making. The summary also includes information on the acceptance rate for international applications, the value of provisional patents, and the importance of focusing on an MVP and initial customers rather than an elaborate IP strategy.

I'm the Startup School course facilitator.

And in addition to my partner duties, I run admissions.

He's in charge of figuring out who gets into YC.

So today, we're going to talk a little bit about, answer some questions that you asked about Dalton's video lecture on pivoting and the YC admissions process.

First question from Danielle at Squid Bio.

At what point does pivoting turn into actually shutting down the company you're working on and starting a new 1?

Usually, founders are too willing to start a new co.

And just think about it in terms of the amount of work you have to do.

It's much easier to just change your idea with the same Delaware C-Corp that you created if you use Clerky or something like that, or if you use Stripe Atlas, and there's no need to redo it.

I think the major reason I've seen someone create a new company from scratch is there's like legal liability on the old company or.

But for the most part, you know, it makes sense to keep the same cap table, to keep the same investors.

This is an example of work that is not productive work.

A bunch of entities and talking to lawyers.

This is a great example of things you could spend time on that are not work.

And you don't have to change the corporate name right away.

that Jawbone was called Aleph for 10 years before they renamed it to Jawbone.

So I think, in your case, Danielle, you were saying that you have some grant money and you're already incorporated.

I would recommend just sticking with that.

Plus, the main goal is to get the product market fit.

Yeah, spend time on that part and not on the corporate shell game.

This is something you could spend a lot of time on and then it has basically 0 value to whether or not your startup works.

Whiplash Gaming is an interactive watch party platform for e-sports.

Stalton, you mentioned in your video that founders might pivot to avoid hard work, like sales.

How would you make the judgment call between the product not being good enough, justifying a pivot, and the sales efforts not being good enough, justifying a different strategy, different channel?

So It's totally the case that there are good startup ideas that aren't right for the founders.

And so it doesn't really matter if the reason is the founders are bad at sales and will never get better, or it's a bad idea.

And so you want to watch out for 3 levels of hypotheticals.

Like sometimes when founders end up in a hypothetical land about why the thing isn't working, again, this is an example of something you can spend a lot of time on, have a lot of conversations about within the co-founders, ask a lot of advice.

And so spending too much time diagnosing it, I would argue is not great.

And again, I hear where you're coming from, but sometimes there's really great startup ideas that require lots of sales expertise.

do, it's not going to work, even if it's secretly a good idea.

Also, at the early stage, sales is identical to talking to your customers and responding to their direction.

So if you can't get past that first stage, that's probably more of an indication that maybe the idea is not 1 That's picking up steam with with customers Have you seen any cases where like very clearly the founder market fit is unaligned?

And there's like a structural problem to getting to that next stage.

Like say you've got a couple customers, some early users.

I just see a lot of times where there's founders whose primary modality is writing code and Maybe they can talk to other programmers, but they end up with an idea that involves selling to insurance companies or something.

Which is why so many great founders are solving a problem that they themselves have.

Because by definition, They are the customer.

We're now pivoting after realizing our solution didn't solve the problem we were addressing.

Yet we're still trying to solve the problem as we believe it's a big 1.

Their problem that they're solving is 40% of returns on clothes are because they have the wrong fit.

So we're now launching it as a discovery tool that shows clothes on your body for new and used clothes.

On 1 hand, we're concerned that we should only focus on newer used clothes, but we also feel that a discovery tool should have both, as 50% of shoppers shop for either 1.

Do you think that we should start with just 1, or offer the full range as an MVP?

Yeah, so if I were doing office hours with a company and they asked me that question, my answer would be, I don't know what are your numbers, what are your sales?

And the reason is you can end up weaving this really complicated web of conceptualism about we think this, we think that, and if the current product has 0 users, 0 revenue, 0 traction, I'd be like, well just go get anyone to pay you a dollar.

Or to make 1 purchase on the platform, in either type at the beginning.

And It's so tricky when I talk to people and they have these really complicated intellectual arguments

they're like, well, we don't really have any customers per se, Eric.

It's like, well, how did you end up with this super elaborate theory of the market and you've gotten no customers?

The numbers aren't in here on this question, but I usually encourage folks to try to make metrics-based decisions about if they're like, should I do A or B?

You look at your numbers, and it's much easier to not get caught in conceptual land when you're doing that.

And what's also true is that we see companies across a wide variety of domains, products, technologies.

Interestingly, this kind of advice actually works at a meta level above the particular thing that you're working on.

Like, you really can't go wrong with finding a few people who want to buy the first version of whatever you're making and incrementally add on top

The best time you know when to pivot is you just fail to get 1 customer.

1 live customer really using your product like that's the bar if you're below the bar So it's not even a real thing and if you're above the bar, then you can have a sophisticated conversation

So in this case, like I wouldn't be too worried about Like whether your product offers everything at the start, I would worry more about can you describe what you're doing in such a way that it gets a couple people excited right at the beginning.

And they really use it, not just excited.

They're like, yeah, oh, it's returns for retailers.

And you can say, wow, we've done 100 returns.

And even if it's just used, or even if it's just new, either 1, as long as you have some degree of traction, then you can iterate and work with that.

Marble is a collaborative learning solution for teams.

Automates learning, turning assets like links to courses, videos, and articles into a learning plan.

Question is, how do we know if we need to pivot the design functionality of the product or continue searching for the right user?

OK, well, I would first say, is anyone actually using it?

There's people on this end of the spectrum, which they pivot constantly, but they never actually make anything.

Yeah, what they mean by pivot is we talked about it.

And then there's the other end of the spectrum, where you write code and you build a product, and it's been live for like a year, and it's not working.

1 kind of actionable thing that I offer to founders here is to set up experiments that they could run in usually about a week or so.

And you can run experiments on the product side by either adjusting or changing the actual landing page or the things that your product does.

But you can also run experiments on who you're trying to attract as users.

So for example, you could lock your product in and say, because every good experiment has a control.

You don't change both things as you're running an experiment.

You don't change the product and the user base.

So maybe you say, OK, the product is doing good.

But we really want to explore to see if teachers are the right people, or whether it's parents, or whether it's whatever.

So over the course of a week, you say, I'm going to try to get 20 customers in this new customer segment.

And then at the end of the week, if you hit that, it's a pretty good sign that you're onto something.

And if you totally fail and you get like 0 or 2 people, well, maybe that's not the right group.

Tougher part is when you hit in the middle.

There's actually a subtlety in the question.

The question, 1 part says, when you pitch the idea to people, they are very excited.

However, when they test it, they're not finding the use case themselves.

So there's a whole variety of ideas that are very exciting and that I would call them mirages or something.

Eric, I have a startup that makes delicious food.

It is the most delicious food you ever had You love it.

You're excited about this delicious food, right?

Why wouldn't I that does not mean this is a good startup idea?

And so, you know, oh Eric don't you wish you didn't have to work as hard and you got more work done?

pretty much everyone you ask the question to will

And So if you phrase this particular idea as like, we're a learning solution for teams, would you like your teams to learn more?

That excitement is not actually a startup idea, in my opinion.

And you have to turn it into an actual product.

Yeah, and usually it's a problem that they're having in the course of trying to have better learning.

So Eric, do you want your team to learn better?

It costs $20 per user per month, it's a Slack plugin, and what it does is it sends them daily reminders about whether, you know, with daily lessons.

Like you have, The second part you really need in there, which is to offer a concrete product and solution and ask for money, the second part is what you need.

The first part is not, it's so like vague as to not even really be a startup idea.

And you can run that in 1 of your experiments.

So you already have these people who are kind of casually interested or excited, you can try to put people on point and say like, okay, we talked to 20 people, how many of those 20 people are actually willing to pay us just for our MVP?

You're seeing a lot of this right now with remote work.

Everyone's pivoting to remote work stuff, I get why.

But just saying, we're doing remote work, that's not a startup idea.

You have to turn that into, here's the thing that we make and how it works.

Until you nail down the specifics, you just saying we're doing a remote work startup is not going to get validation.

They'll be like, oh, that's very interesting.

That does not mean you have validation on your startup idea.

Next question from Joey from the company Nurture.

Pivoting mobile apps is a lot easier and less painful than, say, building rocket ships to Mars.

What are some of the best tips for hard tech companies to avoid wasting time and money to avoid a pivot later down the road?

So we funded a lot of hard tech companies.

And I think you want to think about this in terms of risk.

And the biggest risk for a software company is they don't make something that people want.

And that is the most, you know, when we see our cause of death, that's pretty much it.

They spend too much money and they don't get the product.

Yeah, it just doesn't happen for them, okay?

What's harder about hard tech is you have this other kind of risk, which is financing risk.

And that is to even show up, we need this amount of capital.

And whether or not the product is good, whether or not people want it.

Like there's all these things that apply to software that you have this additional financing risk thing.

And so if you don't have a plan around financing risk, and I would love to hear your thoughts, you're an expert on this, like that's gonna, you need a plan from the get go.

And even having tons of consumer demand, even having prices, even having Kickstarters,

have an expert here, even have lots of people giving money in Kickstarter, if you have financing risk, it's going to be tricky.

So to kind of dive in on this, financing for a hard tech company is not just about paying for the engineers and the people who are working at the company.

It's also creating a business model in which customers can actually buy the thing that you're building.

So we have a bunch of companies that are using robotics to solve labor-saving activities in the workplace.

This is a really hot subject, and we've invested in a ton of great companies.

But 1 of the things that we always talk about with our kind of hard tech, let's say, robotics companies is figuring out, once you design this product, we have, for example, a company that picks, a company that makes a robot that picks tomatoes from greenhouses.

So it drives along the greenhouse and picks tomatoes.

They had to work really hard to build an arm that would find the tomatoes.

They had to build a vision system that would process it and figure out which tomatoes are ripe.

And then they had to improve it to make it faster than an actual human so that it actually made sense to use as a product.

1 of the things that they did really well is they were super lean.

Like, they managed to do all of this on like a couple hundred thousand dollars, which most people, when they think of a hard tech, they think like millions of dollars of R&D.

You can actually hack together a pretty good, even hard tech company.

And then what they did is they talked to customers, and in this case, this is greenhouses, and they figured out what was the buying time frame?

How did the greenhouses run their budget?

So they actually built a model that was almost better than the greenhouse's actual financial model to understand who was paying what and where were the actual ROI opportunities?

And then what they did is they said, we're going to build this robot.

And then we're going to charge our customers in such a way that it fits into their budget, into their line item.

So if you're a hard tech company and you're thinking about, how are you going to finance this from customers, you have to consider how do they earn money, how can they pay you, and then potentially even find people who will finance it or give you loans because the customers aren't used to paying that much money up front.

And final thoughts, everyone looks at Tesla and that's the shining light of people that tried really ambitious things.

I don't know if everyone knows this story, but there was a time where they had a big financing risk.

Yeah, there was a time when he took his entire net worth, which was sizable at the time, and put it into the company because they would have died otherwise.

And so it would be very hard to bootstrap.

What's tricky about certain startup ideas is no amount of the idea being good can help you from financing risk.

And so, yeah, the reason we get to enjoy the fruits of Tesla these days is there was all these times they almost died, maybe like half a dozen.

And you needed like an act of God to keep the thing financed.

And just to touch on, sorry, to touch again on the question that you're asking, so it's like how do you avoid a pivot later?

The main way you can do this is by validating that customers actually want to buy what you're making, very specifically.

People bought the Tesla Roadster immediately at a high price point.

The idea was to start with a very high margin, high end thing that he could just sell to basically his friends.

Validate the market, validate that they could actually build it.

If he started with a Model 3, I don't think so.

Another way to think about it is you can have this large vision of where you want to go, but try to figure out if there's even an MVP of a hard tech product that you can deliver.

Elliot from Unify asks, what is a good way to pitch a market opportunity when a standard total addressable market slide would show all users who take pictures?

I think the question is kind of getting at, how do you provide subsets of markets that you could reasonably say everyone could be interested in?

I guess it comes up if an investor says, oh, well, what's your TAM?

But when you're at a really early stage, it sort of doesn't matter.

I remember 1 time an investor gave a talk that I attended and they were talking about Uber's TAM.

And when Uber was very small, it was only these cars, like limos, basically, like private cars, black cars, that you would.

And they did a TAM, bottoms-up TAM, of the number of black cars in San Francisco and how much Uber could charge.

And I think the TAM, it was something silly like $10 million.

And it just shows how this particular exercise-

Airbnb, what's the TAM of people who want to sleep on other people's floors when they go to concerts in other cities?

I think what we're saying is that the investor who's asking that is asking kind of like a bullshit question.

And what they're really saying is, I don't think this is interesting to me.

And this is a nice way to say it, is, oh, I'm worried about the...

This is really great, but I'm worried about the TAM.

It's just a nice way of saying I don't want to invest.

And the best way to combat that is to have the overall company seem exciting in numerous different ways than thinking you can debate them with a really good TAM slide.

Or here's a potential path that you could take people down.

I always love it when founders are actually, like it's counterintuitive, but when they're actually like very specific about the market that they're trying to go after, because then I can visualize their product being used by that market, and then saying, they could easily knock this out of the park.

If you go after that, no 1 else is going after that.

And then you get into this conversation where the investor is actually saying, well, actually, it could be used for this category.

And as the founder, you're like, obviously, I know this.

But it's this more back and forth where the investors are actually generating ideas and participating in it.

Trey asks, what are the largest best sources for case studies that shows kind of before and after pivots?

Are there any metrics that we should use to track?

I don't know about a library of case studies, and probably the best equivalent of that is literally this.

It's startup school, because in my talk, I'm not aware of there being other people that have talked about this in this level of detail.

So I think this is as good as it gets, unfortunately.

In terms of triggers, it's usually where you're either working on an idea that is not falsifiable and it requires you to work for 2 or 3 years.

I'm exaggerating, but for a very long time with no market feedback, that's a pretty good trigger.

Or you have been doing it and it's just a goose egg.

Like 0 customer, like the most, it's so funny how many people, the signal to pivot is just like, literally no 1 will use it.

Here's 1 of the triggers that I use when I think about products.

And it's if there's always like 2 more, like if you give it to people and they always say there's 2 more features that they would use before they actually become regular users.

See all Y Combinator transcripts on Youtube

YC SUS: Eric Migicovsky & Dalton Caldwell discuss pivoting & pitching