Y Combinator CEO: “The True Test of Product-Market Fit Is 'Drowning in Demand'”
blog.capitalandgrowth.org
blog.capitalandgrowth.org
> It depends your product. My definition of product-market fit is: you are drowning in demand—your product is being used by so many customers that you cannot handle all the new people knocking at your door!
> Absent that you do not have product market fit. Most people use the term too loosely.
It's easy to sell a dollar for 90 cents and be "drowning in demand". The answer seems to indicate that demand is a pre-requisite, but not necessarily sufficient.
Not only are many businesses actually selling a dollar for 90 cents and therefore think they have product-market fit, but even if you aren't one of these, your competitors might be behaving this way.
For example, when doing freelance/consulting work, you can try to match the rates of cheap freelancers, but that's usually not smart- they might be doing the equivalent of selling a dollar for 90 cents.
Always worth remembering that 90% of new businesses die within a year (or something like that). That means there's always a significant amount of competition that is doing things that don't make business sense.
I didn't really dig into this too deeply, but I'm not sure how accurate these numbers are. Or at least, they're very open to interpretation. For example, see my other comment regarding which businesses we're talking about.
But even taking the Bureau of Labor Statistics data, my immediate questions are:
1. What constitutes a small business? LLCs? Are sole proprietarships also counted? (Definitely relevant when thinking about freelance work, after all).
2. What consitutes failure? Sure, there's bankruptcy. But there are lots of businesses that continue to exist, but are not active. You might very well start a company for doing freelance work, stop doing the work and start work as an employee (which is also a question - failure or personal choice?), and keep the company in operation.
Just to dig even deeper into this example from a freelance dev perspective (what I know most about): Let's say Sarah starts work as a freelance dev through her own LLC. She charges a very low rate. This is the kind of competition I refer to in my original comment. After a year and a half, she realizes that she isn't making enough money for her trouble, and decides to start working as an employee for a company. The LLC sticks around for another 5 years for various reasons.
How should we count this? Is this a case of a business failure? She certainly didn't go bankrupt or anything. The LLC is still around, so she probably counts for governmental purposes. And arguably, she shouldn't count as a "failed business".
On the other hand, if I'm a freelance dev in the area who knows that you need to charge twice what she's charging to get anywhere near a decent salary as compared to working as an employee. From my perspective, this is definitely a business failure, in the sense that if someone tells me "your rates are too high, look at Sarah's rates, why don't you charge less", my answer to them would be "she doesn't realize it, but she's selling 1 dollars' worth of value for 80 cents".
Even assuming we're talking about startups, the article is looking at venture-backed startups, and defining failure as returning less than 1x. This gets two things "wrong":
1. If talking about startups, most people (founders, investors) will consider a return of 2x to be a failure as well.
2. You're already limiting the discussion to startups that have been venture backed.
IIRC in venture-land, the number of startups that are considered to be returning a meaningful return is around %10-%20 for the best funds.
I think the rest of my comment stands though. And especially, if you're singling out specifically VC-backed startups, I'm guessing their 1-year failure rate is practically zero, since they usually raise enough money to survive at least 18 months. That's why the details of what exactly we're talking about matter a lot here.
I think, for all VC-backed startups, there are statistics that show a more-or-less 90% rate of startups "failing" in the sense that VC's care about. But it's really more-or-less- I'm assuming it's a made up statistic that is in the area of the truth.
Her advice is to make sure that you base your "absolute minimum" price on what it takes to sustain your business and not just on what your competitors are charging because the competition could be going out of business and just don't know it yet.
It's almost like saying, "The true test of product/market fit is whether or not it's obvious you have product/market fit."
Essentially, the statement is redundant.
There are various levels of Product/Market Fit. There's unicorn-level product/market fit and there's amazing product in a small market-level product/market fit (ie. $25M-$100M exit). YC and others in the VC-world only seem to recognize the unicorn-level because that's the only level that's relevant to them.
The thing is, this is the sort of thing people _actually say_ about p/m fit. E.g. Marc Andreessen: "you can always feel product/market fit when it’s happening"[0].
It might feel redundant, tautological and unhelpful, but there are a heck of a lot of startups who are really, really trying to convince themselves they're ready for the next step because "I think we've got product/market fit - I mean, we got that pilot with a big corporate, didn't we?".
Statements like this are a reminder to bring over-optimistic entrepreneurs back down to earth and remind them there's still work to do.
[0] https://a16z.com/2017/02/18/12-things-about-product-market-f...
I've consulted for quite a few startups, and the successful ones really do feel different. Are there lines of people standing outside the booth at trade shows trying to write 5-figure checks? Are the sales people bringing in big contracts so quickly that people start mixing them up in conversation? Is engineering constantly struggling to keep up with ridiculous growth?
Mind you, not all of these companies succeed. Some find themselves trapped in a market that can't sustain growth. Others are mismanaged, or ultimately suffer from too much churn. Sometimes Google moves into the market and everything gets harder.
But once you've seen very aggressive customer demand, it's pretty obvious.
This would TOTALLY redefine my personal definition of product-market fit. Thanks!
Just because lot's of people say it, doesn't make it any less wrong or misguided. In my mind, it's kind of like the "great man theory" from the 19th century in that is fails to go deeper into the issue and take a critical analysis of what's really happening [0].
There are many amazing techniques that are 10x more rigorous than this statement, which we've only scratched the surface of for testing Product/Market Fit.
"The attribute of greatness is reserved for leaders from whose time onward history can be told only in terms of their achievements."
You can have product market fit but if there are competitors, life still is hard.
*these are just my eyeball estimates from looking at graphs in google search for "slack growth"
I've experienced "drowning in demand" twice in my career. An ecommerce startup and a dating app. It is intense and you know it when it's happening. It's hard to fathom what that would look like in an enterprise software setting. Large deal size, long sales cycles, high touch everything, it just feels like it's geared differently.
Go.com w/ Overture, acquired by Yahoo - that had PPC offerings before Google even had a monetization strategy.
Sounds like a road map.
(Wikipedia)
(There are exceptions, of course. LASIK surgery, for example. In which case something like NPS can be a better key metric. But generally, repeat business is the best proof that you've got something real.)
And the better the market the more they convert (member conversion rate), and finally pay (customer conversion rate).
If you only measure conversions, you don't understand why they happened and what to improve to make them happen even more.
How is Digg?
Y'know, unless you want to be happy.
I'm so doing that next time (and on elance, et al., coding "talent" can be had for even less than $20/hr). I spent a year coding my current product (b2b) myself, and now I'm discovering user acquisition is the hardest thing in the universe.
Having just graduated with one, and meeting Michael at a lunch where he tried to tell us MBAs why we were bad fits for YC, I actually agree.
It’s certainly not a differentiator and while it got me a great job in a very traditional firm, startups need risk takers more than credential seekers (which I’ve come to understand is something I’ve actually become to some extent).
It makes sense for YC to try to filter out credential seekers, because they’re unlikely to have the right motivation. But if your goal is to have a wildly successful career, and MBA is probably a safer bet than a startup. Most startups - even well-funded YC startups - trundle out of the gate, fall over and die.
And the reason it stopped being great is always some variant of "..and then the MBAs came."
So, getting an MBA has had very little impact on my career/life. In fact, considering the time/effort involved, I'd say it has probably given me negative value.
You may not realize it but an education like that will come across in the language you use and the way you present yourself to executives. It won't help in a startup, because there are no executives, only doers, but in a larger company being able to hold a conversation with executives on thier terms and using thier jargon will make you far more effective than someone who can only talk in programming paradigms.
I've seen people grind on a startup for years, pivot multiple times, and have nothing to show for it. Then they try another totally different idea and it's an instant success from day one.
Another analogy is viral music videos, even though some videos have all the ingredients they never take off, while others go viral. Same is sometimes with not-so talented getting popular due to various factors. Another analogy I could think would be movies or game production, no one can really know when it will go viral, but you can do some calculated risks and make sure that it will become at least a modest success. Like sequel to a hit movie or a poker/gambling games that won't go viral but can give consistent returns. That can be thought of as b2b in software markets, while b2c leans towards viral/large success stories..
Isn't that a bit irresponsible?
I, for one, had to taper my startup ambitions for a year after graduating university because I had student loans to pay and an initial bank account balance of zero.
Some people are a bit more privileged to have a bit more of a safety net when they start, so jumping off the deep end is a bit easier and less risky.
Many tech companies don't see profit for several years. I don't see how that's possible for some aspiring founders without someone like YC giving them that initial boost.
Despite this "strike," I'm fully self-sufficient now with my own business that is ramen profitable despite having a bit more of a barrier to overcome to start with (which meant biding my time at a Big Company and legally moonlighting by doing remote work with my cofounder while building my business).
Is it not equally impressive that someone started with an empty bank account and worked a full time job for a year and then worked 8+ more hours each day after work to bootstrap their own company?
Please keep this in mind, if at all possible, when deciding whether someone is "fully committed." I hope this makes sense.
The real challenge is to work out when your product isn't being snapped up by users not because it CAN'T be awesomely useful and demanded and valuable, but because your marketing has not effectively sold it to them yet.
> Foolish stubbornness comes not from working on the same problem, but from an unwillingness to iterate on the solution. Many founders fall in love with their product rather than the problem they are trying to solve.
That is for founders who can personally code their solution. Imagine the situation for founders who can only make a proof of concept prototype and need to raise money for them to see a commercial version.
In this case the desire to see their idea in physical form is so great, they would work unpaid for 1 to 2 years for a 20% chance at seeing it. That means their personal demand is $100,000 for a 20% chance to have 1. (And this is why many engineers work for equity / often called a lottery ticket.)
That is another way to see that the world is drowning in demand for your product/vision :) People will donate hours or market salary to having one.
(I realize that the last three paragraphs are mixed with potential financial reward - however for people coding the solution for themselves, this definitely shows that they themselves are drowning in demand for it.)
Depending on the business that is, but if there's an unsolved need from the market, wouldn't it be that no one is fit for the market? If so, then the person who's "drowning" is the most qualified of all (given at that moment).
And it can have bad consequences: your customers might get unhappy because of delivery delays, lower quality, you might try to adjust the product/price loosing early customers or making a mess, your thinking and doing will be in a hurry, ...
Overall, the trend might not last, and others might get into that market in a better shape than you.
Market-fit is more about the long run: being able to have regular steady growth, of customers, of the revenue, being able to learn and adapt to it, having positive feedback again and again and again, ...
I think this "drowning" idea is a pure bro/VC one: you demonstrate in a short time that you can attract a lot of customers, you get good press (for that short time), investors, millions (or billions) of money pours in, and it's not really about the customers or the long game of making a product.
I'd rather have a positive feedback that my business is growing in the right direction but can't keep up with demand than being the never-ending void of finding a product-market fit. This is purely my opinion (so I don't expect everyone should do/think this way), I just feel the trade-off is better.
"Overall, the trend might not last" - I love this bit. I think this is one of the biggest problems with finding a product-market fit that is often mixed with "catching waves" (think AI wave, sharing economy wave, blockchain wave). How do I know which is which is the driving force of my product when it's mixed up?
You're 100% right, market-fit is about the long run and there's this tendency in tech culture that everything should be fast-tracked to become a unicorn, hyper-growth, press coverage on TechCrunch. It is a purely bro/VC idea.
But iterating your product/business to the best/biggest possible realization that it could possibly be can lead you to get "drowned". At the end of the day, when coming on the crossroads at that point, we should ask if we want to "go big" or "stay as you are". Both are equally accepted answers and comes down to taste and preference.
We all hate the toxic tech bro/VC culture telling us not going big is akin to being a failure and we don't talk/promote enough the idea it's absolutely fine to go slow and steady (my favorite example would be Basecamp). Tech bros are all in to get a few seconds of fame and VC are in maximizing their returns in the shortest possible time. It's unsustainable and unstable idea. I think it's the primary root why there's a diaspora of engineers, VCs, startups in silicon valley.
How would you estimate the chances of success on something like that. I estimate them at roughly zero in spite of the technological prowess, but maybe I’m wrong.
Yes if your goal is to build a Dropbox, an Instagram etc then you don't have "product-market fit" before you have to stay up all night weeks in a row to keep your service alive.
But sometimes it makes more sense (and joy) enjoy building stuff where the product-market fit and growth curve is a lot slower, read Basecamp's post about linear growth companies: https://m.signalvnoise.com/the-world-needs-more-modest-linea...
Background: I have build a VC backed company that ended up on a linear growth curve and enjoyed it a lot, today the company is profitable and proud. Today I am VP Growth of a heavily VC backed company that is trying to hit product-market fit.
Similarly, Dataturks can have demand from a stream of different companies, each doing their isolated projects.
Not a big deal, but confused me at first.
the "demand" is from people who like seeing lines that go like this / in their portfolios
the ratio of people who are actually using the "product" to those who are just speculating is pretty insanely low
2. Acquire users
3. Own distribution
4. Iterate to multiply 1