Makes me think of something I say occasionally... you need a problem so compelling that customers will pay for a buggy, incomplete solution, because a buggy, incomplete solution is what you're building.
It's the dream of every engineer to get to make something that doesn't suck, right? But the 80/20 rule applies. Getting through that last 20% that is suboptimal is 80% of the work. Every hour you spend polishing your way to perfection is an hour you don't spend adding new features or creating new products. Every bug fix is paid for with opportunity cost - at a very high compound interest rate, if you're in a startup.
I've come to the conclusion that if your code doesn't kinda suck, you're doing something wrong. You're wasting effort. Actually, I came to a much harsher conclusion than that - I've concluded that all software sucks. That all software hovers near the line between "barely works" and "doesn't work". Why? Because by the time you get to feature-complete, if it still works, it's so riddled with compromise and bad decisions that you never want to look at that code again. You want to start something fresh and new, and THIS time, you won't make all those stupid mistakes! This time, it won't suck!
I think it's possible to write good software, but I agree that it's difficult to do at a startup. At the very least, there are kinds of software (aerospace and medical software for example) where hovering between "barely works" and "doesn't work" isn't acceptable, and the people responsible found ways to improve their software beyond that point.
It's not really making software that more than barely works... it's changing the definition of "barely works". I've spent most of my career in banking and health care, stuff under heavy regulation. The regulations become part of the requirements. And within those bounds, banking and health care software barely work. Avionic software barely works. The software that keeps your pacemaker going? Barely works.
Remember, a pacemaker that doesn't work 1% of the time doesn't work. Specs are different.
This kind of starving crowd is rare and you should expecting it to be rare. That's why people are willing to pay you millions of dollars if you find one.
If you find a truly hot market, it can be worth billions and most any idiot can make a lot of money in a hot market. If you are good, all the better.
But remember, it's rare and hard to find. Assume that 99.5% of all businesses aren't in that kind of market situation.
It's a charming ideal that I think is vanishingly rare in real life. I also can't see how the hair on fire thing fits with any of the top startups and ex startups I know. In fact I dare say the rapidly rising user count is a separate story from the hair on fire story. The hair on fire tends to be a specialty business problem and still, against all reason, tends to require a pretty rigorous sales cycle. The problems I work on, making the medical record discrete data is of this kind. The other kind is more like Google, Facebook, Uber, AirBnB and those were surprises because no one's hari was on fire. We didn;t know we wanted that stuff until we saw it and then we wanted more and more.
In summary, I guess I disagree with the whole thing
I think there are actually two aspects to the market question. First, there's the level of pain experienced by the customer, and second, there's the size of the market. Lots and lots of startups are solving "hair on fire" issues for relatively small markets - enough to generate millions in revenue, but not billions.
For a sad case in point, there's not a whole lot of research for a cure (or even a treatment) for ALS - a disease that is completely debilitating and almost always fatal. There just aren't that many patients (even though it's common enough to touch famous people like Lou Gehrig, Stephen Hawking, and Jason Becker). Meanwhile, tons of money are poured into Viagra and its competitors - not an actual lifesaving problem, but a huge market.
I know about Airbnb personally and it started with a hair on fire problem - a design conference where all the hotel rooms in the city where sold out. Brian and Joe hosted attendees in their apartment on air beds. That sounds like a hair on fire problem and a brick solution :)
I see your point but that implies that on rare occasions hotels are sold out at a specific time and place. I don't think that is the problem AirBnB solves
I'm wondering if this may dovetail with another piece of common YC advice, "start by going narrow & deep".
I didn't start using AirBnB until 2013, and I used them because they were remarkably polished and made the experience super-easy. They weren't at all a "hair on fire" problem for me.
But I could imagine that at some point in time, there was a group of people who were so desperate for a place to stay that they'd crash on a couple of airbeds at a stranger's place. And that gave those strangers the idea that maybe this was something worth looking into and investing time into perfecting.
The importance of that initial core group of users may be feedback. I've had to give up startup ideas because I launch them and then there is literally nobody who cares. And then with no users, there are no guideposts as to which direction to improve the product in. It's all guesswork, and random ideas don't have a particularly high chance of success.
This may also dovetail with the known advantage that you get from solving your own problem. If it's your own problem, then you know at least one person wants it, and you can use yourself for feedback on improving the product. Then hopefully once it's good enough for you, you can find some other folks that'll find it good enough to try out, and use their feedback to evolve the product.
Great write up. You hear many times the advice of getting MVP out and then "just" listen to your customers and the data. A lot of times, as you said, there is virtually nobody that cares and few feedback. You are just treading through a desert without nothing to guide you...
I don't remember their story in detail but I think they started as a couch surfing alternative. Given the rule of supply and demand I'm guess there was not "no hotel rooms available" there was "no hotel rooms available cheaply". offering a cheaper alternative is great, but I can't consider it a hair on fire problem
Cool analogy. However don't some startups actually open up things that we didn't even know were painful?
E.g I was happy staying in hotels. I wasn't on fire looking to stay at someone's house. I'm guessing the hosts weren't on fire either having random strangers over.
Facebook, Google, Amazon. I was very happy with the alternatives. They just did somethings better. Although Facebook is kind of dead to me. Haven't opened in months.
I do like the analogy of getting overwhelmed by customers and having scaling problems.
If you didn't know you were in pain, you weren't in pain. If you didn't know your hair was on fire, your hair wasn't on fire. I suppose if you squint hard enough you can make any of these support the authors contention.
Hey Michael! Great piece. Signs of product/market fit are often strong and unambiguous, and that needs to be emphasized.
That said, many startups get v1 wrong. Not necessarily because their MVP is too clunky, but because they fail to frame (or graap) the correct problem in the first place. Or because their solution to that problem just isn't the right one. (As you aptly put it, founders can get hung up on their solutions and take their eyes off of problems.)
I'd love to see a followup piece about where to go from there. Let's say you're a founder, and your v1 failed to achieve fit. Your v2, perhaps, failed to achieve fit. What is a creative, but rigorous way to go back to the drawing board for v3?
I've worked for some successful startups, and I've worked for some failed startups. None of them found fit right out of the gate. The difference seems to have been that the successful founders recognized it quickly and took intellectually honest steps to do something about it. The failures either avoided that reality, or floundered while trying to adjust.
So what are some effective methods for assessing and course correcting in MVP customer development and product development? We know that "getting outside the building," to use the Lean parlance, is critical. So is effective brainstorming. But what are the best techniques for doing either?
I strongly second this request. As a founder working through a v1 and currently thinking a great deal about product/market fit, it would be fantastic to hear more about this.
Thanks for writing this, one of your quotes "Founders often hold too tightly onto solutions and too loosely onto problems" really resonated.
If someone were evaluating their next opportunity to tackle, would you recommend they focus less on identifying a specific pain point (that they may know how to solve), and instead identify a large market and then search for inefficiencies (even if they have less domain expertise in it).
Obviously the intersection of the two would be better (large market with a pain point you know how to solve) but curious which you would optimize for if you had to choose.
I think solving familiar problems is a curse of the startup community. Founders go after problems that are familiar, but trivial. Product finder apps for vanishingly narrow segments. Task management. Or my favorite category, startups that are supposed to help startup founders/investors evaluate startup ideas!
My recommendation is that if you're a "technical founder" - a programmer who wants to be CTO - don't look for problems to solve. Look for business partners who already have domain expertise and a problem that they can explain succinctly (so you see the value proposition), and work with them. There's a whole world of people who can see a massive problem that could make money with a good technical solution, but they don't know how to code and don't really know anyone who can code and is willing to help them.
It seems a counterproductive direction given the recent years explosion of startups and e.g. app development.
People and their pain are very different in general, but rather similar in specific groups. Almost everyone spends almost all of their time in a bubble that excludes the pain points of most of the world. And, for someone with the skills, possibility and desire to run a startup, it's exceedingly likely that your market is oversaturated already.
Pick any demographic metric.
For age, the personal pain points of 15-30 people have an overabundance of solutions, but the pain points associated with 53 year olds or 9 year old kids have more gaps and possibly even more paying power.
For lifestyle, the pain points of single young adults are oversaturated while needs of large families (possibly excluding baby-apps) are not, again, due to the peculiarities of startup employee demographics.
For types of location, even the minor pain (inconvenience?) points of people living in highly dense urban areas have all kinds of startup solutions, while pain points specific to millions of people living in USA rural areas are pretty much ignored.
For social groups, the subcultures popular in Silicon Valley have their specific needs filled much more than the various minorities and cultures which are mostly elsewhere. Pardon me for a crude example, but for example, there is a competitive market in gay dating apps/services with a lot of solutions developed; however, the USA LGBT population is comparable in size and thus purchasing power to the Mormon population - if you're starting a new service, which market has the larger potential for you to serve a need that's not already served by existing solutions?
For technological aptitude, the needs of early adopters are obviously served by new tech, while you definitely can build solid businesses designed to serve the far more numerous people who want to avoid bleeding edge tech or even don't have the skills or desire to use industry-standard tech - if they still have the same needs, then making that tech more friendly to them or more distant from them (having you do stuff for them by that tech, instead of having them interact directly in detail) will bring you good revenue.
This (and Andreessen's article) is quite insightful of knowing when you're on the right track or not, but it's more like a "you know it when you see it" kind of advice that is not very helpful in terms of how to get there, or how to design a process that allows you to methodically identify and jump into these market opportunities. Maybe a second article could dive more into that.
I agree with your point 100% - if we could tell you how to get there we would! I think the biggest point is that many companies die because they scale expenses as if they had reached product market fit when they haven't. We are trying to give you more time to find the right problem/solution.
People tend to say the same thing about marriage. They say "when you know, you know". Not extremely helpful information. However, you can optimize for that scenario by dating as many people as you can. If you go on 3 dates a week for 6 months, thats 75 dates of worthwhile info you can use to get closer to that objective.
I'd be curious to see examples of what "hair on fire" problems looked like when they were buggy, incomplete solutions to poorly-defined problems.
For example, I know that Google was a hair-on-fire problem: everybody searched for things on the Internet and nobody found what they were looking for. But was Facebook? I started using Facebook in 2004, before they'd taken VC, and I honestly didn't see the point (and still kinda don't). Perhaps I just wasn't the target market (and still am not), but then, how do you recognize a hair-on-fire problem when you're not necessarily in the target market for it?
(Interestingly, LiveJournal solved a hair-on-fire problem for me about 2 years before Facebook came out: how to keep in touch with friends that I'd met over the Internet but lived thousands of miles away. And yet Facebook is apparently a whole lot more successful than LiveJournal.)
Facebook totally was a hair on fire problem - if you saw a cute girl on campus and you wanted to look her up - all we had was a stupid freshman "facebook" (book of faces and names) and a crapy web version of the same.
Heh, interesting. That may have been my problem - I didn't particularly care about the cute girls on campus.
So then, an amended question: given that you do much better solving a problem that you personally have, what happens if you're a person of few desires? Someone who enjoys having a simple life and focuses intently on a few things that matter to them, but doesn't really sweat or even notice many of the things that annoy other people?
It seems like this personality type would be well-suited for entrepreneurship in general, between having a low personal burn rate and being able to focus intently on a topic, but the inability to recognize and get seriously worked up about a problem seems like a killer.
The concept of good/bad market is attractive, but I'm wondering is this is really a good model. How do you account for innovations that require a market to learn its benefits before they want it?
In the beginning, a market for a innovation looks like a terrible market. Few people wanted a PC in the early days. Vacuum cleaners were a hard sell when they were first introduced, and there are countless other examples. All of these markets required that the buyers be educated and they had to be convinced it was worth the effort to learn how to use the new thing.
It seems to be, saying "don't be in a market where it is hard to sell" is great in theory, but honestly I think every truly innovative product has to create their market to some extent.
To what extent is sustained exponential growth a sign of product market fit? And do you view it as binary?
Also what are your thoughts about the long term viability of this approach to building companies? As someone with a bit of a machine learning background I see this whole model as "greedy hill climbing" aka "gradient descent." That will work until we fill in all the micro-niches at the current macro fitness peak. Then these kinds of "lean startup" style fit-and-explode market opportunities will become depleted.
Exponential growth with retention over time is probably another definition of product market fit. I think the world is always creating new problems to fix and we aren't successfully tackling the hard ones yet (housing, healthcare, poverty, etc).
What would be your advice for a team that has clearly reached product-market fit, but the market is inherently small and their growth stalls out because they've put out most of the fires for people's heads in that space?
At that point it's tempting to diversify the product and start solving less dire problems for customers by adding new features, but that makes the product larger and might slow the team down.
That's tricky - this sounds like you might have a lifestyle business (successful product in small market). You can try profiling your customers and see if if there are other companies with the same problems that your aren't serving.
"If you handed them a brick they would still grab it and try to hit themselves on the head to put out the fire. You need to find problems so dire that users are willing try half-baked, v1, imperfect solutions."
this kind of falls into line with Reid Hoffman's famous quote:
"If you are not embarrassed by the first version of your product, you’ve launched too late."
I think these statements paint a wide brush and fail particularly for products whose main value prop is ease of use.
If your product's core value prop is UI/UX/ease of use, you cannot deliver that with a buggy half finished product. And if you cannot deliver the value you intend to provide, then you are shooting yourself in the foot, spooking users losing their trust in your ability to deliver.
Obviously each product is different, but I think specifically for this type, these methodologies should be omitted. Because the most likely case is that your customer is already solving the problem with a "brick", whether that is your competition or a home made solution that they have created. Handing them another brick will not suffice.
"If your product's core value prop is UI/UX/ease of use" then you're obviously not entering/creating a new market, nor you are attempting to find a product/market fit, are you?
It pretty much implies that you're attempting to overtake existing competitors who have found product/market fit, where people have options to fill their need but do so grudgingly because of their UX quality. Doing that can be a solid business plan, but it's very, very different from the type of new-product-discovery startups that Reif Hoffman was talking about.
If the people are paying for an existing crappy product with poor UI/UX, then that illustrates a valid need and building an easier-to-use product can greatly increase the size of that market.
However, if currently people are not eager to pay for that need, then I'd wager that simply providing UI/UX/ease of use won't change that. If their need isn't sufficient to put up with the minor bother of sucky UX (at least for 10% of most needy part of the market), then that need isn't also sufficient to put up with the much larger bother of paying you adequately.
"then you're obviously not entering/creating a new market, nor you are attempting to find a product/market fit, are you?"
Yeah, and if you were entering/creating a new market, usually people do not understand why they need the product you are creating and thus the original analogy still does not work.
Overall I just feel the brick analogy is not great because there are no low hanging fruits anymore. If a problem is so painful, there must exist a competitor OR a home made solution.
Actually, if I think about this more, then the brick analogy is very useful - if you see someone (or better yet, some whole industry/niche) beating themselves on the head with a metaphorical brick, then this means that they have a very, very important problem and most likely if anyone came to them and told that they have a solution that's simply unpolished but solves most of the problem without hitting your head with the home made brick-solution then the expected response to this description would be "shut up and take my money!".
I won't ask you how to identify these markets because I think it's fairly clear that it takes dedication, thoughtfulness, and experience.
I would like to ask if you believe there are many markets with unserved "hair on fire" customers. Are there thousands of these problems waiting to be solved? My intuition is that there are many people with hair on fire problems, but not many of those problems generalize to products that lots of people would want. What do you think?
> new products (if they are successful) often create new problems.
This rings true to me. And I've been accused of doing just this. In 2004, my "hair on fire" problem was trying to figure out how to test web apps that heavily used JavaScript (which really was a rare thing back then), so I created Selenium to solve my dev team's testing problem. Fast forward four years later to 2008, the next "hair on fire" problem to solve was that maintaining a lab of computers dedicated to Selenium testing was labor-intensive and expensive, so then I founded Sauce Labs as a way to rent out machines for testing. Sauce could only have existed because of the problems created (accidentally, I swear!) by heavy use of Selenium.
Not sure you are still paying attention to this thread, but...
When you have a marketplace, would it require a hair is on fire problem for both supply and demand?
I work at a startup marketplace where we clearly have a good solution for a hair on fire problem at the demand side, but our supply side isn't just that interested. They sign up, but their engagement is very inconsistent to say the least.
What problem does your marketplace solve for suppliers? Does it get them extra revenue? Is the amount significant to them?
Do they have to do something extra/different to work through your platform (e.g. different process, different service level, pay commission to you)? Is it still worth it?
Do they use it once and then stop? If so, just ask them.
It solves a problem, that will only get extra revenue at the long term, for which they have alternatives and has no incentives for continuous engagement (it is good enough to use it only once in a while).
Hey Michael, is it imperative to have a problem to start a company and get to product/market fit? I am thinking of successful companies like Starbucks and P&G whose customers are definitely not in hair on fire mode. Or is this framework of getting to product market fit for tech companies only? Your thoughts?
Having taken the long and slow road (9 years now) to building a business and getting to the place where your last paragraph seems to be where I'm at -- reading was definitely an encouragement this morning. Thank you!
Hey Michael, awesome piece! When I started my first company [that made serious revenue relative to other projects], I didn't spend any money until we actually had sales. Getting to product-market fit took awhile for us, as growing the brand and reputation just took time, but now it's an awesome machine that churns well with some spurts of oil here and there.
I think you're spot on with the half-baked, v1, imperfect solution. Currently trying to do the same with a new startup (https://getbeau.com). Our v1 didn't do it for us, but we listened to the recurring feedback and just relaunched our v2 and are continuing to validate that the problem we solve for our customers is a serious (albeit first-world) problem.
Incase of enterprise tech - finding repeatable pain across set of customers is a must. Solving one enterprises "hair on fire" problem is no guarantee of similar pain existing else where.
My observation from being in enterprise mobility market for 10 years - while one customer realizes hair on fire, many others are diabetic and don't feel the pain.
Great article. Another good way we've had this explained is: if someone is in a burning building, you are the option of jumping out of the window.. There may be a 95% chance they will die, but they're in so much pain that they'll make the jump (even though they normally never would).
Do you think this is more geared for entry strategy (getting the tip of the spear in)? In an ideal world, you could find something where building a small quantum converges with solving this degree of pain.. though of course the stars do not always align that easily...
I'd like to find more resources / books (I love audio books these days) that answer some of the questions at the bottom of Marc Andreesen's original post:
- How exactly do you go about getting to product/market fit if you don't hit it right out of the gate
- How do you know when to change strategy and go after a different market or build a different product?
