Ask IH: Why is there still guessing involved in "making something people want"?
On the bottom of Y Combinator's website, it says "Make something people want." There are troves of data about the American consumer and information about people's desires is so public and easy to obtain. Why is there no easy way to figure out what people want so you can just focus on building said thing?
I think the problem is that people don’t know what they want a lot of the time. No data trove could have possibly predicted the meteoric rise of Instagram or Snapchat, to say nothing of more obscure SaaS companies.
With B2C, it's kind of different story, because consumers react more emotionally than rationally. So it's not enough to build something they want, but to make them actually want it and then keep using it. With B2B, it all comes to how much value you bring and how can you prove them that.
Valid points.
Most of us have heard and said Henry Ford's: "If I had asked people what they wanted, they would have said faster horses."
If you are a non-free SaaS, it's also that users find enough value to pay for it.
I have been piping all of Paul Grahams great hits to "ReadAloud" and listening to them. It's all great stuff. But it is contextual. He is very straight forward when we talks about building companies with no revenue. My perception is that most people, here, are looking to get paid. This upends a lot of the Silicon Valley wisdom and strategy that gets passed around.
As a friend told me: "A bar opens in San Francisco. A million people attend the grand opening and buy nothing. It's heralded as a raving success!"
yes these are obvious exceptions, but I am sure a lot of consumer needs are hidden in plain sight.
The saying has merit, but it’s really a starting point to identifying a need in an identifiable market.
Things were a little less mature in early stage startups (maybe they still are) when Alexis Ohanian started using the saying as part of his talk to startups. It was easy to find a team working on a solution that didn’t have a market.
I still see it today, but not as often. I see a technical team building a product because it is of interest to them. There’s a whole process of market research that is skipped or misread. The team goes out to find a market for a solution they built and quickly finds that they built the wrong solution for the wrong market.
If someone built a product and they are stuck asking “how will people want to use this?” Then theorizing how to convince different people to use their product for their problem, then they didn’t build something people want.
You need to start with the user/customer and identifying their needs.
One company I met built video streaming software, but they did it because of their interest in the technology. After they spent 6 months building their product they tried to think who may need their product. They turned to online education to allow people to host their own courses. They couldn’t find a foot hold there so they moved on to online corporate training and compliance. After that didn’t work the turned to hosting videos for sales pages. 6 months later they were running out of money and couldn’t recover.
Another company started with a focus on building simple e commerce pages for people who didn’t need all the bells and whistles of Shopify and wanted their own shop outside of amazon/eBay. They spent time finding and talking to people who had shops, but only one or two products listed. This led them formulate a product that was for people looking to sell one product online. They were able to clearly identify who they were building a solution for and how to find them. They’ve been in business for almost 2 years and they aren’t slowing down.
Do you know the name of the company selling e-commerce pages? i might be able to get them some clients.
Because it's making what they want, delivered to them in the way they want, at the price they want, as easily as they want, through a trusted vehicle/business that they want, at the time they want and on and on...
underrated reply
this is very true
Short answer: Because people don’t follow my Sales Safari approach. ;)
This is the core of what I teach in 30x500 and I’ve written and recorded lots of free content to explain it:
https://stackingthebricks.com/video-failure-vs-creating-a-product-people-want-to-buy/
Longer answer: Because of psychology, basically. Most people are more interested in doing what “feels cool,” or scratching an itch, or following their first instinct regardless of how well-informed it is and regardless of what the global marketplace shows works. Or they’re obsessed with being an inventor, creating some kind of novel business model, or chasing after the fame of lighting-strike success in fields like consumer social apps which aren’t often even profitable and not truly businesses in their own right.
Lastly, folks are not skilled (and/or don’t try to) actually analyzing what happened and why, so they come up with pet theories like “nobody could explain the success of xyz” even when there existed a series of previous platforms and products that presaged it, and human nature that explains it.
People behave in predictable ways. There’s whole academic disciplines devoted to studying us, and how, what, why, and when we buy things.
After you rule out lucky breaks in winner-takes-all marketplaces (B2C social apps, etc), the rules of economics — how, why, and when people PAY for things — aren’t a mystery.
People have habits, desires and needs and problems, and either do or do not pay for things to fulfill those desires, needs, and problems. Once you’ve narrowed down your audience to a group of people who are inclined to pay for things, all you’ve got to do is figure out what it is they need, want, and are ready to buy, and brainstorm a variety of ways to deliver it, and market it.
Ethnographic observation is the best way to figure this out without skewing your results. Which is why I based Sales Safari on real-life design ethnography.
There’s no need to reinvent the wheel on marketing. You can build trust and visibility with simple, boring steps that work.
This is a great question.
People are primarily emotional beings, no matter how much we like to portray ourselves as calm logical decision makers. Emotions are chaotic and unpredictable, every little stimulus can instantly change your mood drastically.
I could be seeing the best website for the most useful and valuable SAAS ever, but if I'm having a bad day chances are I won't pay too much attention to it. Also, I could be having a great day and even the mediocre product will look magical.
That makes estimating what people want extremelly hard. I would even say it's impossible today - most of the time people don't even know what they want, and we lack all the data and processing power to make an educated guess.
I believe the pendulum has swing too far away from vision and imagination, and too close to lean methodologies, measuring, market research, idea validation. All of those are great and useful concepts, but they often result with products for a smaller audience. That's why instincts and gut feeling still matter a lot.
agreed its hard to find the middle-ground
Tough question. I have nothing to add, but here's Naval Ravikant's words:
I interpret from this that while society knows what it wants, often that want is only clear once you build a product or service to address it.
People are unpredictable.
A reason why there's such a strong debate on why psychology is a science or not, is because it's hard to quantify. People are random, hard to study, and many factors affect each one.
True, but shouldn't it be pretty easy to spot consumer trends if you had a large amount of data? Or at least easier than coming up with an idea without an understanding of its market?
I think it's more nuanced than that: as Ford said, "if I asked people what they want, they would say 'a faster horse'".
Not only people don't know what they really want/need, you also need to latch onto something people are willing to pay for.
Consumers are notoriously difficult to monetize, as opposed to businesses (who most of the time act more rationally)
How far in advance?
As far as I am aware, no-one thought that the first telephones would one day turn into little computers you could carry in your pocket.
At the time the telephone was invented - or the electric telegraph or almost anything - the marvel lay in the very fact of its existence.
So today: new technology arrives and its very arrival reveals further opportunities that even the original inventors may not have envisaged. Alfred Nobel believed his dynamite would be a force for good in the construction and quarrying industries. The poor deluded fool had no idea that its main use would be to blow other human beings into tiny, tiny little bits.
As for market, sometimes the market is created, not served. An example: when Frank Whittle was inventing the jet engine in the middle of the Second World War, he was not taken seriously by the then Air Ministry. Although the jet engine would prove itself far superior to the piston engine and would quickly have demonstrated its superiority in a theatre of war, the main market (the Government) lacked the imagination, foresight or long term commitment to back Whittle at a time when it could have done the most good.
History is littered with tens of thousands of such examples.
You don't even need to look at technology, Famously, J.K. Rowling, author of the Harry Potter series, was turned down by no fewer than 27 publishers who had lots of data on what had sold but no idea at all as to what would sell.
J.K. Rowling is now very rich. 27 publishing houses could have been.
Part of it is that you're often trying to build something that hasn't been discovered yet. And you're trying to reach people who are yet to find the existing options in the space, and/or you're trying to make it work financially when people prefer free.
A trove of data will tell you what people do and what they like.
It won't tell you what they're frustrated by and what they're looking to accomplish.
Those two things -- what frustrates them, and what motivates them -- are what a successful product understands and builds for. It isn't about what they "want" -- it's about taking away pain.
The first step is to get out and talk to people. And when you talk to people about what they "want", ask them not "what do you want?" but "What are you trying to do? And what's getting in your way?" That will give you much more valuable data than any spreadsheet could ever possibly convey.
If it was straight forward and data driven then every business would be profitable.
The problem is that the data you refer to is historic. It references things that actually happened with products and services that actually exist.
What we need to do is find the things that people don’t know they want yet. That’s our holy grail.
Like a lot of things about starting companies, I think discipline plays a big role. It's not that hard to figure out, but people get wrapped up in their pet ideas or technologies. It happens to me too. My version is that I tend to make things that scratch my own itch. That's at least some validation, but often the market as a whole doesn't have the same needs you do.
The other thing is that market trends are pretty abstract, and there's a big gap between a general problem, the specific version of that problem, and a unique solution to that problem. Ok, let's say that people want to save more money. You still have to figure out why they don't do it now, and how to help them do it. Broad market research info probably won't take you very far down that road.
Good point Charles. May you be more specific and list a data sources for people preferences analysis ?
Data brokers like acxiom are a good example. See my comment to @tadas below
Could you link to said troves of data? I'm not sure I'm familiar with specifically what you're referring to, and the context could help guide discussion.
https://www.acxiom.com/
data brokers are a good example because they track using tons of different sites on the internet. A lot of our information is cross associated.
I'm under the impression that data broker data is expensive (am I wrong? acxiom offers only a "contact us" link when trying to get hands on their research). In your original post you said public and easy to obtain. Are you referring to the way data brokers harvest data? I'd argue that planting pixels and doing heavy statistics work like identity resolution, etc, is far from trivial/easy.
I've often thought that huge companies like Google, Facebook, etc, can consistently churn out fairly good and useful products in large part because they can piggy-back off of what they learned from their massive userbases. And they're the only ones privy to that usage and feedback data.
It is probably very expensive, so I guess that is a bad example. That being said, there are definitely ways to get "big data" at a low cost. I don't know those well but if I took a few data mining classes at school I bet I would.