Hacker News Reader

Comment by pg | Hacker News Reader

These are all things we care about, but they are probably not the most common questions we ask. E.g. we already know about the equity split because we ask about it on the application form, so we only bring it up during the interview if we noticed something odd about it.

The thing we care most about in interviews (at least of things one can change) is how engaged the founders are with users. How do they know people actually want what they're building? Have they talked to real, live users? What have they learned from them?

We don't care super much how big the initial market is, so long as the startup is making something that (a) some subset of people want a lot, and (b) if that market is not itself huge, there is an easy path into bigger neighboring ones. Basically, we're looking for startups building Altair Basic.

View on HN

A good corollary question to "How do you know that people actually want this?" is "How are people solving this problem now?"

If founders respond that there aren't really any current solutions, then it usually means that either a) They aren't making something that people really want, or b) They haven't talked to enough users.

If it's a problem people actually have, then they must be coming up with crazy hacks or solutions that are much more tedious/inaccurate/expensive/generally more painful than the one you're coming up with. Very rarely is there simply not some kind of existing solution.

This is such an important test that we have a question about it on the application form. I learned about this heuristic from Sequoia, who care a lot about it. They call it "proxy for demand."

Christensen talks about this too, as "non-consumers" who have a job they need to get done, but can't because they lack skills, or money. Instead, they pay a professional to do it for them, or "cobble together a solution". There are also "non-consuming contexts" where you just can't use a product (e.g. a landline phone in a car).

What I found really interesting was that the reason successful disruptions are "more convenient, simpler and/or cheaper" is not because that's an improvement, but because it enables the disruption to be used by non-consumers... (who lack the skills for a complex product; or the money for an expensive one; or access to an inconvenient one.) They are delighted to have a solution better than what they have now, so it doesn't need to be as good as the incumbents'. Secondly, if it's not good enough to appeal to incumbents' customers, it won't provoke a competitive response.

"Very rarely is there simply not some kind of existing solution."

Yep, when you talk about known problems. Still often both the problem and solution (which is just another view of problem) lie just outside current scope of people imagination, and are only found by prospective minds when the way is open (other conditions met).

There was not simply some kind of existing solution for Internet 50 years ago.

If you solve an unknown problem (ie. an issue nobody is concerned about) you will have a very hard time to market it and show people how they need it.

There were solutions for this Internet thing 50 years ago, even if it's hardly imaginable today. If the problem you are speaking about is communication we had that 2000 years ago, in form of smoke signals. It sucked and was dog slow but it was not an unknown problem.

If I had asked people what they wanted, they would have said faster horses." - Henry Ford

another great quote along these line (who knows if rumor or not): "we never would've invented the ipod with a/b testing"

think there's a good balance between customer development/research and your own instinct/vision

Aren't some of the most world-changing businesses actually businesses where a simple solution already existed, yet entire industries have been created or destroyed in spite of these simple solutions??

For example, before the automobile, horse and carriage was a 'solution' to the problem of getting around. In the early days, was a car a much better solution than a horse? Horses were readily available, didn't need a gasoline infrastructure, they even self-replicated so you could get a new one every few years.

I understand I'm playing a bit of devils advocate here, but I often gut stuck in the mode of thinking existing solutions are good enough, then somebody comes along and evolutionizes an industry with what at the early stages might seem to be minor improvements.

Another example would be the early days of sunglasses. If the question was asked 'how are people solving this problem now', the overly simple answer would have been 'they squint a bit'. With that as the answer, would you go and develop sunglasses?

The transition is never that stark.

Take your example of sunglasses. A quick search on wikipedia shows that the precursor were glasses with a thin slit in them. It was never squinting, and then BOOM, Oakleys.

There was always a gradual change that is only missed in retrospect because of how standardized and popular the winner is.

Name a few popular products and services, and I bet there have been similar solutions beforehand.

Thanks PG. Could not agree more on getting users as early as possible, and getting as much feedback as you possibly can from them.

In our example, we produce and sell incomplete ebooks, knowing full well we may get a few refund requests, precisely because it's invaluable for us to know how to convert a paying customer, and to ask those customers for feedback.

Reply on news.ycombinator.com