I sat down for a conversation with Alex Long. He took notes and sent them to me, and it seemed worth lightly-editing the notes and posting. I’ve left it quite raw, more like a tweet thread than a proper blog post.

what was your entrepreneurial journey / how’d you come up with wave?

After graduation (2008) I started business creating a programming language. It was a multicore programming language with automatic scaling. Didn’t get into YC, stopped after a few months.

I reconnected with Drew in 2011, worked part time together. Applied & got into Y Combinator in 2012, dropped out of grad school & did startups full time.

start of 2014 Drew and I started sendwave, which became successful.

Pre 2014 we iterated through a bunch of different social apps, primarily under the name “Chime”: event discovery, chat, dating, and video chat.

Tell me about “Chime”?

chime was the app we got into YC with in 2011-2012. It was a social networking idea, aimed at college students based on real world. Basic pitch was “find more things to do with your friends in the real world. post an idea, if enough people are interested it happens (like kickstarter).”

we did a good job executing on the basic idea. I built the website and drew was really good at distribution. We still had friends on campus, and we would go in and tell people about what we were making - people definitely clicked and wanted to use the product when they heard the idea. needed a core critical mass group. Recruited a group of students (Drew mostly did this).

We set up a daily meeting to get everyone in the core group posting. within 4 weeks everyone on campus was using it. This sustained only while we told our core group to post every day. Did this push for sept to nov, everyone left for thanksgiving and people stopped posting, when we came back we didnt restart the daily posting, and then immediately traffic fell off and no one was posting organically. So the idea failed, but it got us into YC bc it showed that we could build, talk to users, and execute on distribution.

During YC we made a bunch of different ideas based on the original Chime. For example, what if it were just a really good chat platform with location features. we switched from desktop to mobile sites to mobile apps. Next version - ui was a map, could put pins to say where you were going to be. none of these things worked.

def had the expectation that things wouldnt work. default was that thing wouldnt work. experimenter mindset.

i felt like i was learning. i learning how to build apps fast. just building anywith with a chat app was a pain. ppl have very high expectations for e.g Messages app working perfectly but it is actually quite hard to make the scrolling work as you expect when the keyboard comes up or whatever.

Tell me about starting Sendwave

the advice we got from YC was build something that can make money from launch. any revenue was exciting. Drew had previously worked in Africa, so we had talked about remittances a lot over the years, but they always seemed hard - like, it was obvious that a better remittance product was possible, but we had no idea why nobody was doing it: was it too costly to operate? Did the regulatory framework make such ideas too hard to execute?

At the end of 2013, getting fed up doing “easier”-looking startups, we said, maybe we should try to do the hard thing that could really make a difference in people’s lives. maybe itll be hard but maybe itll be worth it.

To investigate the idea further we came to DC to meet with kenyan immigrants who send money home. they gave really positive feedback on napkin drawings of the app. we decided to commit a few months to building the prototype to see if people would want it. we knew that the regulations would be some challenge. so drew started researching that while i started coding on a product.

Drew had prev started a NGO in tanzania and knew how hard it was to send money there. had also experienced mobile money in tanzania.

We wouldn’t have believed enough in Sendwave if we had just come up with the idea by reading about mobile money. drew’s experience in africa was pretty essential to us believing it would be worth the large amount of work.

Drew started researching compliance and a lot of people told us that getting licensed wasn’t something startups can do.

any high impact startup ideas you think would be cool to implement but dont have time or inclination?

small and informal businesses in africa need software, like for smart phones that just helps them do basic business things like tracking their expenses. like understand how much i am spending my raw materials and how much money im making. ive been kicking around storebuilder idea. most dont have access to computers.

On conviction

most startups have a phase where you need a lot of conviction instead of giving up. but also dont want to waste too much time.

if you just really want to work on something, like aviation or whatever, even if you dont know a lot: Decide “ok im going to spend some time working in the field or volunteering, and spend time talking to people, then i’ll figure out what they need need and then i can build something.” i think you could do this process deliberately, although it is definitely more natural if you already have a bunch of experience in the field.

What resources for entrepreneurship do you recommend?

ycombinator was great for us but (now from the outside many years later) it seems to have changed a lot and I couldn’t give a strong statement about how good the YC experience is anymore. YC’s investor network is really good, mentorship was good in 2012 and has to be worse now but probably still fine. worth doing if you can get in. no reason not to do it basically.

books/blogs: i learned a lot from paul graham essays, sam altman “start up playbook”, peter thiel “zero to one”, elad gil “high growth handbook” (heres how to build a company). Tons of other stuff too, lean startup, etc.

So I’ve been trying to build lots of 1-week prototypes and put them out there to see which ones connect with people. Is this a good strategy?

It can work I guess, but you shouldn’t expect that things that can be done in a week will produce that much value.

The fastest from-scratch build we did was a week to make a first really shitty build, but even then trying to get users was a minimum of a week of work as well. More realistically we could make a still pretty bad prototype in 3-4 weeks, then try hard to go get users, and plan to do several iterations - and the later ones go slower because it is always slower to improve stuff than to build fresh, but this is where the real value comes from. In general we never expected less than couple of months for each idea. 2-6 months to really try something out properly.

Sendwave’s first build was 2-3 months of coding (5 months committing to working on idea -> launch). While I was doing that, Drew was doing design and UX testing (getting feedback on napkin and Photoshop prototypes), figuring out biz dev, licensing, compliance and fundraising. Drew and I both went to Kenya prelaunch to figure out the delivery, legal and bank partnership aspects.

A high % of the work should be trying to get users. Let’s say you have built your 1-week prototype and you think “gosh, this is good, I want to take it to the next level”: Try seriously to get users and commit to making it work better for them. Your growth strategy can’t just be: “hey this thing is out there” - for a really rough prototype, you need to try hard to discover people with the greatest need.

One trick that seems appealing to me is finding the subreddit for the community of people most likely to be interested. You can’t expect to make reddit your ongoing growth strategy though - reddit post is something you can (kinda) only do once. The reason to do it is hopefully find people through that process who can be your power users who can you help you find ways to grow fast: “hey i built this thing, i’d love feedback. ill even give my product to you for free but i want to to talk to you about your experience with it”.

You need to find the people who are gonna love it and work with them closely on it and figure out how to make it good or better.

More broadly, in startups, you’re always running the meta process of “am I doing it right?” It’s very unclear if it will lead to anything, “am i committed to this and making it my job if i can, and choreograph my life around making it succeed.”

The thing that ive learned most distinctly: nothing takes off when you just put it out there. sendwave was extremely well-poised to take off because there were a bunch of ppl who needed it desperately - and still it wouldn’t have taken off if we hadn’t gone into kenyan restaurants and pitched it to people night after night before launch.

As Paul Graham has said, startups are like an old car that needs a lot of pushing just to get it to start1. thats why a week is just not enough time to put in. At YC they would look at you funny if you came with an idea but wanted to give up early. thats part of the structure of yc. As a solo entrepreneur you don’t have that structure.

YC’s default expectation is that the thing youre doing is worthwhile, at least on the order of commitment of a couple of months. To me, even a couple of months seems low. it doesnt seem crazy that if you were really well focused and decided after that to change ideas. but that was still fast for us, pre Sendwave, Drew and my cadence was ~6 month ideas.

Thanks to Alex for the questions and to Drew for edits!

  1. Slight misquote, but this references Do Things that Don’t Scale