Startup School 8: Jan Koum of WhatsApp | Hacker News Reader

Startup School 8: Jan Koum of WhatsApp

startupschool.org

271 pointssama55 comments

Jan is the entrepreneur that I admire the most. Here are some of the reasons:

1. He was so focused on building the product that he just ignored all emails from carriers and investors. Sequoia partners had to hunt down Whatsapp's office just to get in touch with him. Most entrepreneurs would be easily distracted by investor attention.

2. He didn't care about the press (despite a ton of inbound interest). I don't know how or why he could resist the temptation, but he did for a long time. In fact at one point he tried to consciously stay out of press because he wanted to "manage" growth and didn't want competitors to get wind.

3. Very balanced outlook towards global users - even though SV (or even US) didn't recognize their traction, he always understood that there is a big world out there where not everyone has an iPhone.

4. Focus on infrastructure and reliability.

5. Focus on being "lean". For a long time the team was quite small - but the thing I find most telling is the fact that he didn't have "business people" for a long time - anyone who didn't build the product. At acquisiton I think they had a GC and a business operations guy. It it amazing to be able to resist the temptation to build an empire.

Whichever way you look at it, he comes across as someone who is able to resist common temptations and had a great balanced head on his shoulder.

There is Lot to learn from Jan

He's pretty lucky that he was able to just focus on product and not really worry about the actual business much. Many builders just build in obscurity and the world never knows what they made. I'm jealous.

Not to take any credit away from Jan, it is however worth noting that his background (and that of many of the early employees) is in Infrastructure / Backend at Yahoo! The side-effect of that is that reliability/uptime is already in his DNA, as also, just "getting things done without enjoying any of the limelight", and he's done a great job carrying over lessons/behaviors learnt in that previous life over to WhatsApp.

> 3. Very balanced outlook towards global users - even though SV (or even US) didn't recognize their traction, he always understood that there is a big world out there where not everyone has an iPhone.

This impressed me the most, because at the time whatsapp went mainstream globally - it was available on all platforms including the old BlackberryOS, BlackBerry 10, Symbian (S60), Windows Phone and even some candy phones!! which was unlike any other messaging platform and I have a good reason to believe it was instrumental in their success given how many of my friends switched to whatsapp just because everyone in the group could use it (we had WP, Android and iOS users and one oddball BB10 user [me]).

And that is why they succeeded imo. Because Facebook bought Whatsapp for it's 1 Billion users, and Whatsapp would never have hit that mark had it not been for the fact that they focused on market outside the US (where cheaper phones with android, candybar and other OSes were more popular).

While Whatsapp is still not as big in the US, it's literally the defacto mode of communication in countries like India, Brazil etc. to the point where "Msg me" would automatically mean "Msg me on Whatsapp"

That + the fact that it was tied to your phone number was in my opinion genius. I for a thousand years would not have though too much about Global Market (because I'd be focused on revenue and users in 1st world countries would be generally the ones with spending cash for messaging apps).

> That + the fact that it was tied to your phone number was in my opinion genius.

This is an important aspect. Having accounts and contacts based on phone numbers allowed my mother and her friends to use the service in no time - those are the kind of people that don't know their own email address (if they have one at all) and would never register for an account on their own. Phone numbers, however, is the system they already knew since using SMS.

Did not early WhatsApp use a covert channel in sms signalling, just like you can get 2.4kb/s across the x.25 "D" channel of ISDN?

I've no idea if true or not, my memory unreliably tells me I found some kind of confirmation, but common sense tells me not to rely on such recollections.. but I often pondered if such channels exist in the GSM spec, back in the day, when I was shocked to receive a sms, the service having been given free to early subscribers in the UK, I literally had the first (edit)model Nokia GSM handset, and was most sceptical :D

I think you are mixing that up: SMS uses a signaling channel of GSM. It was originally not intended to become that big and just used otherwise unused resources. Therefore it could have been given out for free or very cheap. Later on telecom providers saw that people like text messages and that it has an enormous business potential, so they started to make huge charges for it.

The "otherwise unused" isn't true. Any time your phone is transmitting, it is using some of available bandwidth in the air. It does go other the control channel is GSM rather than setting up a data channel, and is was originally just for testing, carrier announcements etc. The per-SMS charge is far higher than actual costs, but so is everything. Mobile networks are almost entirely fixed costs in building the network.

I hate to sound cynical, but these are only admirable if these features are the most valuable to users/customers. I see way too many entrepreneurs enamored by technology to find a year later there's no one using their software.

I know this might feel sad for software engineers to hear, but there's a lot more Benioff's than there are Jan's in the world.

I agree about Jan. I do admire what he and his small team did with Whatsapp and the sheer mind boggling scale they achieved.

But this is also success bias. If he didn't succeed the same points would be used against him. Winners make their own rules and hindsight is a bad teacher.

These same points he avoided for instance partnerships and press may have worked for other successful founders.

WhatsApp launched at a time when SMS was a paid feature in Europe. It's pretty rare to have such a massive virality in your product. Not only is there a large network effect, your competitor is also charging a large amount compared to your free and superior product.

SMS is still paid in many countries of Europe and even more across world, so unlimited sms are still exception, not rule

Regarding #5, Whatsapp ought to have one of the most impressive employee:user rates ever.

In other words, he passed the marshmallow test.

WhatsApp and Netflix are the only companies who actually have interesting performance engineering solutions instead of throwing incredible sums of money at Yak shaving shenanigans that actually amount to digging themselves deeper.

This was interesting, and well presented. So my beef isn't with Jan but more so with the startupschool initiative. (focus on school)

How do you learn anything tangible from this? It's just a bunch of successful people talking about what they did. Without more context this is just survivorship bias, most of these aren't even serial entrepreneurs.

Dunno - I think there's a lot you could learn. They've had about 7 people on talking about the startups they did, a few were serial, some of them had failures. I'm not sure what better way you'd learn lecture wise?

I find that seeing about how someone else thinks about / breaks down a complex problem is helpful. Sometimes you can also see where the speaker has a blind spot - an interesting thought experiment is to consider how you'd compete with their company.

It's the reason business schools teach through case studies and is the reason Warren Buffett mainly reads biographies. Learning what other people did or did not do and why can help you in your decision making processes.

The talk raised interesting points about hiring. Jan said that most of the early hires were people from their own networks that currently were "unemployed" or goofing around and that's why it was easy to hire them. So it sounds that they hired people that they knew and trusted but who, for an outsider, would look like were not great at execution.

And yet, Whatsapp is clearly a startup that excelled especially at the execution. They successfully build multiplatform product that included OSes that were traditionally seen as cumbersome to build for (Symbian, Java ME). Many, many companies have struggled and failed to do that. It also seems that they scaled really well (compared to e.g. Twitter)

(without reading the article/listening to the talk) I suspect "unemployed" meant something like

"worked at yahoo for double digit years, cashed out shares, and so don't have to take dumb jobs for eating money, so free to tackle new challenges"

not

"so bad at programming they couldn't find a job in a hot market"

Yes, I understood that, I tried to indicate that by writing "unemployed" - i.e. they were talented but taking it easy by their own choice (one guy apparently for 10 years). However, for a person who hadn't worked with these guys, it wouldn't be obvious that they would efficient in the execution, as many times people err to use a filled/gapless CV as a proxy for execution capability.

Super congratulations for successfully reading the "messaging" tea leaves and capitalizing on the trend. I admire the vision. I would have never in my wildest imagination predicted the extent to which text messaging became a killer app on mobile--to the point where we have dozens (hundreds?) of different apps to choose from. I still don't see the appeal of messaging apps, but their popularity is undeniable. People don't even seem to mind the fact that all of these apps do fundamentally the same thing yet are largely incompatible with each other.

We had XMPP, but Google and Facebook decided to kill it.

i just don't understand why we don't replace it with universal email, everyone has it and it is certainly possible to build chat on top of it

I enjoyed this talk and took away something slightly different to others on the talk...

For me, what was clear is that they listened to feedback and took action relatively quickly. I remember when they first launched the Blackberry version and emailing the team about bugs - I think I got a response the following day and they clearly implemented something by the following release.

It wasn't about being devoted to a vision of a particular product they were passionate about. They saw a market/ problem and solved it using their technical ability.

Yes, they had a lot of luck on their side - being in the market early enough to gain the visibility. Also, people naturally promoted the app so they didn't need to spend much on marketing. Other apps that came after it, weren't so lucky. Being well connected and having cash to provide initial runway probably made this easier for him than it would be for most but I think it's fair to say he built up his rep at Yahoo - which probably helped when he needed to raise cash.

How did WhatsApp survive between idea -> first charging users?

Why did they avoid any media attention?

Can you please post a reddit-AMA style proof to assure users that it really is you. It is unbelievable that you weren't aware of HN and that the profile was created just a few hours ago. Thanks. :)

i have no idea what is reddit-AMA style proof is... sorry i didn't get around to making HN account before today.

AMA = Ask Me Anything

And people doing these AMAs on reddit usually post a picture of themselves with a sign, see https://www.google.co.uk/search?q=reddit+ama+proof&source=ln...

A simple tweet from your verified twitter account @jankoum to indicate that your "HN username = jankoum" should suffice.

We detached this subthread from https://news.ycombinator.com/item?id=14251471 and marked it off-topic.

I honestly don't even want to watch this, out of spite.

Whatsapp is so damn frustrating to me and the people I talk to.

The microphone button is so finicky that I lose hours of my week to trashed messages. I don't know if it's the size of the button being so tiny that a man's thumb easily hangs over the edges, triggering the trash feature with the slightest movement, or some other strange bug. A call comes in, message trashed. An alarm, message trashed. I've literally recorded a 10 minute message and lost it, condensed it to 5 minutes and lost it, condensed it to an angry 2 minutes and lost it and then just given up...

Then there's the volume. Why do I have to crank Whatsapp up so high in my car that EVERY OTHER audio event blows my ears out?

How are these huge usability problems not being fixed with the amount of money that was thrown at it?

I can't find anyone or anywhere to complain about it either. The Contact Us area of the application REQUIRES you to give them your full contact list before you proceed.

This!!! I would upvote 1000 times if I could. I appreciate Whatsapp's features but voice recording is just broken. I experience this same frustration... I tried submitting feedback through the app and it failed and lost what I wrote due to the default mail client not being configured. Oh the irony.

Please fix. These are minor changes.

It is very nice to have first mover advantage. Many web and mobile application companies today do not have first mover advantage. This was a good talk to listen to if you have quietly stumbled on something that people want before the rest of the product builders find out.

There were a lot messaging apps before whatsapp so they weren't first. I think the smartest thing they did was use the phone number as the user ID. It really reduced the friction for a lot of people.

But they didn't have first mover advantage. There was ping and a few other similar apps on the iOS app store.

WhatsApp focused (correctly) on distribution over several platforms.

BBM had huge potential as well in the early days. Blackberry practically invented this "SMS-alternative" type of chat application. It was being quickly adopted in emerging markets as a replacement for SMS, which was much more expensive. However, Blackberry screwed that up by keeping BBM Blackberry-only for so long. The BBM should have been available on the iOS and Android stores (and probably on Symbian/Java) since 2009 at the latest (basically when Whatsapp was launched), and it would've dominated.

Possibly. On hindsight, RIM totally underestimated the onslaught on mobile started by iOS and later also by Android. They never thought those platforms could take off, so their decision is understandable (but totally wrong of course).

yes of course - because having a random 8 digit hex pin was so user friendly, easy and intuitive :)

What did you do in the very early days to promote the app (apart from Flyertalk)?

heh Flyertalk... awesome forum btw.

until we added messaging nothing I did worked. once we added messaging it took off on its own.

You mentioned in one of your talks about a small loophole in the early days of the App Store where apps were showcased under "new and noteworthy" if one made an update to the app - Q1: is this true?

And Q2: when you say "took off on its own" does it mean folks learned about WhatsApp only from word of mouth with no effort from the WhatsApp team?

Q3: what impact did charging for the app have on the company finances and growth rate?

1. yes, i believe so - of course it was eight years ago.. i think if you submitted new version of the app and changed the name, it would show up under "new" category

2. yes, word of mouth only... but we helped people spread the word by putting "Tell A Friend" functionality into the app :)

3. it helped us pay for office space, salaries and server bills without eating much into VC or angel money we had in our bank account.

Thank you for the answers.

Re #2: I remember hearing about WhatsApp from my friends in 2009 or 2010. They said, "it's basically BBM that lets you communicate with non-Blackberry phones for free." Since those of us who had Blackberries used BBM non-stop, it was a no-brainer to get WhatsApp.

for the cynics who will invariably pile on with "survivorship bias," would you mind sharing what choices you consciously made that were different from other messaging apps?

there were a few and i probably won't be able to list all of them, but here are some:

- build for multiple platforms. most of our competitors did iOS and Android only and called it a day.

- keep the app simple. most of our competitors complicated not only the sign up flow but also the in-app experience

- look native. our thinking was that hundreds of millions of people use native apps thousands times a day and if we use native look and feel in our app, it will feel comfortable and intuitive for our users.

- focus on speed, performance and reliability. it is easy to build an app that sends images quickly on 4G but you also have to work efficiently in the EDGE environments.

- localization. we translated into as many languages as we could as quickly as we could.

- support. for the longest time Brian and myself were the only two guys answering customer support emails.

- focus on organic growth as it makes your network stronger

there are probably a bunch of other things we focused early on that escape me at the moment...

thanks so much! keeping whatsapp simple and focused on replacing SMS/MMS seems like the key differentiator. was there ever internal conflict about making whatsapp more complex or "instagram-y"?

do you mind also sharing what the burn rate per MAU/DAU was around the acquisition?

thanks again!

You covered product decisions pretty well, but I was curious about organizational decisions. Based on your talk, it seems like you existed mostly outside the SV bubble and seemed to maybe even avoid it. Nonetheless, many of your employees came from the network you developed in SV and it seems you were based in SV at the time.

Do you feel that it helped to have started there? Do you feel at this point that you can build a successful company "where you are," given modern tools (AWS, etc.), or does it still help to do your time in SV first?

I remember a few years ago when people bought 10MB. And WhatsApp for a whole week.

In fact, till today, my dad doesn't subscribe for data. He uses WhatsApp on a PAYG basis which would be financial suicide with other apps or browsing the web.

For instance, Skype can eat 1mb+ just by launching it on Windows.

WhatsApp is fast and easy on resources. I know someone who still uses WhatsApp on blackberry bold!

Skype on the other hand takes forever to open and a few more to connect on a 1GB ram phone.

The only time skype beat WhatsApp on speed of launch was when it was baked into the bareboned SMS app on WP.

One area where WhatsApp stands heads above the rest for me is notification. It's timely, and doesnt get lost. Notification counts are accurate unlike skype which always displays (1) pending even if you have 10 new messages - even On desktop.

WhatsApp understands the market better than the rest who mistakenly believe that the rest of the world has similar high, reliable, cheap network speed/ fast devices.

It makes me wonder why they all collect usage, device info, network data... at all.

BBM blew it though. They had huge traction in 2009 and wouldn't let other platforms access it because they were selling a lot of phones and BBM was one way of keeping users tied to their platform.

That's why in 2013, BBM was released on ios and later android. but by then it was too late, they were losing a lot of marketshare.

Have you tried using Telegram? What do you think about it? I don't care about all the security battle that is going on, but I feel that Telegram has more fluent UX.

Reply on news.ycombinator.com