Hacker News Reader

Comment by 0xfffafaCrash | Hacker News Reader

The idea that a company on the scale of Meetup could be effectively run by 5 people in a basement strikes me as completely absurd.

I've seen similar comments about companies like Twitter, YouTube, Spotify, and even Facebook and I'm really in awe of how completely out of touch with the reality of what it takes to run a global business with millions of users and customers people have to be to suggest these things.

Do people imagine you can just take the UI, have a few people wire it up to scale in "the cloud" and basically call it a day?

View on HN

When Whatsapp was acquired for 20 billion and was already one of the largest messaging services on the planet it had 50 employees total, 30 of them engineers.

Now granted that's more than five people, but that was a planet sized service. Say you're a company that's worth 10 million or 50 million with a proportionate amount of complexity, you ought to be able to built that with a team of 10 people if you chose your software stack wisely.

Paul Graham has an older piece about Lisp and beating large companies by having a technological edge. Picking your software and understanding architecture can be very important. I have seen Clojure teams of smart engineers in particular in distributed / concurrent systems (the sort of thing that can produce a lot of value) who do as much work in a team of 5 as a big corporate Java behemoth does with a 100 people.

Same with Instagram, not sure how many people, but they could all almost fit in a garage together.

I've also seen small teams of smart engineers using decidedly "boring" stack like Django/Rails + MPA on EBS do as much as much work as a larger teams going hard on more cutting edge stuff (React + Redux + Scala on k8s).

I wonder what the deciding factor is. I used to be part of a Haskell team and it was actually net negative vs. just using something really batteries included.

> I used to be part of a Haskell team and it was actually net negative vs. just using something really batteries included.

Could you elaborate on that? What were problematic?

The deciding factor is how complex the code is. You can add a ludicrous amount of complexity into a React app before you've even implemented your first actual feature. People do it all the time. Those people will always get smashed in the marketplace if they have no other advantages and a competitor that's developing PG style with a lean team of actual experts.

I feel like the deciding factor might to do with team size, skill, and cohesion as team. Having a smallish team makes it possible (although not easy) to have a team that are all high performers. And that team can easily outperform one 2x-5x as large if not more, because there is a large communication overhead to scaling to more people. If you have competent people who "just get" each other, then communication flows and doesn't actually take up much time.

The technology choices follow on from this. If you have a smart team, then you can pick technologies that solve the problem most efficiently, rather than the lowest common denominator that everyone will be able to use, or what's most fashionable (you see this in every field: the "good" people are often following trends or best practices (and doing it well), but the very best tend to have a much more nuanced judgement. They know about the latest thing, they'll use it sometimes, but they keep their own council, and they're not afraid to go off and do something unorthodox if it makes sense for the situation they're in.

---

Side point: Incidentally, React is actually on my list of go-to productive boring technologies. It's definitely overused (despite building React apps in my day job, my personal website is built with Hugo, and a lot of sites would be better off with server-rendered Django/Rails/Laravel), but where you need that complex client-side interactivity I think it's the generally best tool for the job.

> Having a smallish team makes it possible (although not easy) to have a team that are all high performers. And that team can easily outperform one 2x-5x as large if not more, because there is a large communication overhead to scaling to more people.

^ This. As the project gets larger you can't find enough "star programmers" in a timely manner and have to fall back to mediocracy. Read "The Mythical Man-Month" and the associated Brooks' Law for context.

I similarly agree 5 is an absurdly low figure. Forget everything else and just consider securing the site, collecting payments globally, tax, and you've got 15 FTEs right there.

BUT - i'm at a VC funded startup and can see how bloated these companies become whilst achieving not much. Our company got VC cash, then the VCs installed a CFO. First thing the CFO did was demote most of the engineering staff down to entry-ish level titles.

Then came product managers, program managers, directors of product management, senior directors of program management, VPs of product, and C-suite.

The Product:SWE ratio is 1:2 to 1:5 depending on the team. That is right, since we're falling behind on coding and quality, the obvious solution to hire senior people to go talk at conferences.

Meanwhile engineering headcount and QA headcount barely moved. When product timelines and quality suffered, instead of considering why SWEs were demotivated, they hired more Directors and more Product Managers to micromanage and oversee the efforts. Except the Product Managers are remote and we barely ever hear from them. We think they are product managing their own careers mostly, we see more activity on their Twitter and LinkedIn accounts than we see on Slack.

So what have people done? SWEs jet out of the office at 5, many are running startups on the side, some are consulting on the side.

Next thing you have a several-hundred-person-startup producing about what you produced with 50 people

Fuck me.

This sort of "corporate bloat" was incredibly common in the "dot-com" era. I'm not surprised to hear it's still happening. You gotta hire a ton of people to justify that VC money.

... especially with the VC's requirement to "grow".

The small company focussed on their core service. The VC corp has to reach out to new markets, increase sign ups, please advertisers and produce shiny growth figured for investors.

For users of the Meetup site it should be slim and I want to spend as little time there as possible (i.e. get invitation mail, ideally respond by mail, done) Meetup however wants me to "engage" trying to convince me to go to more meetups, rate them, comment, ... do as many clicks as possible while not driving users away ... all these deceptions take time to do, to try out, to A/B test. A committed founder who uses the site themselves notices where users got trouble and works on that. Maybe not in a team of five but also not hundreds.

Err, Stripe?

I've worked in a startup which processed 10,000s of different transactions from various countries around the world, we dedicated a whole 0.01 person to handling it. I think they were up to around 7 o 8 countries when I left. Had one full time dev and me a few days a week.

As in it was checked now and then, and we had the occasional query once a month.

YAGNI for a business like meetup. I don't think you realize just how much off the shelf, fire and almost forget stuff there is out there now.

I wanted to jump on this bandwagon with my own experience. I worked for a software company about 4 years ago that had 6 people (including myself). Our company did $40M - $50M a year in revenue. No VC funding. We took thousands of transactions a month on Stripe and we basically never thought about payments. The software auto-handled collections or cut people off if they weren't paying (which in turn got them to pay-up). It automatically re-charged delinquent accounts, it automatically let people know that their card was about to expire, it automatically updated credit cards issued from banks (yes, really), and much more. We really never thought about payments at all. 90% of our transactions were USA, about 8% were Canada. And the other 2% was from Singapore. Never ran into any problem with these.

6 person crew generating $40M a year with no VC funding. It was all operated off the money that came in the door. It can be done.

>> 90% of our transactions were USA, about 8% were Canada. And the other 2% was from Singapore. Never ran into any problem with these.

And I once ran a python script on my local computer and had no scaling issues. I dont know what all these people are talking about nginx, load-blanancers, blah! heck, I just used python -m simpleserver and Viola! it worked! See -- everyone can run a global business right from their macbook with 0 FTEs! I did it! (Please dont mind I picked the easiest script and conveniently based my business in places where Stripe works!)

So, people who ARE doing it are replying on this thread.

People who don't need 15 FTEs, or even one FTE to take pretty large volumes of payments from many countries in the world.

Another reply is talking about $40 million turnover without a FTE.

The latest revenue figures I can find for meetup is $15 million.

So it begs the question, are you the incompetent and inexperienced one for needing to overengineer, or are we, for doing it cheaply?

Its really not about scale. I can process 1 or 1million JSON messages with the same effort. Now also support 25 other message formats? Doesnt scale the same way.

I can support a trillion of US/Canadian/European transactions with mostly the same effort.

I can also do payments myself in "7 or 8 countries". If you can do payments in 100 countries by yourself, please please let me know how!

I think a lot of technologists confuse deploying code with actually handling transactions, charge-backs, local tax rates, etc. Also, just because you are small enough to ignore local tax laws does not mean everyone is.

Move fast and break things is all fun and games until you get charged for tax avoidance.

Five years ago, it was difficult to comply with tax laws in every US state, county, and jurisdiction (now there are services) -- can you imagine doing this globally?

Yes, and if it doesn't need even close to one FTE to do this for 5-10 countries, why the hell would you suddenly need 15 to do it for 100? The other question becomes, that if you're not going to become the next Facebook, why even try?

I think you've never actually done this, have you?

>> Yes, and if it doesn't need even close to one FTE to do this for 5-10 countries, why the hell would you suddenly need 15 to do it for 100?

Because things dont scale linearly? I mean I can support JSON messages in 5min. Another five minutes and I can support XML. So I should be able to support all the historical mainframe message formats in 5 minutes each right? No...

But there is a service to handle that for you! https://www.avalara.com/us/en/index.html

Until you have to deal with customer support. That takes an insane amount of bandwidth.

Its a $10 product, how much support are you expecting?

People expect the same level of support whether they're paying $1, $10, or $100 per month.

You can throw bodies at the front end by having support staff, bodies at the middle end by writing extensive help articles/documentation, or bodies at the back end to ensure you have robust data and reporting for pre-empting issues.

In any case, you need to think about it, and it'll take bodies.

They are probably hired because of their contact network. As an engineer it might appear that new customers just magically show up. Depending on the sales model you probably can't just spend it all on ads.

> Forget everything else and just consider securing the site, collecting payments globally, tax, and you've got 15 FTEs right there.

I think you're over-estimating that by a factor of at least 15. If you're happy to say "let's just use Stripe, we'll support the countries they do, their payment forms are good enough" then that's like...1 FTE for a couple months to build your payments system, and then basically well under 1 FTE there-after to maintain it.

You can absolutely find work for 15 FTEs working on payments (and more!), and you'll probably end up with a better solution too, but there's plenty of startups accepting payments globally with, effectively, 0 FTEs working on payments.

Can all the US mom-and-pop websites collecting payments from US consumers please explain how they would proceed to effortlessly collect payments in Africa? The Middle East? Pakistan?

This is an honest question, because I'm starting to seriously question my beliefs (I honestly though it was difficult.)

If people in those places have Visa/Mastercard credit cards, I'd expect to be able to charge them in Stripe just like any other card. Sure, https://stripe.com/global doesn't list Pakistan, but that's the list of countries your company is allowed to be based in -- your customers can be anywhere in the world.

If people in those places don't have credit cards, I'd wait for Stripe to add whatever payment method they do use. (They've already added systems like Alipay and Klarna that are mainly used outside the US.) In the meantime, they can email me and we'll work out if they can mail me a check or use Western Union, or maybe I'll just give them an extended free trial or something. There are probably not so many people in the world who (a) don't have credit cards and (b) are interested in my software-as-a-service that I wouldn't be able to manually help them on a case-by-case basis.

(Disclaimer: I have worked on a mom-and-pop e-commerce website, but it shipped physical products and didn't operate internationally.)

If you're in the US, it just works; it's been a solved problem for, I dunno, decades? It just works with Stripe, and it just works with all their competitors. Check out their docs, it's interesting reading: https://stripe.com/docs/currencies

What's a bit newer is that it's a solved problem for people in other countries too; 10 years ago or so I really wanted to process payments, but none of the easy payment processors supported my country (mostly Braintree and Stripe back then), so I had to go with eWay (awful) and a merchant account from my bank (double awful). The company I'm with now (same country) uses Stripe, and it's a massively better experience. (Not being a Stripe fanboy; they have some very strong competitors, I just happen to be familiar with Stripe.)

Keep in mind: It matters where you (the merchant) are. Accepting payments from someone in Africa when you're in the US is trivial; accepting payment from someone in the US when you're in Africa is much harder. You can see where Stripe is supported here: https://stripe.com/global (and most of their competitors are in the same or fewer markets).

Also, there's still a lot of pretty bad legacy payment processors around, and they're often a fair bit cheaper. So if your company has been accepting payments online for twenty years, it may still be a pain for you, I wouldn't know. :)

But yes, the mom-and-pop website in Topeka should be using Shopify or have paid a web shop to throw together something equivalent using off-the-shelf packages, and it should be using Stripe (or Braintree, Square, Adyen, whatever), and then yes, they can accept payments globally, and the experience will be really nice. It's not even any extra work; it just comes out of the box for "free".

I don’t know anything about the payments software, but most mom and pop websites in America aren’t hurt by not selling to people in those places, or at least not hurt enough to be worth a FTE.

Stripe accepts a huge number of payment methods, and for the ones it doesn't (e.g. M-Pesa in Africa) there's a good chance you can pay via Google Pay, which it does support and which supports even more methods.

Just based on your off the cuff description I'd be seriously thinking about jumping ship. At least update your resume so you don't have to when the day finally comes (either by your initiative or theirs).

>> So what have people done? SWEs jet out of the office at 5, many are running startups on the side, some are consulting on the side.

I'm not the GP poster but that sounds like a good deal. So collect a paycheck, nod during the corporate circle-jerk of design meetings for weeks while working on your own startup and side consulting?

My ethics don't allow me to work at seemingly-doomed companies for very long after the realization that they are doomed.

I'm working for a company (hesitant to call it a startup) that has ~15 people on engineering, and it's pretty successful in it's space. It's been around for a decade and I think we have a lot more features than what I would assume Meetup has (we have to, because we're dealing with customers who have individual workflows).

I'm not saying 5 people can run Meetup, but I wouldn't expect it to require the 270 employees that Google lists.

5 people can run meetup, but that's not the business or your product with better features would be bigger than meetup.

Most money goes to sales, business, strategy. Growing your company into meetup would take hundreds of employees endless sales calls, conferences, golfing dates, money raising, competitor spying, timing/luck and most importantly closers. Then you need people to manage, organize them, spy of them, babysit them and clean up after them. Nevermind the interns.. if they are paided.

It didn't, it didn't have those numbers when it got big.

It, like so many other companies in the networking space, were in the right place at the right time. And seem to many of us to have grown in proportion to the money thrown at them, rather than the actual operational need.

There are examples of companies that needed to do that, Groupon, etc., but that's because they had to actually do sales.

Depends on whether you want to exist on ads revenue.

You could run Meetup with 5 people in a basement if you had simple subscription revenue, but effectively maximizing an advertising opportunity requires at least ~1000 people.

I can’t upvote your comment enough. Most people questioning why Twitter needs have 2k staff is because of advertising.

I’ve advertised on Instagram, Facebook, Twitter, Google - there is a tremendous amount of work that goes into the “money making” bit of these companies.

Remove ad-revenue and all these companies would collapse. They have no other model.

Meetup does have a subscription model that charges the group owner a relatively significant amount of money on a recurring basis.

$120 to $200 a year, depending on the country.

I'm surprised how expensive it's become. I was paying $60/year in 2018 (when I abandoned the group), that's now doubled.

https://help.meetup.com/hc/en-us/articles/360001620472-Organ...

That's quite a lot considering that nature of the platform. It explain why most of the stuff I see these days seem to be more small commercial classes rather than just the meetups that it started as.

If you're bringing in subscription revenue from millions of people you're gonna need some more customer support. And at some point you're gonna need some more HR then...

> If you're bringing in subscription revenue from millions of people you're gonna need some more customer support.

why can my mobile phone provider outsource its support to $random company in $random country but meetup cannot ?

Contrary to popular belief, having an overseas callcentre run by a third party to answer phones doesn't eliminate the need to have other staff capable of actually making decisions and fixing things.

First tradeoff: training and tooling.

If you go the "minimize the official headcount" approach... now you need to make your software systems and business processs operable by $random users potentially with different first languages, with little context or history with the company. That's gonna require some significant tooling investments compared to the typical 5-person-in-a-basement operation.

partly because oligopolies dont need to provide as good support

Those are massive companies with little competition and millions of global customers so they can have crap-tier support that works most of the time and get away with it.

You're not going to put up with that for Meetup.

The question isn't about effectively maximizing something. It's about whether they can turn a healthy profit. And in my experience you can, if you build the company and product correctly. But if you have VC money you won't because you can't be happy with tens of millions of profit with VC funding.

VC money can win a market and make everyone rich. You just need to be aligned with the goals. Making a little bit of money isn't the goal. It's all of nothing if you stumble there might be a pivot. You agree not to stop until it dies, you die or you hit 2 years and growth doesn't look like a hockey stick and they take it away.

Does Meetup run on ad revenue? I have an ad blocker installed so never noticed. I thought they charged organizers a modest fee.

Organizers pay. It's about $15/month for a calendar and email list.

Also moderation: Meetup isn't gMail, so I get unwanted, sometimes explicit messages delivered directly to the Meetup inbox (and forwarded to my email).

That doesn't even take into account the number of fake groups that aren't really meetups so much as "here's a webinar in which I pitch stuff to you 1-on-1 or 1-on-2, if I can find a second sucker".

Those are problems that can be left alone until they actually become issues, which would probably be the case once the hypothetical service has become successful; in that case, there should be enough cash flow and minimal overhead needed to implement/hire moderation. But if you're spending all your money on building rent, fancy office equipment, lots of bureaucracy, etc., then of course moderation is going to become difficult.

These problems can be mitigated early on by charging meetup organizers. If a $20 a month is too much, then it probably doesn't belong on a meetup site. It's enough that anyone willing to abuse the system will take the path of least resistance and use Eventbrite to host webinars.

Any other meetup communication should be made unappealing enough that moderation rarely becomes an issue. It should be possible for people on meetup to communicate, but it should be a second class citizen to however the organizers communicate with group members. Any forums that are part of meetups should suck just enough that nobody wants to use them unless they are serious about communicating something.

There are creative ways to handle moderation that don't require obscene amounts of money or employees.

> a company on the scale of Meetup

Meetup as a website and app is a pretty small engineering footprint. The fact that you think is absurd shows how much we have been accustomed to this "Grow at all cost" mentality.

If I had to take a guess and break down the 200+ employees at Meetup, more than two thirds must be in BizDev, community relations, and artificial management layers. Out of the 70+ Engineering, a core of 20+ are probably really needed to keep the core app up to date and up and running. The other engineers are working on bets and new products that the company needs in order to justify its valuation and future returns.

Essentially I would say you get 90% of the product with 15% of the engineers. Everything else is small bets to justify the valuation.

Don't forget the pointless reworks of the interface that happens every few years in most startups.

And then you look at places like gumtree or craigslist and for these network effect businesses you realize the UX ain't worth the expense.

I bet if twitter hadn't gone through their SPA/etc. iterations they'd probably still be as big today.

My thoughts is that to do exactly what meetup does they probably need the huge team they have, but they could deliver a perfectly adequate service for much less. We don't need all of the bells and whistles and new redesigns as long as the site works and has content.

Users are more likely to leave due to monetization efforts than a dated UI.

The new rework was pretty awful too, made it much harder to find stuff.

When Facebook brought Instagram for a billion dollars, they only had 13 employees.

Of course, that's the pinnacle of an extremely small and simple feature set, and avoiding anything labour-intensive like moderation, or support, or sales. Indeed, they were iOS-only until 6 days before the acquisition was announced.

I agree you could not take Meetup in its current form and operate it with so few employees.

It’s not absurd at all, Instagram was only what, 13 people when it sold right? If meetup is in maintenance mode, which it sounds like it might be, then I would have expected it to be very lean too. It might be wrong but it isn’t absurd.

Sure, if you ignore monetization completely then that is reasonable. But that Instagram is not the Instagram of today, which actually makes money from an advertising platform.

I'm not ignoring it, I'm saying it's plausible they went into maintenance mode and stuck to very basic monetization techniques to support a skeleton crew to keep the lights on until they sold it.

I've worked at Twitter, Facebook, Uber and Amazon. Of them, Twitter and Uber were wildly over provisioned in engineering headcount. Amazon was under provisioned in most of AWS. Guess which ones had massive in-fighting over projects and ownership, and which one just got things done?

Compared to Twitter/YouTube/Spotify/Facebook, Meetup is a pretty simple app. Most activity is localized to a single group, it's 100% user generated content, and it's pure digital. "230,000 organizers who create an average of 15,000 in-person events per day" isn't in the same ballpark as those other services.

Maybe they have Hard Tech problems like dealing with griefers, but the basic group system looks pretty straightforward to scale with something like Google App Engine.

That said, I'm sure most of Meetup's employees are not engineering.

I don't know if this is still case, but a good friend of mine used to be a designer at Meetup. They had a decent size team on design/brand, from the way he talks about it.

Even as someone working on a really big global app with hundreds of engineers, people also underestimate that although the core of something might not be all that complex, all the auxiliary parts and necessary edge cases add tons of complexity that you have to handle if you want a complete product. You can only get away dodging that for so long.

Not exactly five people, but Craigslist runs a huge, popular global site with 50 people:

https://en.wikipedia.org/wiki/Craigslist

You're right, it can't be that small. But it doesn't need be that large either.

This is HN so commenters are focusing on the technical aspects where Meetup is clearly bloated for what the site delivers. However there's plenty of work to occupy several FT positions, just in customer support alone, when you're selling something to consumers online. Sales, HR, finance, legal and general admin are all necessary to operate at their scale.

Plenty of Fish was being run by one guy out of his condo but still was giving the large players like match.com a run for their money. Of course match's parent company bought up most of the dating market, including Plenty of Fish, so the headcount probably went up since then.

Honest question: what are the issues that a company like meetup faces that would require a large team or significant funding?

Translations, international money handling, local market research / contacts, SEO, marketing/ads/visibility, ...

I don't think any of those teams will be huge, but I expect it adds up. Just those listed ones are probably 20+ people, so add some management layer as well.

Everything you said + content moderation, dealing with policy/harassment etc. Anytime you’re allowing end users to create content on your platform or use your service to interact with other people, you have to deal with all the policy and icky people stuff.

Just remembered, once you have income and presence in multiple countries you also need an accounting/legal team which can deal with that.

Well part of it is having the right five people. They’d need massive experience with scaling websites. But if you know how to wire things up, there’s no reason AWS couldn’t handle it.

If you want to run your own server farm or offer lots of live customer service then the equation changes.

5 people is probably hyperbolic, but I bet it's a lot closer to a reasonable head count than the head count they actually have.

Of course it takes hard work to run a site like Meetup. But that doesn’t mean that it’s a good fit for the VC business model.

Meetup is not Twitter, YouTube, Spotify or Facebook. A single Digital Ocean $5 droplet should easily be able to handle 15 000 created events a day with all their RSVPs.

Businesses are not just a website.

Essentially, yes.

How much work you have to do scales with the number of people you employ. That is, if you keep hiring, people will continue to find more work to do and do it less efficiently.

Five people in a basement could absolutely remake the core functionality of Meetup and scale it to millions of users.

The hundreds or thousands of extra people are chasing long tail revenue, niche features, change for the sake of change, and supporting each other.

Give me a million dollars and a year and I'll have a ready competitor for Meetup.

Give me $10k and I'll have a working crappy clone.

Give me a weekend and an espresso machine and I'll have an mvp.

You need way more than 5 people just to handle the customer support and abuse issues when you have that many users.

> Give me $10k and I'll have a working crappy clone.

... of a product

> Give me a weekend and an espresso machine and I'll have an mvp.

... minimum viable PRODUCT.

That's not a company. That's a product. A company has to handle abuse, support, sales, marketing, expanding into new markets, partnerships, etc.

Well said. to add to that list, real companies often have to handle taxes, international payments, regulations, laws. Also, enterprise companies (not retail app gimmicks) also have sales forces, contract people, and account execs. Those scale with the growth of the business (hopefully sub-linearly, but still scale.)

The difference you will see on the examples being cited here are those companies existing on ad revenue (or in growth-cash burn mode making no $) -- if you can ignore realities of business then yes you can run lean.

Show me an Instagram or WhatsApp that had global payments, regulations, taxes, etc and managed to do it with 5 or 10 or 13 people -- it does not exisit.

Reply on news.ycombinator.com