News for hackers. Stuff that matters.
Re:Why successful startups stumble at 40 employees by mathattack on Monday August 28, 2017 @03:13PM (#15119225)
For enterprise software companies, the numbers matter, as they're roughly proportional to revenue.

For example, if I'm in build mold, I may scale my non-engineering organization as: - Each salesperson I hire a million dollar quota - Every 2 need a Sales Development Rep - Every 2 need a Sales Engineer - To generate pipeline for them, I need X dollars of marketing spend per salesperson (scales linear) and this will require a marketing department of Y people. - Every $2 million in revenue requires a support person - Management is 1 employee out of 5 - Every $2 million in revenue requires engineering support overhead of another body. - Every X employees requires an additional head in Finance and HR.

These #s are all made up, but it's a way of thinking why people use headcount as a proxy for revenue, especially when revenue #s are private.

This is much less true in consumer focused companies, or resellers.

Re:Why successful startups stumble at 40 employees by kakwa_ on Monday August 28, 2017 @06:14PM (#15120549)
Well, at least for the purely engineering part (aside from sales), I've seen the contrary more often than not.

Earlier I my career, I've seen projects understaffed with tight deadlines. When management finally released that the said deadlines could not be met, they hired really fast, and not carefully, a bunch of people to try to reach that deadline. Those new people needed to be on-boarded, trained, and sometimes put aside because they didn't have the required competences. In the end it put us even more behind schedule than before.

I've also worked at former startup bought by a big company, with more than ample resources. Yet there were tons of issues. Many teams having huge difficulties to communicate with one another, tons of features in the roadmap but very few made it into production and those in production were barely working and were a pain to keep up, the core functionalities of the product were slowly rotting away, long, annoying and not understood processes just cluttered everything and pieces of knowledge/information spread more like rumors with all that entailed of inaccuracies and wrong ideas... I could not shake the feeling that with fewer people, we could actually have had a better and faster output. It may be an extreme case of scale-up (in term of headcount) gone bad, but I'm pretty sure I'm not the only one with a story like that...

I'm a fervent believer that flat organizations are way better, you can discuss directly with people, the pain points and challenges are clearly identified and solved together. In the end everything goes way faster and it's far more rewarding as an employee to work in that kind of environment. The issue is that past a certain headcount, flat organizations don't work anymore, however, when designing our products, maintaining the headcount as low as possible should really be a big item in the requirements.

Growing the headcount for growing the headcount seems ludicrous from a purely technical point of view. But I can understand that a larger headcount might make a company more credible and valuable from an external eye. I understand it but I'm far from a big fan of it.

Re:Why successful startups stumble at 40 employees by lou1306 on Monday August 28, 2017 @07:03PM (#15120872)
> When management finally released that the said deadlines could not be met, they hired really fast, and not carefully, a bunch of people [...] In the end it put us even more behind schedule than before.

Nothing new under the sun. This is from the Tao of Programming (1987):

A manager went to the Master Programmer and showed him the requirements document for a new application. The manager asked the Master: "How long will it take to design this system if I assign five programmers to it?"

"It will take one year," said the Master promptly.

"But we need this system immediately or even sooner! How long will it take if I assign ten programmers to it?"

The Master Programmer frowned. "In that case, it will take two years."

"And what if I assign a hundred programmers to it?"

The Master Programmer shrugged. "Then the design will never be completed," he said.

Re:Why successful startups stumble at 40 employees by flukus on Monday August 28, 2017 @08:46PM (#15121346)
It was well known by then even, from the Mythical Man-Month (1975):

> adding manpower to a late software project makes it later

More then 4 decades later it seems like we still have learned.

Re:Why successful startups stumble at 40 employees by tormeh on Monday August 28, 2017 @06:21PM (#15120600)
>Well, at least for the purely engineering part (aside from sales), I've seen the contrary more often than not.

>Earlier I my career, I've seen projects understaffed with tight deadlines. When management finally released that the said deadlines could not be met, they hired really fast, and not carefully, a bunch of people to try to reach that deadline. Those new people needed to be on-boarded, trained, and sometimes put aside because they didn't have the required competences. In the end it put us even more behind schedule than before.

The parent is talking about enterprise software, where costs per customer are very high. These companies live on contracts. They may have a core product, but each new customer requires a lot of development. These kind of organizations can scale linearly because each customer basically gets its own team. Consulting is the extreme version of this.

Re:Why successful startups stumble at 40 employees by krob on Wednesday August 30, 2017 @01:28AM (#15129900)
I feel like we have worked for the same type of company or the same company ^_^
Re:Why successful startups stumble at 40 employees by mathattack on Tuesday August 29, 2017 @01:41PM (#15126089)
Having these rule of thumbs as guidelines doesn't mean that they're actually followed. :-)

One thing I've seen in tech companies is they invest heavily in engineering at the beginning, justifying it as a fixed cost. As the company grows, they stop growing the so called "fixed" cost, and hire more in sales, services and support. (So called "variable" costs) The reality is these salespeople sell custom deals to hit their revenue #s (especially in tough quarters) so the engineering work goes up. If staffing doesn't go up, you wind up with tremendous technical debt.

Re:Why successful startups stumble at 40 employees by hvidgaard on Tuesday August 29, 2017 @02:31AM (#15122507)
> I'm a fervent believer that flat organizations are way better, you can discuss directly with people, the pain points and challenges are clearly identified and solved together.

You don't want it entirely flat, and you don't want it rigorously hiarical. You have touched on too hiarical, but too flat is a thing as well. Someone with the skill and time needs to make descisions, and someone needs to keep track of things and coordinate. Both of those things are important even for small teams for 5-6 people, otherwise you end up with "The Bridge"(https://www.stilldrinking.org/programming-sucks).

Re:Why successful startups stumble at 40 employees by rjzzleep on Monday August 28, 2017 @04:14PM (#15119698)
Does it? Revenue is != profit though.

This is where it bothers me. Startups start out by taking a share of the enterprise companies lunch by doing things that enterprise companies can't because of the structures they built.

At some point when startups get successful they flip a switch a switch and try to become more enterprisy. They might get a "real" CEO, one of the big performance metrics becomes by how many hundred people you can grow the company per year and at some point if they make this transition successfully the company will start to stagnate. Back in the days when I went to some Erlang user group I was complaining about how I could solve a problem and rather than solving the problem the guy wanted to grow the engineering team to 20 people(from 5) some government contractor said that I don't get how the system works. You think that solving the problem is the key performance metric, but the actual metric is how many people the manager manages.

One of the comments I saw below is true though. They grow the numbers and then they don't have proper management to fill the gaps. But what's proper management to begin with?

If they "invest" in their people they will promote their earliest employees to director of xyz. Congrats, now some random crappy junior javascript developer is now your Director of Engineering and it will be legitimized by the $10000 Leadership 1 weekend course he got at Stanford. He will do 1 on 1s with you and try to talk about your future.

But here's the problem though. Where do you get good management? Anecdotal evidence, I once sat at a dinner table and next to me was some old school business guy with his friend, a senior management guy at another company who recently got fired, saying that the worst thing you can do is hire middle management that is weaker than you. He will always try to keep the lower tier below himself. I'm supposedly in a leadership position in my company, yet the thing that no one ever wants to hear in upper and mid management is fixing issues.

Ironically I recently did a PMP certification to see what the fuss is all about(granted, it's not really a management certificate). Nothing I saw in that formal guidebook aligns with what I've seen in real life. So if you can't get good management from other "enterprise" companies, and you can't get good management from your early employees and you can't get it from training, then why strive for that model to begin with?

It seems like the only rational explanation for all of that is to pump up to numbers so you can aim for a nice exit on the ipo.

Re:Why successful startups stumble at 40 employees by klodolph on Monday August 28, 2017 @06:53PM (#15120816)
> You think that solving the problem is the key performance metric, but the actual metric is how many people the manager manages.

Both metrics are wrong. "Solving the problem" might as well be translated as "how well the engineer engineers", and it's just as bad a metric as you might expect. You want to measure impact on the company bottom line, in the long and short term. It's just not easy to measure this metric, so we come up with a bunch of proxy metrics which we optimize for, each of which can easily lead to its special type of dysfunction.

Re:Why successful startups stumble at 40 employees by friendzis on Tuesday August 29, 2017 @03:15AM (#15122669)
In lectures on management focusing on people (not necessarily strictly HR, like what traits you want in certain roles) you often hear that one of the worst thing you can do is to promote the best salesman to head of sales, best programmer to head of engineering, etc.. Their qualities make them good at production, not management. There is a lot of truth to it.

The problem you describe is what some people call "excel managers" - managers who only juggle numbers and have no idea how processes work. I have seen some management course material and quite a lot of those focus entirely on production line management. There is nothing wrong with that. Problems arise when managers adept at controlling production line processes are given control over projects.

Re:Why successful startups stumble at 40 employees by majormajor on Monday August 28, 2017 @06:42PM (#15120748)
> This is where it bothers me. Startups start out by taking a share of the enterprise companies lunch by doing things that enterprise companies can't because of the structures they built.

It's a tradeoff. I don't think it's realistic to, outside of consulting or similar model stuff, build a company that can continue to run at peak efficiency even as the world changes. Internal small exploratory teams and a willingness to cannibalize yourself can help a lot here, but may hit a wall at some point (e.g.: what can Apple build next for future massive growth?).

It's something like: Startup starts small enough to exploit how the world has changed since the big companies built their processes, and start undercutting those companies. Then they have to grow their delivery capacity to build out the rest of the features to capture the rest of the market, but this means more devs, more management, more sales, etc. That makes those of us in the "but it's less efficient" camp uncomfortable, but it's also explicitly what we're being paid for: figure out a way to grow this opportunity to be something really big. At some point it's unlikely that it can be your baby anymore.

Then the world changes enough, and somebody else does the same thing to that startup. Satellite/cable TV is an example: from very innovative at the time, to the dinosaurs lumbering around today.

The biggest thing I've seen correlated to good management is caring (if not about the people under them, at least sufficiently about the product so that they will do what's necessary to empower the people under them to get things done right). Hard to identify in a day of interviewing, though.

Re:Why successful startups stumble at 40 employees by mathattack on Monday August 28, 2017 @06:14PM (#15120556)
Revenue definitely doesn't equal profit. Headcount is easier to measure than revenue, which is easier to measure than profit. :-)

A rule of thumb I've heard for SaaS startups is after they reach $50MM in ARR, Revenue + Margins should equal 40%. So if you're growing 20%, you should have profits of 20%. If you're growing 50%, it's ok to lose 10%. Just a rough rule of thumb.