Not exactly the same, but I have sometimes wondered how many startups are just burning other startups' money, and would collapse if VCs tightened the belt.
For instance, startup X seems profitable but all revenue is from other startups not making money yet. Like a big ponzi scheme of sorts. Most products mentioned in the article are now mainstream, but for lots of them a few years back they were small and only used by other small startups.
I used to think this every time I downloaded a free mobile game. Probably 90%+ of ads in mobile games were for other free mobile games that presumably also have the same ad-supported model and the same sort of advertisers. It feels like there's a few games making heaps from IAPs and everyone else is just advertising in a pyramid toward them while also using IAPs in a way that makes their own game basically unplayable.
There are different genres, or tiers (casual, hyper casual, midcore) that vary greatly in their economic schemes. Casual and hyper casual especially have low CPIs but low LTVs as well, and ads are a very important income source for them. Midcore and up have CTI that is much higher (can be up to $10 or even higher), and of course, they have very small % of paying users as well, but they make it up in ARPPU and especially whales, and almost don't depend on ad income (but since most of their audience doesn't pay at all, they still show ads, of course).
This one isn't so bad. You like web comics, you probably would like other web comics.
What is boneheaded is advertising shovels to me because I was tracked while buying another shovel online. I already have my damn shovel for the next 50 years I do not care anymore at all about shovels. Now if I were served a wheelbarrow ad...
Isn't this just standard targeted advertising? The back page of the book has ads for other books! The protein bar sets up a sample-stand in front of the health-food store! Etc etc.
You gotta advertise to get traction, and the best place to do that is where your users already are..
A lot of these services get quite expensive very quickly once you go over the free threshold. The free plans are just a loss leader to encourage vendor lock-in.
For example, I used to use the MixPanel free plan in a project and went over the limit and suddenly the plan cost around a thousand dollars per year.
It's nothing for a VC-funded or mid-to-large company; it's significant money for a self-funded small business just for enhanced tracking in most countries. Really depends on what kind of business we are talking about.
But the question then is whether you need to do that level of tracking at all. Sometimes of course the answer is yes (“we’re small now and hyperattention to the user demographics will tell us which feature from the list we should implement next”). But often it’s just “well let’s see who’s using our product” without a lot of analysis. Wasted money and unnecessary intrusion into the users’ affairs.
For a bootstrapped business who is talking about their ~$250/month outgoings it is. The problem is that the expenditure doesn't grow linearly with growth so there will be a window of growth where the company starts achieving "diseconomies of scale". This may or may not be factored into their business plan.
I don't think it adds up to that much, compared to the cost of developers. Where I live in Eastern Europe, a single mid-level developer would cost at least $5k/mo once you take into account taxes, office space and benefits. In SV I guess you are talking well over $20k/mo. $20k/mo is a lot of software.
They do, but it only matters if you're working in a region with very low wages or you have a lot of people that only work for a few hours.
I see those tools mostly like a fast computer, good internet, drinks and snacks etc. They really aren't that expensive compared to the employee but they will help the employee be much more effective.
I see this argument get trotted out a lot. Compared to <dev salary cost>, adding service x,y and z percentage wise is just a small increment.
The problem that gets easily overlooked is that compared to the costs being made, the profit in many cases is also just a small percentage of the revenue, and the small % increment in costs can mean a large % decrement of profits. You should always compare incremental cost of revenue to the margin, not to the base cost.
Now if you are running a profitable SaaS business with 80% margins, a 4% increase in spend might not mean the end of the world even if it was wasted on spurious licenses or other frills, but if you are a healthy service business operating on 12% margins then that 4% increase will reduce your profits by a massive 33%.
This doesn't mean that you have to make do with the bare minimum and skimp on decent hardware or development environments, but it also means the marginal productivity/retention/quality increases must in the end be worth it in terms of increased aggregate profits, and not be justified as 'you already pay me 5K a month, so whats another 250 for ...
Sure, buying as an end in itself isn't wise. If the tools aren't improving the metrics, they are too expensive and it doesn't matter if it's $10 or $1 per month.
But if they do improve the metrics, you'll typically get more done, which will translate to more revenue. Software & tools tends to be somewhere between 0.5 and 3% of total employee cost in my experience, but of course it's not priced per market, so that would be significantly higher with the same tools in an area with lower salaries, less non-wage labor costs etc.
This is part of the reason why I like working in a B2C company. B2C has it's own set of problems, but we aren't dependent on a small group of unprofitable companies.
One thing that's been really interesting is seeing how many SaaS tech companies products just don't work for us at all, for example Intercom.
Intercom is a great product, but all of their pricing is centered around how many people use your site. If you have a high-touch sales and support pipeline where a customer is worth $$,$$$s, Intercom's pricing makes sense. If you have hundreds of thousands of customers worth $$s, that's still a workable business, but Intercom's pricing is a complete non-starter.
They aren't the only one. I do wonder if they truly understand how much they are locking themselves out of some markets. Another example was the quote we got from Grafana Cloud that was more than our AWS bill.
You may be correct, but I'm just throwing out the idea that perhaps, the needs of these two different kinds of companies are so different that it is a good idea to specialize in one or the other, else you become mediocre at serving either one?
But the flaw in my logic is that the one, sometimes evolves into the other.
No this is a very good point, being B2B serving B2B means likely lower traffic and different concerns to being a B2B serving B2C companies.
My counterpoint would be that the B2B companies don't always seem to recognise this, but it's not to say they haven't internally decided not to pursue that market. I think it's a shame if they have, it's a very large market!
Good essay otherwise, but you can tell this was written a while ago:
Google has over 82 million unique users a month and annual revenues of about three billion dollars. And yet have you ever seen a Google ad? Something is going on here.
Also, you could add Amazon AWS to the list of businesses that look profitable now but are making their money off a lot of companies that are not profitable and never will be.
There is absolutely a startup support industry. I am wary of using tools that seem to cater to startups only or that are interops between other startups, because I see the whole industry as brittle. Robust in aggregate, but individual services come and go daily.
I want longevity in my tooling, and I personally feel it's irresponsible to put your product into the hands of thirty companies who are all on the same growth or die cycle.
It's a valley thing, and then maybe smaller services run by digital nomads. Most of the startups around the world are trying to solve real local problems by emulating successful ones in the USA or China.
I suspect the bubble is going to burst at some point. Not really an expert in any of this, but it seems conventional wisdom is "invest in 10 startups, see one succeed", whereas by now I'm led to believe that the chance of success is much lower than 1 in 10 given the often times ridiculous companies that are being invested into. But happy to be proven wrong.
For instance, startup X seems profitable but all revenue is from other startups not making money yet. Like a big ponzi scheme of sorts. Most products mentioned in the article are now mainstream, but for lots of them a few years back they were small and only used by other small startups.