This is the underlying problem, and the reason why "disruptive" startups can be successful in the first place; making something more beneficial for the consumer in a big organisation can be at the expense of one of the individual units which can prevent it, even if the change would be advantageous for the business as a whole.
I’ve started and sold multiple B2B startups (and started 1 B2C as well as observing or consulting for a few others), and I agree with Justin’s thesis and most of his points.
Other than for content, getting individuals to spend their own money - even for something they seem to and claim to value - is incredibly difficult. Other revenue sources run into the issues he lists.
We need to get used to paying for online services from the start. It would change the dynamic of dumping hundreds of millions of dollars in VC money into startups that only focus on user growth.
If smaller startups were profitable, we would have more alternatives for all the basic online needs we now have like storage, and search products for digital content we cherish. Offering free products at a loss for years to steal users from competitors would not be as good a strategy as it now is.
Yes, you're right of course, distribution is another problem. I really don't see how something like this would work - if someone made me head of a light bulb startup and gave me $500 million, I'd still be quite lost as to how to go about doing things. I'd probably burn through a lot of money on consultants and all that.
I think some types of business are better suited to big companies, and some are better suited to smaller companies.
This seems like a low-end disruptive innovation. Neither 1 or 2 have to be true for the upstart to become a problem. I don't advocate over-allocating resources towards a competitor who may or may not be a threat. But submitter is already losing business. It may be time to be aggressive.
>Then the company goes well why are we paying for Slack when we have Teams?
To play the devil's advocate, that's the bean counters' job, no? Otherwise, whithout any oversight over finances, the expenses can easily spiral out of control and hurt the bottom line. Probably no big deal in the wealthy US where VC money is plentiful, but in Europe where start-ups need to be scrappy, it starts to add up.
As a consumer, I also ask myself, "well if I pay for Netflix, Prime, etc, can I just get rid of most of them and only keep one?"
There is lots of useful and interesting detail about how the ticketing industry works in this thread, and I certainly sympathize with the post's complaint. But I'd like to briefly (and mildly) chastise the idea for a more abstract reason, if you'll indulge me.
The post advocates starting a company in order to build and sell a new software application. If this resembles every other business app I've ever written or used, the developer will quickly run into a complex set of rules and requirements that is almost impossible to fully anticipate or characterize a priori.
The motivation for doing so is admittedly egregious fees charged by an incumbent vendor. Imagine one is successful in building the app, despite all of the unforeseeable obstacles. All the incumbent would have to do is lower its prices, either accepting slightly lower margins to maintain the fee pass-thru to its partners or eliminating such fees as the hypothetical startup would do.
In other words, the proposal is to dedicate years of effort to a problem in order to compete on price by 10-25%. This is generally a losing proposition. If one can achieve a 90% lower price for the same or slightly better functionality that an incumbent will find difficult to match (the famed 10x price/performance improvement), then one may have something sufficient to build a business around, depending on how difficult/capital intensive it is to create the product--ignoring all of the other barriers to adoption that exist in this or another industry.
But faced with a narrow price-performance improvement, incumbents can either lower prices or, if it's a functional improvement, invest a similar or lesser amount of capital (they already have employees, a brand name, a salesforce, etc. that a startup would have to build from scratch) to match to the feature set.
Don't get me wrong: I don't want to talk anyone out of the desire to start a company, to improve a process or an application, or to return excess profits to the consumer in the form of lower prices. But part of being a successful entrepreneur is picking the right battles and bringing an unfair advantage to bear against tough competition, something I and many others still struggle to achieve. In this case, I don't see the unfair advantage or the 10x improvement, and I think we should challenge each other to be hard-nosed realists about where we focus our collective entrepreneurial talents.
Author of the post here — I agree that the "how" might be what's missing the most. The point I'm trying to make with the post is really more of a "what" question. I sometimes talk to entrepreneurs who want to outsource their distribution through some creative channels. I'm arguing that one should fight as hard as they can to keep owning that relationship with the end customer, and that this may indeed be close to the only thing that matters in the whole chain.
That also explains why startups that have "just" a bunch of users, and no revenue, are still able to raise at massive valuations. Sure the proximate cause is that they'll be able to monetize that attention at some point — but the deeper insight is that attention is inherently valuable, because it's upstream of everything. Every transaction starts with a person paying attention to something.
I see your point, but why would a startup wig out when they scale? Presumably, revenues would scale with demand, so I would be quite grateful that my provider allowed me to satisfy the demand quickly and reliably...
Worse of that, this crazy acquisitions game makes it hard for a startup with a saner plan to succeed. A startup aiming to create a good product and try to survive the old-nice-way charging for it needs to compete with rivals that are not trying at all to be profitable since the goal is to burn money in order to give everything for free and get acquired.
I think this is starting to have serious effects in the quality of web software. An example: if email web clients has a cost, exactly like any other real good, let's say 10$ a year that's ok for everybody, a lot of startups could really enter the market in order to make a better product of gmail (and there are a lot of ways to improve it).
This is happening at all levels. Even worse it's clear that many kind of startups are hardly profitable with advertising. I don't like this system, and I could say it does not seems liberal to me, in the highest sense at least.
Your comment reads weirdly... Are you really justifying a business model in hindsight as a beneficial gatekeeper? Fees are a business model, that's it.
Let's get everyone in the market with small money, test what works for you with almost no money and then when you found your investment strategy, scale it.
Sounds like product development to me or do you think only people are allowed to code and "waste their lifetime" when they got x?
The problem of running out of money, be it a startup, a business, a nonprofit social venture, or a software project, is losing the ability to support the mission and probably losing the ability to deliver the services behind it.
It's wise to survive, but neither so reciprocity- nor goodwill-hostile as to lose your customer base. It's wise to give excellent value for money so that users evangelize by word of mouth and increase referrals with that kind of undeniably-powerful network effect.
Being an excellent business and a profitable one are not mutex. And also it's unwise to defer monetization because it changes the bargain underneath users that feels like a betrayal when that something for nothing doesn't last.
This is all very good but I think it misses a kind of obvious element, in that costs are real and very material in terms of trying to reach customer satisfaction.
It's borderline glib to infer that a startup 'should just do that thing to make it great' instead of seeking funding, because getting to the nice metrics costs money, not just blood and sweat.
This is a very 'software' kind of thinking, and one that applies to the kinds of products that are amenable to value capture at an early stage.
Stripe and AirBnB take a % of the transaction, right from the beginning, they're low-overhead and and 'early revenue'.
Google was the Search Engine 'everyone at Stanford was using' before they took their first check. Kudos to them for doing AdSense to make money, but still, they were 'out of the gates' before the money.
Most startups require material investment before they can get to revenue, let alone metrics, particularly those that require critical masses, or any kind of labour, working capital, esp. hardware.
Imagine this kind of thinking with amazing new Hair Restoration technology - imagine the clinical trials, FDA approvals, maybe it involves some costly tech, new distribution models, or serious economies of scale. Or frankly any Bio/Medtech.
Some industries are really hard to break into without already functional products and enterprise credibility, a lot of companies do not want to deal with fluffy startups with 10 'kids' that could evaporate the next day.
And this is doubly true for Europe, on the Continent you get about as much respect as the glory of your Tech Conference Booth. It's changing, but still.
In this context (successful) YC-model companies represent a pretty thin slice of startup world - those situation which there is the most hype and returns for investors immediately, which is super rational from an investor perspective, but misses other opportunities.
I should point out, to be fair to YC, they definitely invest in a lot of companies that don't fit that paradigm to well, so it's more a criticism of the ethos than their material investments.
It is very good advice these guys give, and every entrepreneur should contemplate it, but there's just too much of a reality gap in there: "Hey, if you have a great situation with customers who love your product, then you can get investors with leverage" - yes, we get that.
In reality, a ton of good companies need 'food', along with the bad ones.
I believe this is where perhaps 'truly model investors' come in, and that should be their ability to decipher between the 'junk' and those that actually do make sense but for which the metrics don't yet demonstrate.
To be slightly more blunt, investors who want in after the 'hot metrics' are basically just basic financiers, more like 'Early Stage Private Equity' or 'Small Scale Investment Bankers'. In an era where capital is cheap, everyone has money and everyone can see what 'hockey stick graphs' look like.
Again - it's very good advice, but I think it needs context.
The problem is, why would you build a startup for a business that can't afford to pay you for the value you create.
I have seen dozens of business start off this way and than eventually become enterprise only platforms or worse, programmatic advertising providers marking up CPMs by ridiculous ripoff margins.
Amidst all this fraud there must be a huge opportunity for a startup with a different model that actually delivers verifiable results (eg. higher fee for a cut of affiliated sales would better align incentives and perhaps encourage a reduction of unwanted spam to end users).
Predatory pricing. The startup business model is to lock customers and suppliers into one another through your platform and then boil the frog with price increases and cost reductions. If you don't successfully lock people in you fail. If you do, you have a nearly unbreakable monopoly and your investors can suck money out of the business basically forever.
Amazon, Google, and Apple are probably the most successful examples of this.
> The startups you are referring almost all lose money on fixed costs, but sell things at per-unit economics which make sense at scale because they are below marginal cost/COGS.
Many startups and new projects operate with effectively zero revenue until critical mass, at which point they start charging. Youtube, Meetup, Reddit, G Suite, facebook, craigslist, twitter, linkedin. The list goes forever. These all started as free services without any meaningful revenue. I don't see any difference between these platforms and Uber, which while not free, is also selling below costs.
Makes me wonder if most web2 startups have taken this a little too far. It almost seems like the startups doing somewhat well(translation: making money) are doing so by charging money for a need that already existed.
You make a very good point. But I would rather have a business model in place which means that if things overload, so does my bank account.
I wish more startups would start up with this premise in mind, as at the moment I'm looking for specific services for my company, and I have the added task of trying to determine if the cool startup I'm looking at has a business model in place so that I don't have to worry about looking for a replacement for them next year when they close down.
(And yes, I'm looking for stuff which my company can pay for, not freebies)
This is the underlying problem, and the reason why "disruptive" startups can be successful in the first place; making something more beneficial for the consumer in a big organisation can be at the expense of one of the individual units which can prevent it, even if the change would be advantageous for the business as a whole.
reply