Hacker News new | past | comments | ask | show | jobs | submit login
What I learnt after burning $10M as an entrepreneur (twitter.com/awilkinson)
485 points by dsr12 on March 31, 2021 | hide | past | favorite | 241 comments



This was, what, 2012? Asana IPO'd last year and still wasn't profitable[0]. In fact, they were bleeding vast amounts of cash and went public with the statement:

We do not expect to become profitable in the near future, may never achieve profitability, and have incurred substantial net operating losses, or NOLs, during our history.

This is the price of cheap money. VC-backed competitors can simply outspend a bootstrapped company on customer acquisition, marketing, development and hiring, and cash out without the company ever actually making a profit.

In the current environment, a bootstrapped company should ideally be: (a) niche (and therefore outside the traditional VC market crosshairs) (b) profitable from day one.

A to-do list doesn't fit that mould at all, which this guy found out the hard way.

[0] https://sec.report/Document/0001193125-20-228462/


I think it's an indictment of our current system.

The toy explanation for how it works is that we have people who take risks, and they have to be rewarded now and again, otherwise why would anyone do it? So it has to be possible to trade that risk too, so that the risk ends up in the appropriate place. It needs to be possible to run a business at a loss in order to search for profits (product-market fit, etc).

We seem to have taken this to an extreme that rewards people who haven't made something useful. If I have a great idea for new kind of lemonade stand, I can get it started at a loss, but if I can convince someone else to buy it, I can cash out with more than I put in. They can go and "develop" the idea and sell that up to someone else. Sounds fine, and should be fine, in theory. Most people could probably think of a few defenses: they're only losing their own money, losses impose a disclipline that forces them to listen to customers, and it's everyone's interest to make money in the end.

The downsides are not so obvious.

Everyone can start a lemonade stand, and sure enough there are loads of them all over. But who gets to try? People who can get access to the risk capital. It's not quite the same as the alternative world where ventures need to make money reasonably fast, or sink. It also means if you try on your own, you are competing with much more patient people who can afford to lose for longer. So now, you can't even try your own without giving a piece to someone else, who gets to decide who gets a shot and who doesn't.


I've considered this question a number of ways. The fact that capital holders are gatekeepers to innovation is unequivocally worse for federated innovation, but it has created an interesting class of companies which may never need to turn a profit yet still have a positive (and growing) net present value.

Take Wikipedia for example. They lose money running a high traffic service (edit: see below reply for clarification), but it's plain to see they hold a huge asset in terms of goodwill, usage, knowledge base, and their contribution to research and knowledge growth. Despite its operating losses, its capital value (which may be in the form of social capital) is huge and will likely remain well financed into the foreseeable future.

The fact that the service is free is not relevant: a startup offering an invaluable service that is based on years of user research, development and testing has developed an asset which helps other companies and companies pay what they think it is worth (or at the beginning a subsidized rate to take a risk to try it). Operating losses at most start ups are from continued R&D; but if they were to just declare the product as "done" and have a sufficient moat/network, they could rent seek on the asset for years - yet in many cases that's not what is best for anyone (company, clients or shareholders) - we continue to want them to innovate for the good of the product and there will be stakeholders that would rather finance this research in perpetuity to grow the underlying asset and thus the value of the product and company.

Inductively, that's a company with negative NOL but positive NPV. In the physical world this might be the same as an apartment complex that's expanding (forever). They may currently collect $1M in rent, but they are spending $2M on new construction. The new construction may bring in $5M over its 30 year lifespan but it will never be enough to outpace the immediate outlay of continued construction cost. As long as the time value of money is correctly attributed, this isn't a new idea - just one that's been pulled to an extreme.


Wikipedia does not lose money in any sense. Its revenue exceeds expenses, and net assets increase year over year. And in any case, the Wikimedia foundation is organized as a nonprofit.


A) Wikipedia is a non-profit.

B) Wikipedia makes money. https://en.wikipedia.org/wiki/Wikipedia:Fundraising_statisti...


Yes, they get enough donations to cover their costs, just like Uber has enough VC cash and loans to continue its operations. It's non profit tax status is not strictly relevant to the fact that you typically need at least as much as money as it takes to run your entity rather than less. In my opinion, this doesn't change the spirit of the point that Wikipedia doesn't make money from its free, high-traffic service but rather from favorable financing for its goodwill and assets similar to a not profitable startup.


>Yes, they get enough donations to cover their costs, just like Uber has enough VC cash and loans to continue its operations

VC Cash and donations are a finite resource constrained by their stock pool and their leverage.

Donations dont have these limitations.


Non-profits rely on donations as a revenue source. As opposed to investment, nothing is given in return. Not sure how you can compare that with VC cash or even loans.


Wikipedia is profitable. The foundation funds other projects from it.


> The Wikimedia Foundation relies on public contributions and grants to fund its mission.

The wiki (https://en.wikipedia.org/wiki/Wikimedia_Foundation) tagline suggests the underlying truth. Wikipedia gets ~half of its revenue from the investment-based endowment managed by the Tides Foundation. The leveraged capital is largely at the charity of large organizations (like google, amazon, etc) who have donated to that endowment over time, plus the remnants of their initial investment portfolio afaict.

Wikipedia would inevitably scale back in size without the continued charity of individuals and organizations around the world.


Being able to toss up a banner requesting donations on the 10th most visited website in the world is extremely valuable.

The Wikimedia foundation brought in US$104.5 million (2018) and only spent US$81.4 million (2018) even as their funding many projects independent from Wikipedia.

In the end donations are just revenue.


"Profitable" ads may sell products or services, or as in Wikipedia's case, solicit donations that make up half its revenue.


>We seem to have taken this to an extreme that rewards people who haven't made something useful.

My unpopular opinion is that the diversity push in tech ultimately comes down to everyone looking around nervously as they realize that who you are and what you look like counts more than how you think and what you can do - and what that means for the viability and sustainability of a space built on a mythos of meritocracy. There is an awareness that there are a number of perhaps raw, but good, products and potentially even careers (which may or may not require extra guidance to thrive) which are being passed over for gussied-up trivia and tripe, because the founders or team don't look the part.

That said, I don't know that profitability is ultimately the be-all-end-all of merit anyway. There are many services and institutions that should be run, even if merely at cost or in the red, because of the dividends they pay to society. I think we need to get better at identifying what those are, and especially at determining how much leeway to give them.


> I think it's an indictment of our current system.

Do you mean the USA, or tech in general?

In the UK, whilst there is VC money, it tends to be less, and more focused. I believe the bankruptcy laws in the US vs the UK are significantly different to the point where attitude to risk influences business start ups.

In the US, from this side of the pond, venture capital investment appears very cavalier, with VCs throwing money left, right and centre, hoping that they'll land on the next unicorn. In the UK, VC money is much more targeted on start ups with a plan to profitability.


The longer the easy money, stock-market-may-not-go-down regime continues, the more the lagging VC cultures like the UK and Europe will tend towards the US.

I'm not sure how the bankruptcy laws come into this though? Limited liability corporations exist everywhere, and there's mostly no suggestion of piercing-the-veil fraud or dishonesty.


That gave me an interesting thought. This would probably tend to make other countries catch up in the software industry, who don't have VC money, because they have to actually compete on merit and profit. This means they tend to take on problems with a real world application.

This isn't to say that they don't have funding, but the funding takes a more strategic approach of requiring real developments and profit, whereas a lot of US talent is tied up in adtech giants who cut products or in VC companies with no real path forward.


What we see is the opposite. Money gratituiously thrown at things like Uber and AirBnB destroys any profitability in those market segments and affect business models and profitability in related market segments.


Bankruptcy in the US has always been very generous. It’s one reason why many people have showed up here over the centuries.

Switzerland won’t give you a visa if you’ve ever had a bankruptcy. In some countries bankruptcy is a crime.


You are talking about personal bankruptcy, which is irrelevant. Businesses can and do go bankrupt every day in every country in the world.


Sorry I wasn't clear. The vast majority of businesses operating anywhere (not just USA) aren't organized as limited liability entities.

In the case of this particular tweetstorm, the business was likely not incorporated as the author kept feeding it from his own bank account.

Also note a parallel comment referring to the consequences of corporate bankruptcy in the UK. I used Switzerland as an example. The US is an extreme outlier; this has long been cited by many people as a factor in its success (though the belief is hardly universal or there would by now be many others who copied the example).


> In the case of this particular tweetstorm, the business was likely not incorporated as the author kept feeding it from his own bank account.

That is a terrible assumption. Of course the business was incorporated. There is nothing stopping you from investing money into an LLC or corporation. That's how VC funding works.


Regarding bankruptcy, under some circumstances you can’t become a company director ever again. In the uk it’s not something you can just do on a whim, and wind up a company.


The dynamics is very similar in cheap money world to non-cheap money world.

Investors always look for the best risk weighted return. Startups are risk-apetite constrained not money constrained.

Also, it's a big employment program. Think of it like that. The central bank injects money into the system, it gets allocated based on expected risk-weighted returs (adjusted for some other factors like social hierarchy, access, ESG, etc).

The formula is the same. Simple product sectors (where there are no network effects, and no other kind of moat either) are basically pay to own markets. Now in cheap money world it's an endless fight between companies with enormous warchests.

Still, as others mentioned in the twitter thread there are bootstrapped successful apps (eg. todoist), but they are completely at the mercy of user preferences. When a next such app comes some of their users will switch over. Sure, maybe the market is stable enough for them to have a small slice for the foreseeable future.


> We seem to have taken this to an extreme that rewards people who haven't made something useful.

Asana is useful. That’s why it’s growing.

The company continues to lose money not because the product isn’t useful, but because it’s so useful and popular that the investors choose to spend more money capturing market share than organic growth would allow.


> Asana is useful. That’s why it’s growing.

So is the use of opiods and painkillers, at the right scale.

VCs and the Softbank's of the world willing to spend so much of "other people's money" like they have a never-ending fountain of it make it really hard to find out what is the right scale of things.

Even with Amazon which is often criticized for not paying dividends or running without profits for so long, they knew that the revenue was there and that Bezos could turn a few knobs whenever he wanted to slow down on growth investment and start collecting some profits. Is that true for Asana? How much more market share do they need to get to actually being profitable? What would happen if they slowed down their marketing spending?


If it's been nearly a decade for a glorified to-do list app and they still aren't profitable, then it is in fact not useful enough to justify the effort invested. That's what market signals mean.


>Everyone can start a lemonade stand

Lemonade stands are a perfect example to illustrate business concepts.


To put a bit more nuance (or perhaps just spin) on it, the problem is not VC itself. VC can play a vital role in supporting innovation. Unfortunately, what it does far more often is let a bunch of marketing/financial dorks with strong connections to the money pipeline enter a young market and obliterate all of the actual innovators. It's not only innovation that gets lost, but diversity as well.

Case in point: continuous data protection. If you were around ca. 2004 you might have heard of this. Like backup, only to any second in the past. Nowadays we might say git for your entire storage system. Anyway, I was a fairly early employee at one of the original companies in the space. It was our marketing VP who coined the term. We worked really hard to promote the idea, and often collaborated with our competitors in that (much like Asana in this story).

Of course, that initial spirit of shared mission eventually disappeared. At that point, VC-backed and BigCo competitors were all spinning up their own products, backed by many times the engineering and (even more importantly) marketing budgets. They didn't actually win head to head, because they didn't actually have any products, but they totally froze the market. Storage is a hard market for a startup in the best of times, and these weren't the best of times, so that company ended up in a fire sale to Veritas/Symantec. End of story. It's pretty unremarkable in itself, except that I could tell almost the same story about half of the other startups I worked at.

None of this is necessarily bad. Just "creative destruction" and all that. In the end good ideas do get implemented and benefit users and make someone a lot of money. Unfortunately, those people are rarely the ones who innovated. When it happens over and over and over throughout the industry, it starts to seem like VCs are setting up incentives to be a copier (at best) rather than an innovator, and I'm not sure that's healthy in the long term.


I think its just the nature of innovation. It takes 1/4 the effort to copy as to create new and this is true in marketing as well as engineering. Call it the second mover advantage, where someone else has already paid the cost of innovation, risk reduction and marketing (convincing people that they want x) and the second mover capitalizes on the new zeitgeist that they created but with lower overhead.


> Call it the second mover advantage

That is in fact exactly what I called it when I wrote about this on my blog. It's definitely a part of how things happen "naturally" but that's also why I think VCs should act to counter it instead of magnifying it.


While I agree with your general sentiment, I should caution that every S-1 (as well as 10-K) contains such horrifying language; if not regarding the firm’s present losses, then regarding the many ways the firm could begin losing money and/or value. The entire purpose of these statements, and the sections in which they reside, is to provide appropriate cover for the firm when potentially desperate investors begin to think about legal action against the company. This way, the business can point to their compliance with their legal obligation to disclose potential deleterious outcomes as well as to the fact that the equity holder was aware of any associated risks at the time of their investment.


Well, yes and no. Every disclosure document says something along the lines of "profits aren't guaranteed". But not every disclosure document says "we've lost shitloads of money and haven't made a profit".

I'm not sure where along the line IPOs changed from "profitable company raises money from the public" to "private investors parachute out and dump their losses on Joe Schmuck".


Mark Cuban can probably tell you.


That's quite a lot of hyperbole...

Private investors "parachute out" only 20% of their stake in an IPO to investors who willingly buy into the story because they believe (rightly or wrongly) in the upside.

It's not exactly like Joe Schmuck is unwillingly forced to hold on a hot potato


Sure, that was 100% facetious. And I agree that Joe Schmuck is buying with eyes wide open.

But the underlying point is that the markets are disconnected from reality, and - fair play to them - the VCs are capitalizing on it.

One day I'll actually set aside a few hours to go through a list of all tech companies that IPO'd in the last 5 years, and see which ones have actually made a profit. Gut feeling is that the "flashy" ones are still burning through cash reserves (Asana/Slack/Uber), but the more boring ones (Rackspace/Pivotal/etc) are probably sitting pretty.


> One day I'll actually set aside a few hours to go through a list of all tech companies that IPO'd in the last 5 years, and see which ones have actually made a profit. Gut feeling is that the "flashy" ones are still burning through cash reserves (Asana/Slack/Uber), but the more boring ones (Rackspace/Pivotal/etc) are probably sitting pretty.

You might find site helpful. https://postipononprofits.com/

This is outdated a bit in terms of numbers (doesn’t include recent years and excludes many companies), but the broad point remains. You are right, there are ton of companies built around the idea of dumping it on someone else before turning a profit. This has become the ‘formula’ for building companies, unfortunately.


Indeed the precise wording itself is basically boilerplate that one sees all the time with startups.

Though such language and sections aren't solely about cynical "legal CYA", but valuable warnings to investors. I would not want to take money from investors who think they are getting a sure thing, only from people who understand the risks.


It blows my mind that so much money is being spent on companies that seemingly have no road to profitability, or expect that their users will stay when they rise the prices to a profitable level.

I wonder how many more mobility startups we can dump money into before accepting that it will fail like the five identical companies before it.

I always joke that my humble little website is more profitable than Uber, Airbnb, Lyft, Pinterest, Slack, Dropbox and a few other companies combined, and was so since day 1. (E: not Twitter)

Maybe I think too small, but when I divide these companies' losses by their number of users, I can't imagine how they could possibly work if they weren't sustained with massive amounts of money.


I used to joke that I make more money selling cars than GM does.

Also consider all the secondary effects as hopeless companies pay for B2B services from other companies in the chain. I think the value of ads and tracking is vastly inflated as a result of this bubble.


Yes but GM does make money. They're not selling cars at a loss. Their moat isn't VC-subsidised cars.


Well their moat was a government bailout. A decade or so ago they were losing billions, meaning thousands per car, given that they sell a few million cars.


GM is still different. Automobile manufacturing is incredibly capital-intensive and it is incredibly difficult to weather reductions in demand. They still have to pay for the factories and equipment, regardless of of how many people are buying cars. They have to keep suppliers going during lean times because they can't manufacture every part they use. And they also have separation agreements with both dealerships and most workers, making it costly to reduce production capacity.

It's trivial to send an auto manufacture into a financial tailspin. Frankly, it's amazing that so many manufactures have survived for this long.


That's a good point. Ah the things you can do when you hold the economies of entire cities hostage.


Twitter had very good years recently. (The user experience has not improved though.)


You realise that Twitter is profitable, right?

And Facebook wasn't profitable once, too.


Facebook is a rather extraordinarily bad example to hold up. They lost money for a short amount of time for a VC backed company - about 4 1/2 years. Asana by contrast is 13 years into losing money.

Here is Facebook's early financial history (2004, 2005, 2006 all showed losses):

2007 | $153m sales | -$138m net income

2008 | $272m sales | -$56m net income

2009 | $777m sales | $229m net income

2010 | $1.97b sales | $606m net income

2011 | $3.71b sales | $1b net income

By the time Facebook was 13 years old it was running at ~$16 billion profit.


Sure.

My point was exactly this really.

Both Twitter and FB were founded at roughly the same time and went public at roughly the same time.

One was a huge success, one wasn't.

Asana hadn't been a big success, but Trello was.

There is some luck involved here, but the VC game is about trying to find the outsized returns, not avoiding losses.


Has it made up for the decade of losses?

EDIT: Yes, it's apparently up ~53 million now. I replaced Twitter with 5 other giant companies that are losing money.


This does show you the fallacy in your line of thinking, right? Twitter was on that list, and now it's not. In 5 years, maybe none of the companies in your list will be in it; sure you will easily find "replacements" that are just as popular/well known, but the thing is that that's exactly the model: "grow aggressively, capture the market, become profitable once you're market leader". It doesn't always work, but it's hard to dispute that it sometimes works, and when it does, it yields tremendous profits. It's not a bug, it's a feature :)


But 53M is not much for a decade of business at that scale. For the size of some of these VC investments they could just put it into the market and get similar returns, I'd think?


The VCs made their (large) returns on the share price of Twitter, not anything to do with profits Twitter made.

Profits are just money a company has been unable to invest in growth.


Twitter hasn't captured the market, but has become only slightly profitable compared to its industry peers. Meanwhile Doordash or Deliveroo or Uber could capture entire markets, but that would rapidly make them prone to regulation or outright bans.


> In the current environment, a bootstrapped company should ideally be: (a) niche (and therefore outside the traditional VC market crosshairs) (b) profitable from day one.

The fact that it is bootstrapped prevent it to operate without being profitable. That is tied to the definition of bootstrapped.

On a more nuanced tone. You can be bootstrapped in the note-taking space. Yeah this specific guy did not manage to do it. But he is a specific case (Excluding the fact that he had 10M of cash, who has that ?). He wanted the differentiation point of his company to be the product and only product. This is a pretty risky strategy to follow. Usually you want to differentiate yourself in several areas.

Seriously any company that goes out of business because of another company doesn't tell you the full story. There are very very few areas in which you will go out of business because of another company competing with you, and certainly note-taking, a space populated with thousands of companies


> He wanted the differentiation point of his company to be the product and only product. This is a pretty risky strategy to follow. Usually you want to differentiate yourself in several areas.

What ways can you differentiate yourself? I mean he admits that Asana was VC funded and started with a crappy product that later became fleshed out to be better than his by hiring better devs, better designers, etc. While he was focusing on marketing and whatnot, Asana was burning VC money for marketing and hiring a better team. So yeah, honest question, did I miss out on any other factors?


Early investors probably don't care (much) if Asana will ever become profitable. They probably got back 10-100x their investment via the IPO, so they have already "won". That said we need a more level playing field for smaller companies, as large VC-backed startups suck the air out of many markets that could support a much more diverse ecosystem otherwise.


There are a bunch of competitors.

And just as the twitter thread demonstrates users will ask for more features and you need a load of cash to deliver. And Flow is still around.

The problem is not diversity, it's productivity of startups and more directly the cost of product development (which is of course itself inflated due to endless money).


It's a pricing issue too. Asana may have over-invested relative to what people were really willing to pay for those features. However VC backing makes price-based competition irrelevant, because your competitors pricing isn't real.


the problem is insanely cheap VC money


Sounds like Snowflake. If you dig into their income statement you see on 264m revenue, they spend 293m on Sales and marketing. [1] Whether Snowflake is actually a great data warehouse product or not, the main driver of its success is that they spend more than a quarter of a billion dollars per year selling it.

This is the state of play in the SaaS business.

[1] https://www.sec.gov/Archives/edgar/data/1640147/000162828020...


That is pretty normal on high scaling SaaS solutions. If you earn 100k a year on a solution, and you know they in average stay for 5 years, you might spend 150k to acquire them as it is still a good business.


By that logic, as long as you're growing you'll always be losing tons of money and never be profitable? How is that a good business model? I mean we're only talking about revenue vs sales and marketing spend here -- the business has a load of other costs too.


That's a great business model if you have investors. You get a proven long-term return of 25% per annum.

If you're doing it out of your own bank account, you'll bankrupt before realising this return: it's not a lifestyle business.


You hit a tipping point with enough market grab that customers start to approach you without advertising.

There is volume discount as you grow.

There is upsell potential on the customers you've signed.

As your signed customers grow their pockets grow to spend more on the solution.

Takes about 3-6. If I use Snowflake today at company X without issue and I switch jobs in a couple years to a new company that hasn't picked their platform, what do you think most people will pick?


You're effectively gambling on running out of new customers at some point. What a world!


> as long as you're growing you'll always be losing tons of money and never be profitable?

Basically yea. This effect is just assumed in the world of physical products. As long as you're growing you're nearly out of business because you need to use N revenues to buy your next N+1 of stock.


Snowflake is basically the alternative to Google big-query. It's slightly cumbersome and needs a better Web UI but pretty good. It meshes the idea of public and private databases.


Atlassian is also targeting them with Jira Work Management: https://www.atlassian.com/software/jira/work-management

And could well be priced as a loss-leader purely to upsell the rest of their ever growing stack in the future.

I wouldn't be betting on Asana surviving such a direct, sustained attack from a well capitalised company.


And everyone will wind up using it, paying for it, and hating it. Lol.


Some companies are willing to sell dollar bills for 75 cents[0]. It's fine if you take on a business risk and it materializes (or pays off), but I wonder if there are companies/founders/executives who maliciously spend investor money with no expectation of positive return for the company.

[0] https://www.nytimes.com/2018/05/16/technology/moviepass-econ...


Absolutely. Everyone does it though, including the VCs. Greater fool theory is dominant investment theme at the moment. It’s terrible.


Why is it terrible if it works? What would you recommend? How could we change it to something better?


It depends how you define terrible. My background is actually in trading, and so from an ROI perspective it is what it is. In markets there are winners and there are losers. I’m fine with that.

But from a broader social consideration of allocation of capital, I think it’s terrible because (as another comment says) it’s not sustainable. Eventually the market will find the way forward, this is the exuberance period of free markets. We don’t know of a better system imho. But it doesn’t make it any less terrible.


It doesn't work because it is unsustainable. This will never hold. There is no such thing as free money.

We should try to get (back) to the point where companies actually need to make a profit in order to survive.


It is sustainable and there is such a thing as free money. Ultimately all this useless activity traces back to governments and their obsession with GDP. Asana has never made money but it spends money so it counts as economic growth. The money itself comes from QE and government manipulations of the markets via printing. There is in fact no reason why governments cannot keep forcibly reallocating money from savings to startups forever, and given that sound money has vanished from the political conversation for many years now, it likely will.

It's a depressing thought, really. There is no use in attempting to play by the rules and building something useful for an effort that can justify the results. Governments don't care about that. What they care about is that you're doing something. Exactly what you're doing, doesn't matter.


The cost of money changes. If your company is only sustainable with free money it will inevitably fail when money costs something to borrow.


Like most pyramid schemes it works until it suddenly stops working.


because eventually someone gets left holding the parcel when the music stops, and that someone is usually you, me, and the rest of the taxpayers


definition of a bubble: people buy to sell


Power and revenue aren't the same thing.


I hate it when this happens. Because 9 out of 10 times the VC backed company eventually goes belly up but the bootstrapped ones that would have made it work will be collateral damage on the way to acquiring that little bit of knowledge.

I had this happen with a company called Spotlife that competed with us in a niche (live video over the internet for consumers, something we now consider to be common and easy, but which was neither back in the day). Fortunately we had some very loyal customers and held on until they died but it was pretty ugly.


Do they go belly up? As far as I can tell bankruptcies of Valley startups stopped decades ago. I can't remember the last company that literally went bankrupt. Failing firms are always bought for undisclosed sums. VCs will not let their firms fail, they will always force one of their more successful bets to buy the less successful ones under the cover of some plausible-but-dubious strategy.

I'm pretty sure the majority of all acquisitions in the VC-backed startup space are like this. They aren't real in the sense that they'd be happening independently. They're the VCs bailing themselves out.


One of my favourite companies is teamwork.com, a great tool set in the same space, makes the fact that they were bootstrapped from the start and are profitable all the more impressive. https://www.irishtimes.com/business/technology/teamwork-pays...


> Asana IPO'd last year and still wasn't profitable

This is interesting to hear. From my research last year, it’s more expensive compared to Jira. I had to use Asana at work for high-level planning and it was, in my experience working as an engineer, an inferior product compared to Jira. The only upside I could see is the reporting/project overview and search. But these features are only useful to the manager. For me, it’s just a cumbersome tool. I’ve tried to get used to it, but I wasn’t sold.

Some things that made it difficult to use Asana: - No markdown support for the longest time. - No code block formatting. - Old comments collapsed by default, which hides the full context. This caused people to miss information and ask questions that were already answered.

I do realize Jira and Asana are not trying to address the same problem space. But to manage a team of engineers I would pick Jira every time.


Startup advice ;) if you are not profitable just find a SPAC. All the dumb money (institutional, hedge) will follow it


How the heck a company can continue to run without being profitable is beyond me.


Because investors believe that the equity of the company is undervalued compared to the future cash flows. I'm sure plenty of poor investments have been made on this basis and you can't keep it going forever, but it seems pretty sensible that in some cases a company has great future earning potential without present profits. Especially on a site run by a firm specialising in such investments!


I think it's the assumption that there is a market there and if they can outlive and outspend the competition, eventually, they'll have it all to themselves.


I understand this, but at some point you would think investors would cut their losses.

I mean... _years_ without profitability. That's a long time.


Real estate is worse. I've watched developers sit on property for decades. Slowly buying up property around them as it comes up for sale. Just so they can build a fancy strip mall.

If you try to bring rationality to money you're going to lose. None of it really makes sense and the rules get rewritten all the time.


It doesn’t sound like he was on the bootstrapping koolaid as much as he thinks he was. Dude just vc’d his own company.

So his first conclusion doesn’t fit. You can’t play the vc game, get out vc’d, then conclude that bootstrapped can’t compete. Well I guess you can, but you shouldn’t!

It’s entirely possible there were enough people who liked Flow and didn’t need mobile and native apps for Flow to have been the Basecamp of his dreams. We/he don’t know because he didn’t try for it.

Edit: looks like he’s a vc now. Can’t help but wonder if instead of horribly misinterpreting his own story, he’s intentionally misrepresenting the conclusion. Frighten bootstrappers in to getting in touch to sell or take on funding.


Not really sure I follow. Pretty sure bootstrapping includes spending money out of your own bank account otherwise how else could your small business pay for anything on the first day?


But the key point of bootstrapping is that you have very low spend initially and you really only grow at the rate that your revenue allows you to plow capital back into the business. I agree with the parent poster, plowing tens of thousands into your business month after month isn't really bootstrapping, it's a VC size of 1.


Fair point, I wasn't sure what the distinction but I guess it's minimal investment vs dump every dollar you have?


A bootstrapped company should be able to fund itself without outside investments. "Outside investments" here means from a VC, or anyone's personal accounts.

For example imagine a hypothetical company that made $100k and has one founder and one employee. If the company paid $60k on salary ($30k/each), then it'd have $40k left-over basically. A bootstrapped biz would re-invest that $40k. Next year if it made $150k, it might have more to re-invest. That's bootstrapping. This is also important for tax reasons. If it's always the company's money, no taxes. If the founder takes his $30k salary, pays income taxes/payroll taxes no it, then re-invests it in company, now he's paid taxes and the biz is not better off vs having its own money.

Now to be clear – I think most people are fine with a founder putting in a bit of their own cash and still calling it bootstrap. But if it happens month-after-month and the biz is not self-sustainable without it, then it's not bootstrapped.

To spell it out a bit more, "bootstrap" comes from the phrase "pull oneself up by their bootstraps." It's originay meant to be an example of something physically impossible, but today in practice what it means is self-sufficient growth.


I think it's really less about the total dollar amount than the timing.

That is, obviously bootstrapped founders need to put in some money (i.e. they'll probably need to pay something to a cloud or infrastructure provider, etc.) but the point is that once they have those initial, relatively low costs, they don't increase their spend until they have the revenue to cover it. That was not the case here, where the founder was enabling more growth and expansion by plowing more of his money in month after month.


Bootstrapping is about the using income the business makes to run itself from the get go, and at most starting with a very small fixed amount of money. Self funding is taking money from your own bank account to fund the business. Very different things that are often conflated and confused for one another.

If you decide to bootstrap a business and set a budget, that's it. You don't get to add more to it, there are going to be very small essentials, but you don't spend any more money. Perhaps you spend a little money to buy domain name, but then you run on a dev server till it brings in income. You absolutely don't spend on marketing or paying anyone, until it brings in income to cover that cost.


I think the difference that the OP might be getting at is as follows.

The low-cost style bootstrapping would involve low to fully covered with savings personal living expenses for one or two people, and ramping up slowly, prioritizing profitability over growth.

It seems like (I don't know the whole story) in Flow's case, the decision was made to compete on an excellent consumer product. Those are hard to build. The fact that 10mil was wasted _probably_ indicates that they hired very good designers, programmers, etc at market rates to produce world-class work for a consumer product ("better than Asana"). Meaning, the typical bootstrapping to profitability ASAP was not a priority in this case.

However, if you sink that much money into a not very profitable business, after 12 years you might expect _something_ back. Maybe not the VC happy path 10x return, but after so much resources your expectations on return and growth rate etc etc are higher than a low-key bootstraper who might choose small-scale profitability over growth. So yes, it is self-funded, but he basically can't end up with a mom-and-pop style $200k/year niche business after he sunk so much money into it, so he de-facto pushed himself into "expected VC returns" territory.


There is a rule of thumb in SaaS that you should eventually reach annual revenue equal to the amount of capital invested. Flow reached $3M ARR on $10M invested, which is a really poor outcome.

I have definitely seen many venture backed companies that never reach this benchmark. That’s just life.

But at $3M ARR, there is not much hope for recovering their $10M, let alone $900K ARR...


Bootstrapping means starting the company from a small sum and having it sustain itself from its own profits without further infusions.

This company was never profitable, and the author invested $10M in it over time. That is definitely not bootstrapping.


That's an excellent point about definitely spending way over what's usually understood as bootstrapping. However, one could counter-argue that bootstrapping would have made this even more painful, and this market is just not cut out for it. Kind of like you can't really bootstrap a Facebook replacement today, VC money, personal wealth or not.


> you can't really bootstrap a Facebook replacement today

why not?


Probably because server usage alone will be $50k/month for just few million users. Customer oriented products (B2C, C2C) rarely makes money in initial days.


You're assuming you would give away your Facebook replacement for free, and not make any money from it up to millions of users.

Just because Facebook began that way doesn't mean its replacement must.

Google Docs was free when Word was downloaded and paid for.

Whatsapp was paid for when other chat was free or included in your contract.

There are many business models and ways to grow. It's hard, but I think you could bootstrap a Facebook replacement... over a very long time!


Andrew runs Tiny Capital. They buy internet businesses similar to a private equity fund. They do very little traditional VC investing.


At the time of this story they were doing a lot of agency work. They tried to start their own company and poured $10M of investment into it, VC-style.

That’s how they burned $10M, as per the title.


Don't miss Dustin's (Asana co-founder) response: https://twitter.com/moskov/status/1377007980063809541


> Even OUR OWN Google keywords were plastered with “Asana vs. Flow” paid links.

I don't know if this shock is in good faith, but it is very naive for a founder of a SaaS to not be aware of such marketing strategies - brand PPC campaigns are basic.

On a side note, I've often wondered how much money Google must make from people paying to fill up the ad slots for keywords which are their company name and they already rank in position 1 for. "If you don't bid for them, then your competitor might - so give us your money".


The real story is how much they charge company A to defend their search terms vs how much they charge company B to usurp company A's search term, ie they are using all their data to extract maximum rent and not running a "fair" auction.


What do you mean?

I’ve run plenty of brand campaigns and the clicks tend to be an order of magnitude cheaper than running keywords on competitors keywords.

My experience typically lies with local companies though.


Cheap on a per click basis. When you take into account the typically high click-through rate, Google is making more on a CPM basis(per 1000 pageviews). Google is making a much higher rate than even $5 non-branded clicks-- and this is for traffic that otherwise would be "free" with organic.


> When you take into account the typically high click-through rate, Google is making more on a CPM basis(per 1000 pageviews)

Is this your direct experience? In my campaigns it's typical for me to see 5% CTR for the keywords I target and then a ~30% CTR for branded terms (depending on the brand name)

Branded clicks are more than 6x cheaper than non branded so the CPM is still lower.

Again, that at least has been my experience. And regarding the "would have been free" argument, I've gone back and forth and have seen better results with branded campaigns than not having them on. But I've seen accounts where that was not true.


My experience is the same. I've worked with tiny startups as well as giant enterprises and the brand campaign is always super cheap.


I'm curious, what kind of CTR do you usually see for branded campaigns? I tend to keep call extensions off of branded campaigns (so existing customers scroll down to Location Listings).

I see ~30%.

And again, this is only for local service businesses.


There's a line of thinking, popularized by DHH of BaseCamp, that skewers Google for allowing competitors to place ads on a brand's keywords. I can totally see how this would be infuriating. You have to pay money to just "protect" your own brand's name.

But I don't think we should ban this sort of thing, though it's a bit of a toss-up when it comes to how it may promote competition. It allows new competitors to come in and compete with established brands. But, it also allows deep-pocketed incumbents take customers away from smaller companies.

It's similar to the argument that political ads should be allowed on Facebook because it is an effective way for fresh candidates to challenge incumbents. It can also be a way for incumbents to drown out challengers with their warchest.


This. It sheds some light on how this whole situation actually went down, and in hindsight, this disconnect between perception and reality may well be the root cause of Andrew's failure.


This whole story sounds like another Andrew Wilkinson embellishment story. Between this and claiming Slack's only reason for being worth as much as they were was because his design team work early on.

My takeaway from this story isn't nearly what Andrew believes his is. What I see is he was so tunnel visioned on trying to crush Asana and be not only the market leader but only option that he neglected his customers needs and wants until they just bailed on Flow for alternative products.

Death by a million papercuts, sure, but mostly from his own errors.


Or, Dustin is lying or remembering himself in the best light.

We'll never really know.


Spending an hour to gloat is cartoonishly villainous. I thought Andrew was spicing up the story there.


Hey, he works on projects that help humanity thrive. Like facebook, for example.


It's all a bit "He says, she says", isn't it? I think it's not really the point of the Tweets anyway


“He says, she says” generally implies disagreement.


But does the Bond villain ever know they are the Bond villain? Or are they the protagonist of their own self directed story.


Please deal with Harris situation: https://www.youtube.com/watch?v=U6cake3bwnY


I'd say that was when he misread the signs, but probably was never going to be topdog in the deal. Maybe ego got in the way.


The whole thread really boils down to missing this key point from 37 Signals:

| 5. Spend less than they make

Had they done that, they could survived long enough to find their niche. But throwing out multiples of your revenue out of your own pocket means you're on borrowed time.


Yep, that's how Fastmail remains bootstrapped too. Our hiring plans are basically:

1) "how much profit have we managed to make?"

1a) (modulo exchange rates - it's more balanced now, but a few years ago almost all our expenses were in AUD and our income is all in USD)

2) "how many people can we hire with that profit?"

2a) (save a bit for a rainy day)

3) "hire the next roles off our roadmap while fitting within our means"


Right, and beyond the value proposition of the email service, you are also serving your customers _by_ being profitable. Because this suggests that your service will be around in many years and hopefully won’t change in ways that users won’t appreciate (which is common with funded companies these days).


That's definitely the plan! One of the SaaS providers we use recently did an "upgrade" and now our team is struggling with everything being much slower and worse than it was before. It was a good instructive point to tell everyone in our all-hands Quarterly meeting yesterday that we are that SaaS for many others, and we have a responsibility to keep providing them a good service.


One area that is missed is that 37Signals didn’t really compete with heavily VC backed startups.

The question becomes what happens when your growth rate goes to 0, and your customers are churning because your VC backed competitors is cranking out features? It’s not clear that “just” spending less would have saved Flow.


That's true.. though you could also look at it the other way which is that they didn't capitalize effectively enough to stay competitive with Asana. The niche is not always a safe place to hang out.


You are basically describing the Ben and Jerry's vs. Amazon decision from Joel Spolsky's blog [0].

> Both models work, but you’ve got to pick one and stick to it, or you’ll find things mysteriously going wrong and you won’t quite know why.

[0]: https://www.joelonsoftware.com/2000/05/12/strategy-letter-i-...


20 years on, it feels like everyone should be strongly considering the Ben and Jerry's model. Even if there is literally no competition in the field, the threat of Google or Amazon or Facebook throwing buckets of money and top engineers to put a company in the grave before it's even properly born is more of a reality than we ever could have imagined in 2000.


Asana, the company that crushed them literally spent considerably more than they earned and beat them.

The story is not 'save your money' frankly the story is 'have the most leverage'.

There's probably a lot more to it obviously, but it doesn't matter what this CEO did, he was going to be outgunned most likely.


Both approaches can work, but lead to different outcomes. The market for "todo list" apps is absolutely big enough there's room for more than one company. The beauty of the bootstrapped model is you don't need to "win the market" or "beat the competition". Is it making a profit? Congratulations! You're now the owner successful company. Want to grow the company? Re-invest the profits, not your personal life savings.

If the goal is to "win"... ya, you're probably going to need to spend money like it's someone else's, and should take that VC money.


This idea that you 'only need 1% of the market to be rich' is not always so great.

There are not that many companies that can do this, usually early movers, people who caught a wave of interest.

For something just a bit more complicated than 'To Do' he was going to need to take on VC money.


The idea that a todo app is a capital intensive business is laughable. I have more impressive ideas that can be realized with much less money and that even includes hardware projects that would require hundreds of thousands in machine tools. All these guys really needed to do was make one really good iOS app and make sure they don't spend too much money on it.


> The idea that a todo app is a capital intensive business is laughable

It’s not just the building of the app that’s capital intensive - it’s the customer acquisition cost.

A highly capitalised startup with a war chest can afford to spend $100+ per customer. You can’t.

It’s like the OP thread says - you can’t field of dreams this. “If you build it they will come” is not a business strategy when your competitor is spending millions of dollars on ads.


> 4. Every developer in the world wakes up thinking “I should build a to-do list app” and people love jumping between productivity apps and workflows. There is no moat in productivity—avoid it if you can.

This was the elephant in the room so am glad he acknowledged it. Just seemed like an inherently foolish thing to base a business on and then to toil on it for 12 years...


At first I was thinking he just wanted an easy project to get his feet wet and expand his business with relatively low risk if it does fail. I still cannot comprehend how he thought he has to compete over something that other developers use to practice their software development skills and kept doubling down. This is like the "make the best pot vs make as many pots as possible" pottery anecdote.

He also completely failed to follow through the bandcamp thinking: You're supposed to put in the bare minimum effort needed for your own company and then try to spin it off as passive income to pay for the development costs and then use the profits to improve the app. If there are no profits then that's when you have to stop.


The best product doesn’t always win, and product is not a longterm competitive advantage.

I have trouble swallowing this one. He decided to compete on product, but instead spent more of their limited resources on other things.

We started burning money on ads and hiring sales people

In order to stay competitive, we had underinvested in our engineering team due to cash constraints and stretched them across mobile, desktop, and web.

We started to get an endless stream of bug reports from our customers.

Our clients were unreliable and had syncing issues

This smells to me like neglected product and over-stretched developers. Chasing new features instead of building solid foundations by selecting the most important ones and getting them right first.

As a customer, as soon as I see something like "sync issues" that's a dealbreaker. If I can't trust your product to keep my data safe, I'll stop using it.

It does look like they eventually figured that out:

We had to hit pause and spend years — literally years — rewriting all of our clients

But by then the competition had decisively out-producted them:

One day recently, I looked at Asana and it slapped me in the face:

It’s better.


Completely agree! The best product does always win. The founder did not focus on making the product better but jumped from marketing to design, to raising money, to expanding the team. The product did not keep getting better and they lost.

1. If you are in a competitive VC-funded space, it’s foolish to compete without raising money. Don't bring a knife to a gun fight.

Again, not true. You win if your product is better. Think outside the box, and make something that no one else has made and users will love it and you will win. I'm not saying that is easy to do. Just that's what wins.


I hate seeing this very thing play out in real time, but I find it particularly sad when investors and VCs then takeover the board and outmaneuver the founders and Csuite to the point they barely run the company anymore. They take advantage of technical founders without the board level chops to do this. Then short term focus or otherwise misaligned incentives mean the board slowly, but unknowingly, runs the company into the ground. (by doing stuff like hiring a bunch of marketing people while the technical debt starts piling sky high...)


Wow, one twitter thread worth reading.

I think this is one of the reasons that sincerely the browser should be normalised, have a consistent API across all platforms (including performance characteristics, and native things like movement, sound, etc) and continuously improved, it just evens so much the playing field - by extension it would allow people to actually build novel and interesting ways to use a computer instead of figuring out how to build the same goddamned buttons across platforms. You can't just outcompete the deepness of some pockets when you have to juggle turds across so many layers.

I never tried flow, I did use Asana and liked it compared to whatever else was there, I think it's a good product, although they might be a bit overboard on the rainbow confetti scale, but I still think all of these platforms sincerely lack an approach to digital organisation, they all still behave like their paper based flows/ideas.

Millions upon millions of $ plus human hours, spent on making a todo app working across platforms. It's a waste of human and societal potential.


It's even worse when you consider that in a very real way, the differences in these platforms largely exist to moat themselves.

Apple creates their own programming languages then implements proprietary frameworks inside these languages.

So you have developers that really have to specialize in doing that kind of development if you want to do native. So you have an "iOS" team and an "Android" team all to deliver the same thing as your web app but on locked down proprietary devices. It really only serves the phone vendors themselves.

But I think React Native/Electron have done a decent job at solving this problem as much as it can be solved. It's important to acknowledge that web standards for html/JS really still aren't geared towards creating fully interactive applications. A lot of what exists is hacks. So I don't think you can fully lay the blame at companies for not wanting to build their platforms entirely on web technologies.


Yeap, the biggest problem in my view is not the usage of other technologies other than web, is that those technologies don't transfer. I would be fine with other technology that allowed the interop, I've written quite a bit in js and interesting stuff but it's far from being my fav thing for sure.

> But I think React Native/Electron have done a decent job at solving this problem as much as it can be solved.

Well in a way they do, and I'm not dissing the work done on those fronts, but it's not the same thing. These solutions and development frameworks (even the native ones from the OS venders) are mostly cookie cutter things. They don't really allow for exploring the breadth of what is available on our computers (desktop and pocket) in any meaningful or innovative way when it comes to UX. In the case of Electron and others in the same vein they still circle back, because they're implemented exactly using browser technology.

> It's important to acknowledge that web standards for html/JS really still aren't geared towards creating fully interactive applications

That also shows how much better it could be if a "building blocks" OS layer/API that is responsibility of the vendor but the same across platforms existed, because you can still build amazing experiences using something that wasn't made for that all. (edit* I mean building blocks not as in "here's an accordion widget, or a button", I mean building blocks to build those things, stop thinking you somehow reached nirvana)

Even ancillary projects like browsers would benefit from something like that. Probably this is not in the interest of many different factions due to many different issues and objectives.

The internet, the www, and the browser is proof, even with all its warts, that this interoperability is good. Transferable skills are good. Having better foundations is good. Most of the interested companies make billions, they should push this forward. There's plenty to differentiate between competing OS/hardware as it is. Better, more maintainable projects is good. We'll fill the gap of what we don't need to do with more useful work for sure. Native will still make sense for a whole host of things.


Google is also trying to imitate Apple here, with Dart and flutter. The only reasonable explanation for these (especially Dart being exhumed from its thumb so it can become the only language supported by flutter) is vendor lock in.


Are you concerned about being locked into Dart as a developer?


> It's a waste of human and societal potential

What a ridiculous comment.

Productivity apps are the most important apps because they enable others to do great work. It's impossible to collaborate (especially hybrid/remote) without them and when they work poorly they drag good teams down.


> What a ridiculous comment.

Only topped by a ridiculous english comprehension.

Indeed productivity apps are essential for collaboration specially in async and non-local environments. They should be invested in because there's a lot of uncovered potential there still regarding modes of information organisation.

What does that have to do with the fact that to deliver that consistently and performantly across different computing platforms you need to produce N variations of your product?


I don't think your comment was particularly clear.

Flow is not just a todo-app. It's a full project management app similar to Jira and I don't think they would've saved a ton of money not building an iOS or Android app.


I have nothing against todo apps, on the contrary! In fact any improvement in that area/tools can easily affect thousands/millions of people, because organising work/projects/collaboration is needed everywhere.

What I would wish for was that to develop things that don't require native functionality (e.g. only need storage/ram/connection) that there was a unified, consistent, performant, API that allowed one to program against that, and have the owners of the OS provide that API, in this case the browser would be the closest (and many issues people have with it against native apps are solvable).

While it's almost sure that the money and time spent on providing native versions wasn't the sole reason, it just adds a gigantic overhead in many cases - as mentioned in the tweet, it's either time you have your devs dedicate to it instead of all other things, it's the continuous maintenance costs as now instead of having a source of bugs you have N, and new features have to be implemented across N, it's the limitations that it imposes in how and what can be implemented (it has to basically work to the lowest common denominator), the hoops you might need to jump to bring it on pair, and then if you use "wrappers" many times their performance is bad. And if you're using a tool to simplify your work you don't want to be wasting time on every interaction you have with it.

Or you can hire a team to do it for you, which brings costs and still management overhead and dependency or you can choose to develop only for web, but this then comes with the issues of not being normalised well enough.

Have an WebGL interface that scrolls a container and needs to interact with mouse/motion events? Well, outside of Chrome it's going to suck. And etc...

As a consequence, given two ideas that are similar, those with much bigger pockets can outrun the UX and quality of their "adversary" products easily. In the process wasting thousands of hours and $ on just replicating the same thing across platforms.

This was my main point!


This is a pretty hypocrite thread. The guy is successful, has a lot of money. He starts a company thinking that because he has the best product he is going to win the market. Things don't go as planned. Blame the VC and competitors for it.

First, his company was VC founded. He poured 10M in the company that's very close to having raised 10M

Second, you dont go out of business because someone created a competing offers

Last, he his a successful guy, it is difficult to go openly on Twitter and says "I failed because we were not good enough" it is easier to follow the trend of our time saying that VC is evil.


Hm, my conclusion of the thread was less “VC is evil” and more like “We underestimated the impact VC had and overestimated the impact “organic growth” would have for us.” What’s hypocritical about that?


I read it as a pretty reflective thread of him admitting he was multiple strategic steps behind Asana at pretty much every stage of growth because of his own ignorance about where VC money can actually be useful.


He claims they were running a bootstrapped startup, but they were never profitable. So they were funded, just not funded well.


This is an important comment. When a company gets to profitability, the worst-case outcome is the company stops growing, it loses momentum and closes with a tiny payout. There is always an option to do that. By not aiming for profitability first he bet on spending big in pursuit of growth instead, which is the VC model. He was effectively his own VC in this case.

Someone with less funding availability might well write the tweet thread about how they were crushed by the person who had $10m of their own money to spend.

I do think it’s great that he wrote this up and I have to say, as a bootstrapped (profitable) startup founder in a space with heavily VC backed competitors, it made me pause.


This in an incredibly valuable Twitter thread. As an entrepreneur, you have to decide very carefully about the market you want to conquer. Or, in the words of Sun Tzu: "If a battle can't be won, don't fight it." - For companies backed by big VCs, money basically does not exist. They can spend ten dollars to earn one as long as the story of building a leading brand can be told. Vanity metrics play an important role here. The primary target is to go public to cash out. There is _no way_ winning against such a player if you're not backed the same way.


Take this charitably, but I’m glad to hear someone fucked up their software business bigger than I did.


One glaring issue I see, is the 12 years part. And I don't mean in the simple banging-head-against-wall versus know-when-to-quit sense. Even if the sunk cost was zero, once decades start passing, even the slowest and most bureaucratic dinosaur will be able to catch up with you. In my opinion he needed to move on to some new adjacent product or aspect of the market where he could have an advantage again. Not just piling on features that those competitors could do too.


The things about SaaS is that it is as easy to do for others as it is for us. So if we think that something was sort of easy to build and get loads of money for, then competition is inevitable. There are a few ways to survive the competition:

1) Do something that you have special experience or skills in so that not so many people could copy it if they wanted to. Maybe you know lots about the financial markets or aerospace.

2) Go into a specific market that the General Practioners (especially the Unicorns) are less likely to target. For example, UK Healthcare or German Finance, which could be super-targetted at the niche and provide a lot of value for money, even over a relatively small number of customers. Some finance software houses only have double-digit customer numbers but very high markups.

3) In some cases, solve a problem in a novel way that once people get hooked into it, they cannot directly compare you to the competition. If it looks just like a t-card task system, it is easy to compare. If you give the entities different names and have a workflow that solves the problem in a roundabout way, maybe people won't know that they could swap to Asana.

4) Add lots of integrations and make it really sticky with people's other systems. Many customers today will literally choose a supplier based on whether they offer a Salesforce integration or not (even if it doesn't make logical sense for your product!)

5) If you really want to go big, you have to go early, well and fast! AOL and Yahoo might have gone early but not necessarily well or fast. AirBnB went early, fairly well and fast. It sounds like Flow did not necessarily go very early, went fairly well but could not go fast enough.


Once you get past the David vs Goliath ethos, we are left with "I had invested millions of dollars, without even realizing it" and "my CFO was freaking out" was a big part of the problem. What was omitted was, "and I didn't listen to any of them". I also liked how everyone else was drinking the cool-aid but thinking you're going to make a billion dollars on your to-do app isn't.


Well, it's worth noting that there are a ton of hugely successful project management platforms launched after Asana - Bitrix24, Monday, Clickup, Notion, Airtable and so on. Asana did not crush other players.


Since neither Flow nor Asana are innovative ideas for a startup, they basically ended up competing in a zero-sum game trying to out spend each other in marketing and features.

That is why innovation and finding a niche with loyal users are key ingredients for startups. But in SV, you can keep pumping money onto a mediocre ideas/products round after round and increase it's valuation. Basically drying out the competition (our author in this case) and pushing their products down the market throat using sheer force powered by marketing and developers VC supplied capital.


Taking this at face value, it feels like a VC market is not a particularly efficient way to produce value for the general public. Success is way more contingent on how long a company can shoulder losses for rather than how good the product is. Similar situations seem to be happening with rideshare apps: although there is some innovation, right now it feels like they're all hanging on by their nails and hoping for their competitors to run out of money.


He still didn't learn how to communicate succinctly.


I loved to read the thread, but your comment is spot on. Marketing is not his piece of cake.

I would read the same story as: we spent only 10 million dollars over more than a decade to build a better product than Asana, who is bleeding out money and spent hundreds of millions. In the meantime we have also built great customer satisfaction, very good retention and an amazing distributed team years before the WFH hype.


He told a story. Imagine if Star Wars was just summarized as, "Some guys with laser swords fought a bunch and the good guys won in the end".


I had this pop up in my twitter feed independently.

When I went to go look at it's basically a lite version of something like Jira. Far less featured, maybe not targetted as much at software development.

In competitive sales analyses (easy to find) it completely loses... far less features, integration, etc.. and yet quite a bit more expensive. It's base starting price is quite expensive compared to the competition (not just vs Jira).

It seems like the product had a lot of things they could have done to improve it and they just didn't execute very well.

It's a weird product area.. I kind of hate Jira, but I also kind of hate every other project management software I've ever used.

It seems like they came at it from the "we're expert UI/designer" types as opposed to project management/development experts. Maybe they cut down their features and kept it too simple out of a sense of design purity.

It also sounds like he had success in other areas that were easier and didn't stress his business management skills, but this product did and he didn't listen to his business partners or seek out better managers to help make the right decision. Perhaps he was blinded by his previous success.

It's an interesting read. As an engineer I would have been very very skeptical to join a team working in this area because my gut feeling would just be bad due to how many products have failed or been mediocre in this product space.


>competitive sales analyses (easy to find) it completely loses.

Could you please share? I wasn't able to find anything.


Just search "Jira vs Flow" or "Jira vs Asana" or "Flow vs Asana" and you will find a bunch of sites showing competitive breakdowns.

Flow looks way more tailored towards simple SMBs that you would think would have trouble dealing with something like Jira.

But Jira starts with a Free tier for < 10 users and the next level up is $7/month/user.

Flow starts at $53/month for 6 users. With far fewer features. The next tier is just "only by quote".

Flow is the least featured but it's starting price point seems to be the most expensive.


I disliked Jira until I tried any alternatives. I think it's the combination of (1) the tool workflow (either rigid or the wrong template) doesn't match how you work , and (2) way too much process layered on by more and more cooks.


And now even Asana isn't doing that awesome.

Good this story is shared. Running a startup is one of the toughest things you can do, it requires balls of steel and virtues of Buddha/Hitler and still luck will be the deciding factor. You need to do everything right a single misstep can kill your company.


Wow, just looked at their earnings reports and it doesn't look like they've ever turned a profit either. In fact, they are losing over a hundred million a year. No wonder this dude couldn't compete.


WTF? how is this comment downvoted

EDIT: Oh got it... surprise surprise guess which company's tech team is hitting refresh every five seconds on this page


How one can see the up/down votes in this place?


you can't, but when a comment goes below zero it turns gray.


That $10M wasn't wasted for me as an Asana/Jira user because they forced Asana and Jira to get better.


I want to point out that Asana CEO meeting OP for coffee and basically starting a competition is one of the healthiest and most amazing things I have ever read/heard about. The market was a game. Amazing, and something you only read about happening in startup land.


One thing 37Signals is very good at is limiting the scope of what they're trying to accomplish and market those limitations as a strength. Instead of doing that, it sounds like they tried to compete directly on features against competition that could outspend them.


One aspect of bootstrapping I imagine is the "fail-fast" principle. If you can't compete while spending less than your revenue you should take it as a sign that your current business is not "boot-strappable".

Doesn't sound like he bootstrapped.


For people tired of scrolling Twitter thread here is the link in blog format.

https://mythreadreader.com/awilkinson/1376985854229504007


Reading the same text on twitter is unbearable.


You are comparing your profit versus Asana revenue. After all these years, does Asana is profitable? If it doesn't I think they didn't win either. You both have lose.


We're in a competitive SAAS market, raised no money, don't do marketing, work exclusively on the product, and offer good customer support. Our competitors raised > 100 million dollars, and daily we get customers from them. And we are not named Jason or DHH; our blog posts usually get 100 views...

We just reached $1M in ARR.

Yes, it is all possible; their problem is they tried to compete directly on the same terms, you can't outspend a VC-backed business.


Have you written about your early days? I’d love to hear more. I’m at the very beginning of what I hope is a similar journey with my side project.


This thread is missing a link to the actual product, so here it is: https://www.getflow.com/

They launched a new version last fall: https://www.getflow.com/blog/flow-x-is-launching-in-septembe...


Wow it actually looks really nice, if this came out today it would be a hit... but because it has been around for a while it it has the sheen of being boring on it.

Otherwise I cannot tell why it is not more succrsssful


I feel this guys pain. I too did the same thing with a startup, only at a much smaller scale.

Failure, sometimes, really is the only way to learn.


As the founder of Toodledo this really resonated with me. When I launched my todo site it was just me and Remember the Milk. I had one of the first iOS todo apps in the store and I rode that wave. Grew to 7 employees. Then came Reminders and Flow and Asana and Wonderlist and Things and Todoist and ...


But you were the one with the pluggable API if I recall, and so able to work across platforms and with other apps, while also being rich enough to implement a “GTD” type system. I could tie a number of things to you, making me a happy subscriber!

Thinking back, I think the thing that eventually pushed a shift was not the ones you mention, but Trello, thanks to Kanban w/ card flipping mechanic, along with “team” boards w/ permissions.


Notably, Basecamp is famously anti-native app, and pro-web/hybrid experiences wherever possible


Thanks for sharing!


That thread made it click for me what his, my own, and other company problems were. We usually don't know why a product is a product, because acknowledging it would make its effect disappear. Thank you, I get this perfectly now.


1. Don't compete with deep pocketed VCs.

2. Keep costs minimal, offshore if required.



A great and painfully honest and upfront read.


i got an impression this is just a 'submarine' promo for whatever is their current project,

which is not surprisingly related to bitcoin, but could have been SEO or some other marketing heavy domain.

I see these stories periodically "7 mistakes I made while building a startup" and it always turns out to be just an elaborate ad.


> burning over $150,000 per month ... Flow did about $3MM of ARR

How did they run out of money? Something doesn't add up.


I think burning here means losing $150k per month. So they had $400k in monthly expenses.


Presumably paying salaries, hosting/hardware, and he specifically calls out marketing costs.


5 tweets deep it hits you with the signup or login to read and then it’s tab over


Where did he get $10m without VC money?


He was the founder of a profitable design agency called Metalab https://www.metalab.com/about


As per his Twitter bio, he's the co-founder of https://www.tinycapital.com/


Some of us (well, “them”) have that kind of money!


What is implied but not specifically stated in his lessons learned is that selling your time while also selling a product is VERY difficult to pull off.

Fried/DHH eventually turned into a product company to focus solely on Basecamp, and I think the Ruby on Rails thing also makes their case entirely off the table for basis of comparison for why your consulting business could generate some quick and easy extra cash from a saas product build by the billable employees on the bench.


I can't stand using Asana. The search sucks, the organization sucks, you can never find what you're looking for, the UI has some of the worst SNR I've ever see. I hate that I'm forced to use it, and that it's being used as a horrible ticket management platform.

VC money is a hell of a drug.


It seems to be very good for creating mediocre products and/or toxic products with a high social cost.

Because it's a meta-market. You don't win by providing a quality product/service, you win by persuading VCs and investors to give you money - which is a completely different game.

You "win" as long as that continues. The product/service itself is just a shop front that gets investor money through the door.


Also it's so damn slow. Feels like it downloads a movie in bluray quality when I log in.

I've had so many friends complain the same, why don't they fix it and why do companies force us to use this :(


Bloat is one unfortunate side effect of trying to monopolize a market.

80% of users need the core functionality, but catering to the other 20% creates a monster.


Agreed, had to use Asana on a recent project for the most basic issue tracking. Used it for a few months, really didn't like it at all, found the interface confusing and never got used to it. I think trello is the best I've seen, but of course has issues of its own and needs add-ons to do some of the things I want.


Also they outsourced many integrations to third party developers.

So Asana is terrible for software development as it lacks the deep integration for Github/Gitlab etc that you get with Jira. And all the third party offers is a basic sync between Asana Tasks and Github Issues.


remember when they started, they had a few videos showing there amazing new framework "luna" that was going to change the world?

https://blog.asana.com/2010/02/lunascript-our-in-house-langu...


Asana feels like some designer got drunk on the minimalism kool aid. I just want a ticket id, not a 128 bit guid.


What does minimalist design have to do with database sequence number generation?

Or am I incorrectly assuming that once sentence is intended to flow on from the previous one and they are just two separate random thoughts.

I do agree with your assessment that the UI is a bit too slick for its own good. Slightly annoying in fact.


As a general observation Asana feels intentionally stripped of features for aesthetic reasons, rather than any actual process reasons.

A specific casualty of this “Remove all the things!” approach is ticket ids. Perhaps the logic was that since you have a title and a link, that’s all you need. Well, no. Titles can change, but ticket numbers tend to be fixed within teams. Without a short ticket id, it is very difficult to coordinate say git branches/prs to tickets. Whereas if you had a short stable id, you can just name your branch WEB-1234, and know it is associated with ticket WEB-1234. Short stable ids also make it easy to make summary tables, text docs, sms messages, paper notes, and other forms of communication refer to the ticket without constantly copying some excruciatingly long URL, or meaningless hexcode.

In my brief experience with Asana, I found it to prioritize form over function. Or as a friend of mine said when comparing to Jira to Asana, “Jira may be a hard configure mess, but at least it knows what problems it’s trying to solve, even if it’s not particularly great at it. Asana doesn’t know what it is.”


Ive never even heard of asana or flow


Wait. What is a "to-do list app"? What else do you need to maintain a to-do list apart from a single text file and a text editor?


It depends how complex you want to get. First, you have tasks and there a few different types, time-bound (do by friday) and ongoing (do every friday), then your tasks have to prioritized, then they have tobe linked to related tasks, often you need to manage dependencies, which can basically be a directed acyclical graph (or worse, a directed cyclical graph). Of course you often also want to be able to get various retrospective views on what tasks have or have not been completed.

A sufficiently complex to do list is more like jira than like a text file.


I wonder if the person you're replying to is correct in but in a different way.

Since text files and excel sheets are "outdated", we don't stack them up against bloated webapp that Asana is.

What if, really, excel sheet can just work? Its like we get so wound up in existing status-quo, we get narrow vision and forget the larger domain space for solving problems.

It's like the guy that ran a bunch of ETL jobs on a Macbook Air faster than a Hadoop cluster. KISS is beautiful sometimes and eye-opening. When someone brings up a ridiculous "outdated" idea, I try to keep an open mind. May be... just maybe we're wrong about all this?


How do you use a single Excel sheet across hundreds to tens of thousands of people ?

That's what apps like Flow, Asana, Jira etc are designed for. Not one or two people.



I don't see it as "outdated", but mostly as avoiding unnecessary overkill. But then again, my to-do lists rarely have more than 5 or 6 items, and I don't manage other people, so I'm certainly not the public for such apps. For my use case, text files are perfectly appropriate and I would feel ridiculous using anything else. Like using a tractor to work on my tiny lawn.


It's mentioned in the thread:

> I was a huge to-do list junkie, but back then all of the task apps were either single-player or weird desktop apps with syncing issues.

> I decided to build a shared to-do list app for teams.

Yeah, simple single-player to-do lists are great beginner projects to do to learn CRUD and a frontend framework. But enabling real-time multiplayer ups the ante because now you have to manage state and updates between multiple clients and a server.


Honestly I don't know and I'm amazed by the ability of people to burn cash like this.

Software is weird in that way, because a todo list is just some "mindmap", linguistic, psychological way to organize data. It could be done with paper too, or with some simple process one could teach himself or learn another way.

It's very difficult for me to really grasp and trust this startup economy. The good thing is that it feeds people and that it cycles this whole ecosystem of "throw it away and try again".

I wish software was not so chaotic and uncertain. When you add up agile, funding and decide to stop spending months to think about the app you are going to make, it's a recipe for disaster.

I'm wondering if this whole skewed startup model is somehow profitable for some people, like finance, etc. Sometimes it really feels like organized scamming.


The problem is when you want to collaborate with other people who may be distributed or asynchronous.

>I wish software was not so chaotic and uncertain.

The best thing you can do is learn how to make decisions under uncertainty.

If you don't want that, go work in a regulated industry at an established company. That work will be planned and predictable as much as it can be.


Classic HN!


Typically just a CRUD app that manages TODO items. Type in an item, store in some some data store, provide deletes, edits, etc. It was (is?) a pretty common app to build for someone trying to learn how to build CRUD applications in [insert stack].


a to-do list app is a sure fire way to get on the front page of HN


Either that or building a lisp parser in rust

I kid, I kid - these projects are all good because they are simple and can be achieved by “hackers”, using the original meaning of the term as curious programmers

Unfortunately the big projects are hard to do, but that’s why they can charge money for them


> Either that or building a lisp parser in rust

Hmmm, what about a rust parser in lisp? Now, that would be front-page worthy!


Indeed! That would be amazing..

Very tempting, must resist going down that rabbit hole :-)


I kept reading for 3 minutes and still couldn't get a grasp on what kind of "agency" he was talking about. Real estate agency? Literary agency? Travel agency? Can anyone who finished reading tell me?


Yes I didn't get this either.

My guess is he meant software contracting or something related, making software to order rather than making their own products.

And presumably he could've switched to a SaaS business model without leaping into a completely different market. Instead he should have stuck with some B2B area he was already working in and knew well. That was the high-risk part in my opinion.


Metalab is a software/design development agency. 37Signals (now Basecamp) started as the same kind of agency.


He has a link in his twitter bio - it's software agency. It's also possible to get it from context - he mentions grabbing devs from it to build Flow.


This is just a techbro(well, more like middle aged) story, full of all the usual ingredients:Narcissism, overestimation of their impact on society, virtue-signaling, extrapolating personal experiences to the whole society. Give me better a story about a Turkish butcher(including he calling names to the mom of his competition) any day of the week.


I don't know the guy. And I might be wrong, but I think this guy is selfish, arrogant and the I only listen to myself type. Not a great formula for a running a startup.


A person like that would blame other people of the failure. Clearly this person owns up to it.


Weird I got none of that.


Well, he did just spelled out why his entire business failed on Twitter. So I don't know how you conclude that he is being arrogant.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: