I sold TinyPilot, my first successful business
mtlynch.io
mtlynch.io
Great write up overall but as someone who has bought a couple of businesses, it is not just the seller who loses out if a buyer backs out. The buyer can also spend lot of time/money vetting the business including legal fee etc during due diligence. On a recent deal, I spent over 10k as a buyer until we finalized the APA. If the deal fell through after that (possible), I would have lost that money and the months of time of course.
In this case, the buyer probably spent more on the deal than I did, but that wouldn't really comfort me if they had backed out last minute.
If I were selling to a direct competitor, I would have had to be a lot more protective about trade secrets. I could make them sign contracts saying that they wouldn't use the information against me if the deal falls through, but I'd have a hard time enforcing it, and the cost of suing them would probably be higher than anything I'd hope to extract for a contract breach.
Here it does not. The goal of a sale is to make money, if there was no deal, than it was no success by that metric.
Broker commission: $88,900
Legal fees: $18,297
Most people don't realize how much get eaten up in deal/closing costs.
Congrats to him for the successful exit. Reading his blog post over the years should be eye opening to anyone just how hard starting/running a business is.
In this case you have a seller with no experience, and ultimately a buyer with no experience. In either situation a broker is valuable. In the case of both its an enormously important moderator who can keep both parties on track to a successful conclusion.
Book learning in such cases only gets you so far.
Most of this value is in the soft areas, like how hard to push and how to manage expectations with the buyer.
One of the core skills of being a founder is learning when to delegate.
Books tell you what they did in their situation.
They don't tell you what you should do in your situation.
This isn’t AdWords where your landing page sucks, and you can just quickly pivot and get some more traffic. There’s only so many serious buyers of a business like this why risk squandering any of them.
It’s a 15% commission to massively reduce the risk to seller and time commitment. Assuming broker is competent they’re going to get the deal done right. Seems well worth it to me.
He also mentioned having a kid soon, so why drag it out? Dude already made almost a cool million what’s 90k in grand scheme of things to get deal done and money in bank?
Skipping a broker in an extremely high friction transaction is penny wise pound foolish. If residential real estate hasn’t figured out yet how to cut out brokers, business acquisitions definitely have not
For me, the value of the broker was just someone to guide me through the whole process because I'd never done it before. It was good to have someone whose interests were pretty aligned with mine and who answered questions quickly and thoroughly without me having to pay by the hour.
Another big part of the value was finding the buyer, as I probably couldn't do that on my own. I could have thrown a hail mary and listed the business for sale on the TinyPilot website and my blog, but the person who ultimately acquired the business definitely wouldn't have found me that way.
Having gone through it, I would consider foregoing a broker in the future if I had some way of finding my own buyer, but I have no regrets about the broker for this sale.
In real estate brokers typically get around 2.5%, so even if they find a buyer that is ready to pay 100K more, it's only $2500 extra (so probably $1500 after taxes). It's not a lot of money esp if they can invest that time and energy in parallelizing other deals. Time is of the essence in such transactions.
And then of course a Buyer side agent's incentives are aligned exactly opposite to the buyer, but even if they would want to push for an early transaction above anything else.
But even aside from that, they make more money the larger the deal, so they work to get a good price. They also feel the pain if the deal falls through, so they have incentive to keep the deal alive and both sides happy.
The broker isn't perfectly aligned with me, but of everyone else involved like lawyers, accountants, and the buyer, the broker is by far the most aligned with my interests.
Perhaps so, but you can buy a lot of legal hours with $89,000...
The marketplace connection seems the part that gives the most value.
If you're already knowledgable, experience, and well-networked, there is less value.
If you're not...well, have you ever sold a business?
Compare the serverCo offer with the final sale. The difference is the value added by the middle man.
That it is, but here Michael is explicit that most of the pain [0] was from doing hardware, software, and e-commerce all at once. Surely, there are other relatively easier businesses in tech?
[0] Reminds me of Avery Pennarun's A profitable, growing, useful, legal, well-loved... failure (2012), https://news.ycombinator.com/item?id=3754531
It's always good to get these reality checks from time to time.
They do if they've sold a house :)
OTOH, it's possible the work a broker does doesn't scale down with small deals, and he wasn't willing to work for less. Honestly, I'd rather accept $50k less and leave the broker out of it if that's the case.
That's not how it works, though. Without a broker you may not find a buyer at all. If you do, you might not know how to price the business. Even if you could find the same buyer at the same price, they might be uncomfortable dealing with someone who doesn't know how to navigate the sale of a business and end up backing out.
For some very small web businesses, you can usually get away with a broker. Nobody is dropping $90K on a broker to sell their $50K side project on one of those business buy/sell websites, obviously. However, once you get into the territory of niche hardware businesses, it's very helpful to have someone on your side who can help navigate the situation.
> Without a broker you may not find a buyer at all.
This only applies if your business isn't super profitable and you're looking to get rid of it.
If you have a wildly successful business and you enjoy running it, you don't need to find a buyer, you can wait until one comes along.
Even if the business is wildly successful, a broker is going to help you shop for competitive offers and get the best price on the open market.
And if your business is wildly successful and you enjoy running it, hiring a broker will let you focus more on running the business. Trying to learn how to sell a business is going to take you away from this hypothetical wildly profitable business that you enjoy running, which is neither good for your profits nor your enjoyment.
Exactly. I once went with a broker (not the US, so lots of options not to) because the broker (despite his fees) will come out on top of other options. I guess, in that case, the broker did earn his commission fully.
I think most people have a negative idea of brokers because in some jurisdiction they can make it difficult for P2P to happen. This makes the brokers a defacto monopoly on the market; and they usually all follow each other on the rates.
The buyers on the other side are professional buyers. They get promoted by buying great companies for as little as possible. Most founders will only ever sell one business, if that. Buyers don’t care about their reputation with you, they do care about their reputation with a banker they may interact with multiple times.
I feel this. Being a founder is hard, and raising kids is very hard. I was a little bit sad when I saw this headline show up on HN, because I've been following you for years, ever since the beginning. I always admired your hardware business, and the yearly retros, and I was sad to see it all end. But after reading the post, I do understand. Congrats on the sale, and on the child. I hope you can find some relief from the stress of being a founder, and I hope your new family will bring you as much joy as my 2 kids have (with +1 on the way for me as well).
Would you build another hardware business, or are you going to try for a software business next?
>Would you build another hardware business, or are you going to try for a software business next?
No, I don't think I'd try hardware again.
It's a shame because I learned so much about hardware and selling a physical product, so I do have an edge if I wanted to do something else that involves both hardware and software, but bootstrapping hardware from zero is just so risky. I caught a lot of lucky breaks for it to work out with TinyPilot; but simple one-time events like a manufacturing error or a lost shipments could have sunk the entire company in the first two years.
I'd like to try another educational product. I released a blogging course at the end of 2020. I didn't have time to promote it much because TinyPilot swallowed all my time, but it still has made $10k of profit. I'd like to try something like that again.
But I would also like to build some kind of pure software business if I land on the right idea and the right market for it.
Seems he exited the Google grind about 6 years ago. I guess salaries were between $180-250K during that time frame.
Google grinder for 6 years: $1,080,000-$1,500,000
Entrepreneur: “$920k over four years (with sale)”, but doesn’t include the 2 bombs he had prior to successful exit.
(None of these figures taken into account taxes or benefits)
Seems like he took an L, but in the end at least he didn’t end up in the salt mines. The education of learning as you go is truly underrated though.
However, he kept trying again before job hunting or returning to big tech. This tells me the monetary factor was smaller than something else.
You mention the underrated educational experience in your last sentence, but there's so much more than that. He was probably never worried about his autonomy, being laid off, working with (or for) people he didn't want to work with, corporate politics, or anything else corporate bureaucracy introduces. This likely freed up his brain to be more creative and actually be used to 100% of its capacity.
I believe the obsession with TC in tech is highly problematic and there are a lot of talented folks optimizing for promotions via internal politics, rather than solving real problems for real people.
I also wish healthcare wasn't tied to employment, but that's a post for another time.
- get health insurance from your employer
- get covered by insurance of your spouse
- buy insurance from your own pocket
- don't buy insurance and risk loosing a lot of money in case of health emergency
It is expensive, but not insanely so. Family of four, living in Oregon, non-smoking, our premium is ~$1k/month. Granted it’s a high deductible ($12k), but it protects us from the kind of catastrophic emergency that could wipe out our savings. We are fortunate to not have expensive, ongoing conditions.
Granted, $12k/year is a lot for less affluent families, but government subsidies can bring the premium down considerably if your income is low enough.
And while the coverage is far inferior to my old FAANG luxury health plans, it still works well enough.
Family of two, living in the EU, non-smoking, our premium is ~€0/month. Granted it's a €0 deductible, but it protects us from the kind of catastrophic emergency that could wipe out our savings.
Also, turns out there is a deductible if you don't want to pay a fortune for self-insurance.
Well, at least there is in Germany. And the premium ends up being ~$1k/month.
US health insurance is crappy in many ways, but EU health insurance isn't a magic fairy unicorn either.
That being said, the average professor salary in Germany is around 84,000 Euro, so it's closer to $560 a month.
You also ignore that after the high deductible is paid in the US you're still paying co-insurance and co-pays. (I love how US health insurers describe these as "your contribution" to your healthcare costs, as if you weren't already paying premiums and deductibles, but the magical insurance fairy is...).
Now you can argue that health care costs less on a per basis, or even per productive taxpayer basis, but that's a different matter. Maybe the US could recreate your healthcare COST system, but it doesn't have to also take on your payment system (tax instead of direct)
In the next-door Czech Republic, it's even mandatory to buy health insurance if you're unemployed, unless you go and register with the 'Ministry of Labour' (which requires you to spend inordinate amounts of time jumping through insane bureaucratic hoops, and is ultimately time limited). Consider for a moment the effect of these laws on people with mental health issues, the homeless, and itinerant minorities like the Roma.
Had a tough year out on the streets, but now getting back up on your feet? Congratulations, here's your back-debt for the 'public' health insurance you failed to purchase, you criminal. Want to take a few months off between jobs? Gotta go down, in person, to your local health insurance office to purchase yourself some public health insurance. Want to start a new business, but haven't made a profit yet? No worries, here's your 'minimum rate' of mandatory health insurance - prepare to shell out several thousand euros a year and spend time every quarter filing paperwork with the government health insurance bureau.
There are excellent public health care systems out there (e.g. Australia, and probably the Nordic countries) but much of continental Europe has truly terrible ones. And that's before you even discuss the difficulty of securing a doctor, or the actual quality of medical care received.
What makes the issue so terribly entrenched, apart from our peculiar association between health insurance and employment, is how utterly opaque everything about healthcare costs are. Both systemically, in that it’s much more difficult to craft policy, as well as practically, in that for many (most?) individuals it can be impossible to know in advance how much any given course of treatment might cost, and no opportunities for “shopping around” or price comparisons.
For most Americans, and companies, I think our healthcare system is essentially a very short-sighted and inefficient tax. For others—people in situations such as mine, and very probably for a meaningful portion of the ~30M-50M Americans still uninsured/underinsured—it’s an unreasonably high burden that accounts for so much more than just the money spent, and a constant specter clouding their lives in uncertainty and risks. For the self-employed and entrepreneurs, it’s a crucial consideration that defies simple calculations, and just by the averages will have prevented a significant number of potential endeavors from ever leaving the initial planning stage and/or denied startups access to otherwise valuable talent. We should be pursuing bigger picture policies, and while I’m not arguing that the federal government is a good answer to everything, it’s clearly the answer to this kind of management of universal needs with clear societal benefits.
Sorry for the lengthy comment, my hamster brain was apparently restless. For anyone who did, thanks for taking the time to read.
It’s a tragedy that universal healthcare and tougher regulation against unhealthy food is still not achieved.
If you don’t mind me asking, what advice would you share to deal with such diagnosis and making it through?
In other words, buy it on the market, like the GP said.
With ACA subsidies, you'll never have to pay more than 8.5% of your income in premiums for a mid-range plan. And that's in the worst case -- it'll typically be less (if you make less, you'll qualify for larger subsidies or free healthcare, and if you make more, the proportional cost of insurance premiums will be lower).
I was paying <~$300/mo on that plan. I now pay over $700 on a directly-acquired plan. So that's problem one.
The bigger problem, IMHO, is that quitting means changing your insurance, which isn't just a financial change. It can also end up requiring you to change medical providers, which is no small deal!
This adds massive friction and stress to the act of quitting/changing jobs, a major source of leverage on the employer's side.
And that's to say nothing of the fact that it's effectively an overly burdensome tax on new entrepreneurs. It's a reinforcement of the concept, "you have to have money to make money".
JMHO on the topic; for reference, I'm a new entrepreneur very early in the process. I'm just lucky enough to be able to risk the attempt.
This is relatively small pickles. Your employer paying a few hundred, maybe a thousand dollars on your behalf if you are old, per month. You can buy the same on healthcare.gov
They only things that I expect to change are total premium, deductible, out of pocket expense, copays, and network. Otherwise, the coverage itself is dictated by the same law.
I can see right now that the BCBS gold HSA premium from my employer (total, including employer portion) is the same as the BCBS gold HSA available on healthcare.gov.
I used a hospital for a new baby birth. Because my plan was on-market, the provider doing delievery didn't accept it (Despite it being on the in-network list). So I got to be stuck with a ~13k bill. Was super cool.
This was back ~8 years ago, so perhaps it's all been sorted.. but I have such a bitter taste in my mouth after having to pay this out of pocket, I won't touch on-market plans with a 100 foot pole.
I now pay $120 for me and my wife for a plan that is better than the $1000 plan through the ACA in Oregon because I'm no longer self employed.
That's almost a 10x difference.
After working for myself for 12 years, hustling consulting and contracting gigs, I'm way happier working for a company that pays well, has great benefits, and I always know I have a paycheck coming.
Look at box 12 code DD on your W-2. That is the total amount you are paying for your health insurance (it’s just that your employer is paying it directly). It’s still counted in the cost of employing you though, obviously.
Assuming you have a silver or gold level plan, total cost to insure you and your wife is the same on healthcare.gov or if your employer buys it.
The only broad savings are if your employer is self insuring and their risk pool (employees and their families) are disproportionately low risk (young and single).
Health insurance has a 2% profit margin, and premiums are tightly regulated by state insurance commissioners. So a wildly different premium indicates a change in coverage, or a change in the underlying risk pool.
For an annual out of pocket maximum of $10k to $18k and age rating factors capping age 64 premiums to 3x age 21, the minimum bronze/silver monthly premium should be somewhere in the ~$400 to $1,200 depending on age. For example, see NJ’s premiums here:
https://www.nj.gov/dobi/division_insurance/ihcseh/ihcrates20...
Where did you get that 2% profit thing? I've been googling for that too and insurance profits and all I can find is that ACA only allows 20% admin costs. Non ACA plans indicate that their profit is ~15-20% of their revenue.
These are all publicly listed health insurers, so all their financials are public. It is not a very profitable business, as pretty much all insurance business.
"More freedom" sounds amazing right up until the point you remember that you aren't getting a salary every month but still need to pay rent and buy food and stuff...
“…it’s where I buy my food, and most of my stuff.” ~ Ron Swanson
Some people like the FIRE approach
Each month of vesting you can absolutely move shares out of GOOG (minor restrictions on insider trading, but regardless of your level you can set up automated strategies), and most employees probably should given the correlated risk between Google taking a nosedive and employees losing their jobs.
The vesting is unrelated. With no promotions or raises the 4yr bump is uninteresting. With those, they match your TC, but they happen _during_ those interesting career events. They describe some subset of your compensation for the next 4yrs. You can still invest them however you choose. You can still leave at any time (unless you believe that in your personal case the promotion train is better than the job-hopping train). You could have just as easily job-hopped and invested the surplus in GOOG.
> should have stayed at Google
You should feel privileged that you've never been "stuck" at a job that you don't find personally interesting and stimulating. Wasting your life to see the number in your bank account increase.
welcome to the world of capital gains and QSBS!
I suppose if he can parlay this outcome into an L7 role at Google, E.G, then in another 6 years he would break even (compared to staying at L5)
Everything has non-monetary costs associated with it. Chasing L7 means more politicking and less actually making things if I understand it correctly. Maybe that's invigorating to some people (am convinced people get off on fighting over artificially scarce resources), but I don't understand them one bit.
According to Levels.fyi - and my personal contacts - https://www.levels.fyi/t/software-engineer/locations/san-fra... Median TC - something just-south of $300k TC (not salary) is more likely.
Furthermore, the higher you go, a greater proportion comes as RSUs, not cash - even if it was $400k TC, a good chunk of that is money you literally can't spend for a couple of years.
But you're right that the TC of the average Google (or even FAANG) job has become somewhat mythical on the internet lately. I'm in a Slack where people have been asking about offers and TC for a long time. There are a lot of disappointed people who think they're walking into $500-600K job offers only to realize that they're "only" getting $300K from multiple companies.
Honestly quite frustrating, given that $300K with excellent benefits is an incredible job offer, but these people have been primed for disappointment due to unrealistic expectations.
I get being disappointed about $300k. Many companies are offering that for non-seniors (Google L4 equivalent) now. I distinctly recall TikTok offering well north of $300k cash for just 2 YoE a few years ago, and Meta offers have risen similarly.
I'd also be pretty sad if I interviewed elsewhere as an L5 and got offered less than what I made as a 2 YoE L4 (and many years ago, prior to inflation at that.)
$500k for senior (L5) is kind of the expected offer these days. If you are interviewing at FAANG you are doing a huge disservice by using Levels and settling for $300k when you can pretty easily get $500k.
I joined a FAANG a couple years ago at $550k TC or so at 4 YoE for senior. It's grown to about $950k due to stock growth. I'm not sure what you mean by "I can't touch the money for a few years," I get a portion of it every few months.
Vesting cadence is typically quarterly or monthly.
Refreshers only being you up to the "standard" target TC (as defined by the company). So after their initial grant fully vests, most engineers will see their comp drop by a couple hundred k, even with refreshers.
This is a huge contributing factor to the low retention at big tech - it's why most people leave by the time their initial grant runs out.
...I never knew I could feel so insecure about my career.
May I contact you directly for some advice and questions?
There is a very high probability he gets offered a CEO position at some company that is looking for an exit path. That type of thing will usually come with a nice salary and a minimum 7 figure payout on exit.
As of this specific date, his bank account might be slightly lower than had he stayed at Google. 5 years from now it will almost certainly be significantly higher, and if it is not, it is more likely because he chose a path that he preferred.
Also 100% sure that his next business will be many times more successful than Tiny Pilot.
That’s the hint you aren’t comparing like kinds.
It also appears you haven’t met anyone truly wealthy, because I can tell you they don’t share your point of view
I actually thought it was a big W for him when I saw this post. But I guess, if you consider the opportunity cost of Google employment, it's a financial L.
He learned and experienced so much more than he could have at Google. He actually built and sold a profitable company.
What’s life if we reduce it to how much money we could have earned?
Your numbers come out to $250K/year at Google and $230K/year as an entrepreneur. That's a small price to pay for this kind of real-world education.
There's no guarantee his next business will be successful -- but there's no guarantee that those kinds of Google positions are going to keep existing, either.
On the other hand, there's also the chance that his next business could be massively more successful because of lessons learned, and open the way to much higher amounts than he could have been earning at Google were he still there.
Overall, I think it was a massive win for him. And it may have been more beneficial to society than the work he was doing at Google.
I'm happy to answer any questions about this post, the sale process, or my time running TinyPilot.
As a sometime-bootstrapper and having failed at a previous hardware startup, I can relate to many of the emotions you narrated through. It may be too early, but my biggest curiosity is whether you think you will bootstrap something again?
(In my case, after the hardware business and some time off, I found that taking a swing at a “pure software” idea was the right balance for me, after the considerable challenges and occasional joys of building physical things..)
>you think you will bootstrap something again?
Yes, definitely. Not hardware again because I think getting to the scale where you can use external vendors is so difficult and risky that it requires more specialized hardware expertise or VC backing.
I'm going to start with educational products because I liked my brief experience with that and never had time during TinyPilot, but I'd eventually like to build a SaaS that I can grow in a calm, sustainable way.
Thanks for all of the content!
>What do you think is next for you?
Next is either an educational product or a SaaS business I can build either fully solo or with 1-2 teammates in customer support roles.
>Would you do it all over?
No, not knowing what I know now about how difficult it is to succeed in hardware.
I'm grateful that TinyPilot worked, but there's definitely a reason why there are so few bootstrapped hardware companies.
In the first few years, there were so many things that could have clobbered the business, like supply shortages, manufacturing errors, lost shipments, design mistakes. I did a lot of things to mitigate these risks, but a lot of it just came down to being lucky enough to avoid random disasters.
For example, there were definitely times in the business where a critical part could have been lost in shipping, and we would have been dead in the water for months if it went missing or got delayed.
As the business matured, we were able to mitigate those risks better, but I wouldn't want to go through those first two years again unless I had a huge amount of investment or co-founders with more specialized hardware/manufacturing expertise.
I do have to say, I was somewhat surprised by the low multiple on the valuation, but perhaps that's just what the range is for a business of this nature. We're spoiled in the world of software and recurring revenue.
What about the tax implications of the sale. Have you figured out how much of the sale you'll be able to put in your pocket?
>I do have to say, I was somewhat surprised by the low multiple on the valuation, but perhaps that's just what the range is for a business of this nature. We're spoiled in the world of software and recurring revenue.
Yeah, from what I've heard, SaaS businesses sell for a much higher multiple, often selling as a multiple of revenue rather than earnings.
The other thing that's really reduced valuations is interest rates. When interest rates were <1%, private equity was bidding up prices of businesses because it was a decent place to park money, but now that you can get 5.3% from a money market, the additional return from buying a business isn't worth the risk.
>What about the tax implications of the sale. Have you figured out how much of the sale you'll be able to put in your pocket?
Still working out the exact figure with my accountant. His expectation was that I'll keep a pretty large percentage after taxes because of Section 174. I had a large amount of expenses each year in software development from overseas contractors, and with Section 174 changes that went into effect in 2022, I had to amortize them over 15 years.[0] But with the liquidation of the company, I can count those expenses immediately, and they offset the income from the sale. So Section 174 is still a bad deal for software founders, but at least when you liquidate the company, you don't have to wait out the full 15 years of amortized expenses.
> Yeah, from what I've heard, SaaS businesses sell for a much higher multiple, often selling as a multiple of revenue rather than earnings.
Adding some color as I have been through the process of selling a SaaS. Based on what you have written, at that stage of development, multiple on SDE is most common. 2.4x is on the low side for growing SaaS businesses, but this business is hardware with real COGS so the economics and buyers will be different. We also don't know the growth rate of the business, which is important in assigning a multiple.
Additionally, full cash payment at closing is a reasonable ask but can lower the overall sale price. (I don't know if it did in this case.)
Again, congratulations in taking this big step on your journey!
Valuation also depends heavily on components of a particular business, so it's not useful to just use one range. Does the business require the owner to run, or is a management team in place? Does it generate its revenue from 1,000 paying customers or 25? What is churn like? How fast is it growing? And bigger companies tend to have less risk in general than very small companies. FEI guides to a range of 7x - 10x for SaaS > $2m valuation (yes, this is somewhat recursive). The high end of that range is about a third the multiple Microsoft gets, but then again these are small businesses without Microsoft's competitive moats.
FE International[1] and Acquire.com[2] routinely publish excellent valuation guides for SaaS companies. I highly recommend to anyone looking to build or buy a SaaS.
1 - https://feinternational.com/blog/saas-metrics-value-saas-bus...
2 - https://blog.acquire.com/acquire-biannual-acquisition-multip...
I have a question about this part:
> I’d also risk TinyPilot’s sales slipping after so many months being distracted from the business.
What was the time split between the sale process and running day-to-day operations? I know bigger companies have dedicated teams for that, but I'm curious how does it look like in smaller ones.
>What was the time split between the sale process and running day-to-day operations? I know bigger companies have dedicated teams for that, but I'm curious how does it look like in smaller ones.
It varied from week to week, but I was spending 10-25 hours per week on the sale for about five months.
But beyond wall time, the time I spent on the sale was often much more stressful than anything else and left me mentally drained for anything else. The sale involved preparing reports that I'd never created before, and it was extremely important for me not to make any mistakes because I didn't want the buyer to think I'd provided fraudulent information.
I also had to focus much more on short-term results than normal. Part of this was a lack of time to oversee complex projects, but the other part was that it's to my detriment to invest in something that pays off in six months if I sell the company at month three. But I can't run that way forever, so I knew if I tried to run the company like that indefinitely, I'd pay the penalty for always focusing on the short-term at the expense of long-term.
BTW tinypilot.com is down: https://www.isitdownrightnow.com/tinypilot.com.html
I share more about the finances in my annual review posts:
https://mtlynch.io/solo-developer-year-6/#tinypilot-became-2...
What does this mean? How does this work? Shouldn't the buyer take out the loan? Or does it look something like
- tinypilot takes out 600k, which you pocket
- you transfer ownership
- tinypilot is in debt, not the buyer
?
Yes, you're correct. The buyer takes out the loan. This was an asset sale, so assets transferred over but not any debts.
The seller (me) can approach a loan broker and show their financials, and the loan broker can say, "I think it's likely that the SBA would approve this loan, and I can connect you with lenders I think would offer the SBA loan for acquiring this business."
The SBA pre-qualification isn't binding or official, but I guess the loan broker's prediction is accurate enough that the brokerage will list businesses as SBA pre-qualified based on the loan broker's assessment.
My question is, did you change anything on your strategy after you were set on selling? Or the goal was still the same, optimize earnings?
>My question is, did you change anything on your strategy after you were set on selling? Or the goal was still the same, optimize earnings?
Once I started the process of preparing the company for sale, it did change a lot of the strategy.
For example, one of the things I'd been thinking about was how to allow users to purchase recurring subscriptions to their TinyPilot Pro software licenses rather than having to remember to re-purchase every year. But once I started the sales process, that basically became non-viable because it meant I'd be investing thousands in software development, but it wouldn't pay off for another year or so, so it would weaken our profit and loss statements and result in a lower valuation.
There were a lot of things like that, where I had to reject any investment that wouldn't pay off in three months or less.
The other thing that was a bummer about preparing to sell was realizing that everything effectively becomes so much more expensive. Normally, if I want to give someone a $5k bonus, it costs me $5k. When I knew I was about to sell and wanted a 3x multiple, the $5k bonus effectively costs me $20k because it's $5k for the bonus itself and then -$15k for the sale price on a 3x multiple. I could try to argue that the $5k is "discretionary" and therefore not part of SDE, but I think the buyer would be in their right to argue that it's not discretionary, as it sets bonus expectations for future years.
[0] https://mtlynch.io/retrospectives/2023/08/#what-would-make-r...
What if you had moved to outsource the office tasks sooner? Could you have better invested that time and stress elsewhere?
The $100k per year improving hardware? Was that the best way to spend the money? Maybe it would have been better spent on marketing, or investing in reducing production costs with better tooling. Whatever the differential from what you spent was lost at 3x.
Gains often compound over time too. Cost reduction begets cost reduction.
Of course, thinking like this results in madness and moral hazard. I’ve seen companies do crazy, catastrophic things gearing up for sale.
I actually hate the 'flip' side of this. Over the last few years I've lost track of the number of services/providers who think it is just fine to renew annual subscriptions (in some cases $400+) with zero warning of upcoming renewal. It's not the amount of the service that bugs me, but the fact that that isn't a small chunk of change, and even a "Heads up, we'll be billing your card in a week for your annual subscription" email seems too much to ask for, to some.
Why wouldn't you go back to a corporation and try to make a software product there, as an employee?
>Why wouldn't you go back to a corporation and try to make a software product there, as an employee?
Oh, boy. So many reasons!
The thing I really love about running a company of my own is that I don't have to get anyone's approval to do things. If I have an idea, I can work on it, and I don't have to convince anyone or justify the time or money.
I also really love being able to choose my own hours, tools, and schedule. At Google, I found real-time chat distracting and net negative (I get that they bring value to some people, just not me), and I found that most meetings were a waste of time. At TinyPilot, we never used real-time chat, and we were deliberate and efficient about meetings.
At Google, I often found that things that were good for the company or my team or our customers were often not aligned with incentives for me (e.g., it was easier to get promoted for launching a complex feature than for a trivial feature that solved the problem elegantly). With my own small company, it's much easier to align incentives between me, my teammates, and our customers.
I could go on and on, but I guess the underlying issue is that I place a high value on independence and autonomy, and no employer can match what I get on my own.
There are a lot of comments here comparing the financial outcome of your TinyPilot journey Vs what could have been had you remained at Google. Can you share your perspective on that? Are you happy you chose the path you did? If so, why?
Thank you
What a feeling! Congrats.
Also, can you recommend or refer the law firm that you used for the transaction paperwork?
Every successful bootstrapper who I know personally says this.
The ultimate success didn’t happen instantly but the product/market fit was obvious on day one.
And they didn’t iterate their way to that point. They walked away from whatever they were doing before, restarted with a blank sheet a paper, and the new thing immediately resonated.
It's not uncommon to become attached to something.
I've done a lot of B2B tooling and the percentage and risk of failure goes down significantly if you are ruthlessly focused on only solving problems that cost time or money for a business, and being flexible to pivot.
In the B2B space, there are absolutely 100% instant revenue generating needs and solutions. I'm packaging a rather boring one right now. Getting out of the building before writing any code, what so ever is critical.
Building before learning is the cart before the horse and limits shelf life, runway, and ability to learn. One could easily cowboy code or over architect something that prevents you from trying the very thing that will take off because it's too difficult to now implement in your formerly simple codebase of patch work that is now too complex.
They often can get discovered while doing a bout of consulting (aka paid market research and customer discovery), or solving one problem sheds a light on something near by that look subtle but critical and a no brainer.
My personal experience has been people will actually say the phrase "I'm trying to find a way to give you money to use this" when you may show them a prototype. This didn't seem like it actually happened, until it did.
The really honest ones will tell you how much of a role luck played. Most, however, will be quick to take credit for luck's role, often deluding themselves.
Indie-hacking-projects have the benefit of a dignified exit by open-sourcing - a lot of whatever value in any technology/code/designs was created will be preserved - which is a good thing.
It's a shame that oftentimes the "best" outcome for a startup or small software/tech company (where they get acquihired by Apple specifically) is not only their product/service gets shut-down, but all traces of it are systematically removed from the Internet and everything the company's founders do is effectively memory-hole'd - the _Dark Sky_ app is the textbook example of this.
-----
Separately, when a company is small, stable and profitable it isn't a failure as far as the bank is concerned, but not exactly a hyperscaling unicorn - so when the founders/owners want to exit their only real option is to sell-out to either their competition for a managed wind-down-and-migration - or worse: an M&A firm (which will only make a lowball offer, then replace staff eng with a skeleton maintenance crew, and mis-manage the company's products/services for a decade as they hemorrhage-out loyal customers due to declining product quality).
Required reading: https://pivotal.substack.com/p/the-worst-outcome-is-a-medioc...
Wow, I've done technical due diligence for a number of large acquisitions in the past few years (quite a few big enough that most people on HN would have heard of them) and the usual timescales are in the order of a couple of weeks and involve constant and in some cases daily question rounds.
Taking what sounds like months and months of time to handle the process sounds super laid back. Certainly seems far more stressful than anything I've ever seen.
That said, when I talked to Chris at Quiet Light about how it played out with ServerCo, he said it was rare to see strategic acquisitions come together successfully at TinyPilot's scale. He suspected that because I approached ServerCo and acquisitions weren't on their mind at all, it just wasn't a priority for them, so that's why everything went so slowly.
The immense care and diligence you have put into setting this up means the new buyers will have a great launching point with few surprises.
Suffice to say, I would have liked to buy this thing for $600k, but I guess you dropped off my radar for the past year and I didn't realize it was up for sale.
Congrats!
Lock-in or additional hours from Michael wouldn't be needed, the product isn't awfully complex, and I'm sure everything is more documented than you would expect.
What I would have liked to do would be to use the company as a scaffold to develop a larger mid-market offering. I don't like mid-market that much, but they are much more willing to cobble together a better, cheaper solution, rather than just buy whatever package that HP or Dell is offering.
I would also try to sign up and train some government focused resellers to see if we can make progress in that segment. USG is very favorable to small businesses. Control over the supply chain would be a selling point to USG as well.
I would also try to partner with someone like System76 or a similar US manufacturer to make the existing TinyPilot product be a recommended KVM component for their server and small compute offerings.
You have a procurement onboarding process that's just sitting around right now? I've been looking at small, local companies that will be closing due to lack of staff/interest once the owner calls it quits.
Now that we're almost in June, does it feel like the deal happened a million years ago? I know when stressful things happen and time passes, it always feels like it was so long ago even when it was only a month or 2 back.
It was a pleasure having you on a few years ago in the Running in Production podcast: https://runninginproduction.com/podcast/105-tinypilotkvm-let...
>You know it must have been stressful when you were compelled to list your desserts in order.
Haha, that's true, but I also get more excited than the average person about desserts.
>Now that we're almost in June, does it feel like the deal happened a million years ago? I know when stressful things happen and time passes, it always feels like it was so long ago even when it was only a month or 2 back.
Yeah, I hadn't thought about it, but you're right. We're only about six weeks out, but it feels like it happened six months ago.
It was funny during due diligence, I was really stressed out, but I could already tell that I'd look back on it and wonder why it was so stressful. At the end of the day, it's mostly just preparing reports and answering questions.
I think the feeling that's hard to recall is that even though the reports I was creating weren't that complicated, they were stressful because the stakes always felt so high. I was always so worried about making errors and messing up the deal or opening up myself to liability claims after closing.
>It was a pleasure having you on a few years ago in the Running in Production podcast
For me as well! It was fun to chat about TinyPilot's stack and tooling.
https://news.ycombinator.com/item?id=33945360
https://nickjanetakis.com/blog/taking-an-indefinite-break-fr...
Any plans to get back into it?
Currently no plans but I do plan to keep the site up in case things change in the future.
I switched to pikvm which has a little less polish, but you can put together much cheaper. It is even more hackable. For example, connect to 4-port kvms that have RS232 and you can run 4 systems.
tinypilot seems to have some serial or order number that you need to download an update. pikvm seems to reserve some features for hardware they sell (for example, they don't seem to want to support HD resolution, even though you can just move an edid file and support 1080p at less than 60hz).
Both of them are or at least started as raspberry pi kvms with an HDMI-in digitizer (either csi or usb) and hardware hacks to use the usb power port to emulate usb devices (keyboard/mouse/drive)
It may have turned out better for this founder that they didn’t go with FE, but I urge people to not always take a single no for an answer.
And my interest is piqued by the Pizookie Trio, whatever that may be...!
As a side project, I'm currently working with a hardware founder (I'm primarily software) who exited twice before having his "clock cleaned" by overseas competition steal and undercut his patented products on his third venture, which caused him to all but close shop and retire early. He's back I think due to boredom, but I'm seeing just how difficult hardware is. Why do you think hardware doesn't generally have the same level of 'respect' as software? From what I can see, it should be held in much higher regard with a correspondingly higher level of compensation. Is it literally just the current ability for hardware to be reverse-engineered and undercut?
>Why do you think hardware doesn't generally have the same level of 'respect' as software? From what I can see, it should be held in much higher regard with a correspondingly higher level of compensation. Is it literally just the current ability for hardware to be reverse-engineered and undercut?
Oh, my perception of hardware engineers is that they are held in similar regard to software developers.
If they do get less prestige, my guess is that it just comes down to money. The margins on pure software products are so high, so software businesses can afford to pay their software engineers really well. I've heard that typically hardware products sell for about 3-4x their bill of materials, so that eats up a huge chunk of your profits. And then there are all the other costs like storage, shipping, fulfillment that are nearly zero for a software business but significant for a hardware business.
I think for similar reasons, we don't see a lot of successful hardware startups. The VC money is probably flowing much more heavily toward software, so we hear of fewer hardware unicorn companies.
Fresh out of collage? The average appears to be $100k [1], but this would include senior developers too. I think their sense of wealth is warped by Google...
[1] https://www.glassdoor.co.uk/Salaries/us-software-developer-s...
2) Congrats on expecting.
3) Thank you for sharing. I found this insightful, and hopefully throughout the journey you found more fulfillment than you would have had you stayed working at Google making software for such amazing product manager decisions like a $240 smart watch for 8 year olds...
I also own a small bootstrapped company (pure software), and I am now on year 21 (!). Someday, my business will run its course. When it does, I will probably write a blog post about the entire experience of starting and owning it, and if I do I will have been inspired by your posts.
Best wishes.
And 21 years is amazing! With everything that's changed in the last two decades, it's awesome that you've continued adapting and kept competitive all these years later.
You mentioned having trouble working with vendors, was this for ex: having boards assembled? I guess I've probably gotten "lucky" with relatively low scale so far, but been happy with the assorted online PCBA services.
>You mentioned having trouble working with vendors, was this for ex: having boards assembled? I guess I've probably gotten "lucky" with relatively low scale so far, but been happy with the assorted online PCBA services.
Oh, it was a challenge to find vendors for pretty much everything. The cost of switching vendors in hardware is pretty high once you've built a pipeline around a particular vendor, so it's important to have reliable vendors. But we had trouble finding vendors for PCB design, PCB assembly, third-party logistics, and packaging. We eventually found vendors that worked well, but even the good ones would have occasional wrinkles that needed smoothing out.
I wrote previously about how poorly I chose a vendor to do a redesign of TinyPilot's website:
You mentioned that building tinypilot suddenly sparked way more interest than your other attempts. Do you have a plan for your next product? Try out different things and hope you find something that gains similar traction? Or do you have some sort of evaluation criteria for your ideas?
>You mentioned that building tinypilot suddenly sparked way more interest than your other attempts. Do you have a plan for your next product? Try out different things and hope you find something that gains similar traction? Or do you have some sort of evaluation criteria for your ideas?
Yes, the market's reaction to TinyPilot will definitely impact how I evaluate future businesses in the early stages. That said, I don't want to overindex on this one experience, as I know other founders who found product-market fit more gradually.
But my experience with TinyPilot does support lessons I've been taught that you should launch early, potentially without even a working product ready. In previous projects, I'd gotten too excited about the engineering and convinced myself I needed it first, but TinyPilot was a really small amount of engineering before I started selling.
My other takeaway from TinyPilot was how much of a difference it makes to sell a product that's aligned with my blog audience. I'm not really an IT blogger, but a lot of my readers are IT people and enthusiasts, so my blog was a huge advantage in finding early customers. My other products were things that had nothing to do with my blog like parsing recipe ingredients or finding live comedy.
I don't have a strict plan for my next product, but I want to focus on things that I can test quickly and drop if I can't find paying customers within the first two or three months.
Cool to see the full lifecycle from launch to sale.
Just wanted to say Thanks for that. Best wishes in becoming a new parent.
> With every other project, I had to beg and plead with people even to try my product. With TinyPilot, there was so much demand that I struggled for months to keep the product stocked.
Anyway, to mylynch: Congrats on the sale and on starting a family. Also: Thanks for documenting everything so thoroughly. It's a rare behind the scenes view into how a bootstrapped startup works in the real world. Very valuable for others to learn from your experience!
Questions business brokers ask is how many hours the owner spends each week, and if it needs to be on site.
Lifetime profit from business
Thanks for including this. Startup hype always demands a focus on the size of the exit, ignoring the other funds extracted from the venture in the form of salary and covering other costs.
But in this case, it seems the buyer found him thru his own content, which is wild.
I see how it's confusing because I say in the post that the buyer read my retrospectives, but that was after he had discovered the company through Quiet Light.
I've just edited the article to clarify that the buyer discovered my blog from Quiet Light.
>Do you ever calculate your GOOG salary and multiply it by the number of years you've been bootstrapping and have any regrets?
I've definitely thought about the number, but no, honestly no regrets at this point.
I got an email from a Google recruiter a few weeks ago offering to rehire me back at my old level with no interview or interview for a higher level, and it wasn't tempting at all.
I miss things about Google, and there are things I hate about bootstrapping, but overall, I definitely prefer the autonomy of running my own company.
that's pretty impressive! as a recent ex googler i cannot imagine them ever offering me that deal
>How did you get people to trust your product? If I see something like this from a small outfit for a cheap price, I will assume that the price is being made up some where else. How did you assure you customers you cant access their machines when you want. Or that your service will offer the reliability they will need?
Good question!
Customers didn't express concern to me about me accessing their machines. I don't know if I did something special to engender trust or the average customer isn't so worried about that threat. I could imagine that there was some trust in the fact that my face was on the website, and I was clear about who I was and that I was located in the US. For a lot of customers, I'd imagine that I'm less likely to have backdoored the product than a faceless corporation who operates from overseas.
I also found that many of our customers found us through word of mouth. So there was trust that the product worked well because they heard recommendations from friends, co-workers, or bloggers that they liked.
All the best to you and your pending addition to the family.
God, I hope the buyer lives up to your expectations.
>Once manufacturing (and other tedium) was outsourced, I'd have thought that's "making it". More or less "passive income" presuming the employees took care of everything else. If the revenue truly was 208k, you'd just need to "hang in there" for 2.5 years and you'd make whatever you got in the sale, then everything else coming in later would be just "extra".
Yeah, I definitely understand that expectation, but I don't think I could have gotten it to 100% passive income.
Even if everything ran perfectly, everyone still needs a manager. So at the bare minimum, someone has to check in with the team, make sure everyone's being paid, etc.
But with so many moving parts in a hardware company, things are never running perfectly. There's always something going on like some part is no longer available or some process has changed with our vendors or one of our vendors has made a mistake and we have to work with them to fix it.
I theoretically could have hired a COO to manage operations, but that person would have to be comfortable managing a dev team, support teams, and making hardware/logistics decisions. So even if I could find a qualified COO willing to work for a $150k salary, the full cost after payroll taxes and benefits would likely exceed any profits the company was making.
I agree this was successful and assume it's rewarding in many other ways!
But still...
There's a lot to it that has nothing to do with earnings though and of course there's a lottery ticket that's correlated to how hard you work.
It’s hard to learn a lot of these lessons without going through it, so I really appreciate you sharing
If it's not about money, I would understand. But then why sell a growing business?
> I missed writing code....
and
> My wife and I also wanted to start a family. TinyPilot only occupied about 20% of my time, but it occupied 90% of my stress. I would have done a terrible job juggling founder stress with new parent stress.
Changes in life circumstances changed his priorities. It wouldn't be the first or last person I've heard of that wanted to go back to coding after ending up in mgmt/founding something.
1. Press on making OK money doing something you're not that enthusiastic about
2. Just pull the plug
3. Sell it
I mean, obviously #3 is the least bad of all the options. You don't know what something's going to be like until you try it. This is a "negative result" from the "what kind of business do I want to run", but now he has a better idea what might not work for him. It's probably a break-even result from a financial perspective, and a positive result from an experience gained perspective.
Fascinating that you had such a poor experience with FEI, but good to hear Quiet Light was a better fit. I also thought about potentially selling my business a while ago, and the broker, while nice, was just not interested in conducting anything more than a commodity transaction. So obviously we paused, but good for folks to know this is quite common, and if you're selling, always make sure your broker is aligned with you, otherwise find another.
I also want to point out, for posterity, that the HN commenters are often amazingly confidently wrong in their strong assertions. For Michael's 2023 retrospective post which got shared here, someone blithely wrote off TinyPilot as being worth "close to zero"[0]. Obviously that was not the case then, and this deal shows it.
Congrats Michael! All the best with next steps.
It was over three days (dinner, dinner, lunch), and I was splitting them with friends. Although, I do love dessert and probably could have eaten those all solo.
I understand it's hardware, but that's a bit rough.
> A month ago, I sold the company for $600k.
?????
Also confused here, the math doesn’t seem to be mathing up
Businesses are typically valued as a multiple of their earnings (in the stock market, this is the P/E ratio). For hardware/e-commerce companies, 2.4x is in the range of fair value.
Again, you multiply by the earnings and not the revenue. That gives you the actual sale price.
I don't want to offend, but 2.4x earnings is disappointing to me. I dream of selling a business for 5-10 times earnings and retiring. The assumption being the 5x earnings assumes the business will grow, perhaps doubling every 5 years?