Hacker News new | past | comments | ask | show | jobs | submit login
We've got a million dollars and we want to buy your micro ISV (businessofsoftware.org)
56 points by spolsky on April 21, 2009 | hide | past | favorite | 25 comments



I like their checklist of items they're looking for in a web app. It seems like a good list for anyone developing a product. This item stood out to me as insightful:

"If you’re selling your product then it must have at least a 10% conversion rate. In other words, if ten people download it, or trial it, then, on average, one person should buy it. For us, this is a sign that you’ve got a product that works. We don’t care how many customers you’ve got - the fewer the better, in fact. It’s the ratio of trials to purchases that counts. If your not charging for it then we’ll look for a sign that a significant proportion of your users are actually using it regularly."

I haven't seen so much emphasis put on conversion rate before. Since they have a lot of experience with this sort of thing I think that means it is a better indicator of success than I've previously thought. From reading articles online, I had thought of conversion rate as more related to shiny buttons and A/B tests. That stuff matters, but it really does come down to how good the product is.


Conversion rates are important to them because they buy expensive keywords via Google AdWords.

Imagine this scenario:

* say you have to pay $20 for an ad click on "SQL Server backup"

* the conversion rate is 10%

* with a 10% conversion rate, each sale will have $200 ($20 divided by 0.10) in marketing costs

* If the product costs $300, you have $100 ($300-200) in "profit"

However, if the conversion rate is 5%...

* each sale would have $400 in marketing costs ($20 divided by 0.05)

* you lose $100 per sale

Given the high cost per click for specialized keywords, they need a high conversion rate to use AdWords profitably.


In what world does a product that generates $10k/mo have a lifetime value of $400k? I'd think if a business was attractive enough to be purchased they would expect more than a 40 month lifespan.


Depends on the discount rate you apply. If I was producing a SQLServer backup product, I would apply a very high discount rate to projected cash flows because Microsoft might add a button and unknowingly halve my revenue. Or, some open source project might provide sufficient value for some would-be customers. It's like this to some degree for just about any product, software or not.

Also, I read "lifetime value" to mean NPV of revenue, not positive cash flows.


It all depends on how much of that is profit.


Perpetual revenue is never worth as much as one thinks.

First, even if your product generates $10,000/mo forever, it clearly doesn't generate infinite profits. In fact, money that comes 5 years from now is worth less than money today. Why is that? As a simple example, let's say you have the option between $380,000 today and $10,000 per month for 40 months. Which is worth more? The $380,000 is worth more because if you put it in a CD for that time period, it would become more than $400,000 by the end of that term. So, money in the future isn't worth its full value.

Second, there is risk in a product that generates monthly revenue. A competitor could come by and undercut you and you have to reduce your pricing (and therefore profits). Heck, things could get so bad that your product becomes worthless. A cash buyout doesn't carry that risk.

Third, there is money that you have to put into a product over time. So, you might have something generating $10,000 worth of revenue each month, but you have to work full-time maintaining it and improving it. As such, you are forgoing other ways of earning money. Let's say that other way would get you $4,000 per month (low for a computer programmer with such skill - remember, that $10,000 is pretax as well). So, would you rather $4,000 per month plus $400,000 or $10,000 per month? At 10% interest (the rate of growth of the S&P 500, you'd have an additional $3,333 from the interest off that $400k bringing you to $7,333). So, it might not be as much, but you still don't have the risk of your business being undercut, made obsolete, etc. And if you're earning $80,000 per year (a reasonable salary for a programmer making such valuable programs), you'd be getting $6,666+$3,333 = $10,000 in a much more stable way where you don't have to worry about where the market might go. Plus, those types of jobs usually come with at least $10,000 in benefits (health care being a big part of that). And you have no expenses since your employer is paying for the servers and such.

So, it's not such a terrible price. There's a lot of risk they're taking as the buyer and they're taking over all of the maintenance. Maybe you believe in your product more or maybe you enjoy working on it which gives other value to the project for you. But it isn't that it will only last 40 months.

Again, this might be different if this is a project that requires a few hours a month of work in which case it doesn't change your potential for other salary. It might be different if you see a great way to expand the project to get more revenue. It might be different if there are competitors nipping at your heels. Maybe I'm too risk-averse, but it's important to see how it all fits together - from the present value of future money to the risk of competition and obsolescence to the amount of time you need to put into the project and how you could alternatively earn money with that time.


There is a flip side to that coin: preserving, let alone growing at the rate of 10%, $400,000 is a lot harder than most people realize.


Those are two different conversion rates:

* The one in the article is conversion of someone who has downloaded or tried your product to paying customer.

* I assume the one you've read about in articles online is conversion of a visitor on your website to: download/trial/paying customer


I agree with the first comment on the page - 10% conversion rate is almost unheard of. It makes me suspect that this is a publicity stunt or some kind of market research gimmick and that the bar is set to deliberately exclude them from actually buying anything except something that is already a money-printing machine.

And as another commenter points out - by even talking these guys you're basically then giving them a product to clone and take your own customers away. I'd be very careful about talking to them.


Given that list, if I was as successful as they require for this, I wouldn't want to sell!


On first blush I agree, but after thinking about it for a few minutes I can see a scenario (actually, recently experience several) where once the interesting engineering problems have been solved and the major challenges overtaken, my interest in incremental growth and feature releases dwindle and I wouldn't mind turning the product over to a company interested in growing the product gracefully (so I can move on to the next challenge).

I would like to be "on retainer" though if in the future some compelling feature was needed or some great bug was to be addressed that proved challenging and/or interesting...


Answering the questions they require without getting a signed NDA is stupid.


I guess some people may be completely unable to grow their own business, but f you have a business that already has happy paying customers, then why sell out. It feels to me like you've already done the hard part.

http://www.jacksonfish.com/blog/2009/04/21/sell-or-sell-out/


I stopped reading at "We specialise in Microsoft platforms".


A lot of businesses use Microsoft platforms, and make a hell of a lot of money doing so.

Not to mention that Microsoft has come a long, long way since VB6 -- C# is a respectable language, ASP.NET MVC works pretty damn well (it powers Stack Overflow), and many of their lightweight dev products (SQL Server Express) are free-as-in-beer.

I may be a Unix guy, but I've still got a healthy amount of respect for what you can do on the Windows platform, even though it's not as 'cool' as an iPhone, Rails, or whatever-else-is-popular-right-now.


No offence meant edw519, but i thought it might be good to reproduce their entire text here, so that people like me who often read comments first might get a better picture.

  It’s got to be software aimed at software developers or sysadmins. 
  We specialise in Microsoft platforms, but will consider others too.


No offense taken.

I stand by my original comment.

If they specialize in Microsoft, that's all I need to know. I'm not interested, no matter what they have to say after the "but".


Microsoft software: That's where some money is. Especially on the Desktop.


"That's where some money is."

Not for me.

I don't object to Microsoft because of personal preference. I object because of software quality. Which translates into, "less money for me".

Without exception, every experience I've had with a customer who had Microsoft products on their server has been profoundly negative. I've had to account for things I would never dream of in a "-ux" environment. I have to regularly stop and deal with things normally outside my control. Which slows down progress and diverts attention from the critical path. In addition, these customers "think differently". I can't quite put my finger on it, but they just don't see possibilities they've never experienced.

So when someone says that they specialize in Microsoft, I have only 2 thoughts:

1. Good luck to them.

2. Not interested.


That's surely a rational response. I took care to write "some money" not "all money".


I'm not sure about that. Piracy of Windows desktop/end-user software is rampant. Maybe it's less for Windows server-related software?


Piracy is rampant but a lot of people and especially businesses don't want to be pirates. A lot of software businesses have been successful in making casual piracy difficult and confining piracy to places (usenet, hacker-or porn- themed websites, bittorrent, limewire) that most of their target customer won't feel comfortable going. They can't absolutely stop piracy but they can make it a lot less appetizing than the "all you can eat for free" that it used to be.


That's not something to brag about.


Nor are $80 PDFs on how to raise one's parakeet. But, it's apparently a profitable business. By choosing MS platforms, those users have already signaled that they're willing to pay for software. That's not a bad market to target.


From a marketing standpoint (as well as R&D), this is rational. Common marketing mechanisms for software like this are very segregated -- Hardly any conference interesting to people using Open Source stacks will have much Microsoft content, and vice versa. Blogs, magazines, etc. are all similar in this regard. So, the marginal cost of marketing another developer utility for the MS platform will be much lower than the cost of marketing the first, say, DB2 utility acquired.




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

Search: