You see a lot of press these days about seed/VC money, 2 million for this and 5 million for that. I thought it would be interesting to add up what it actually cost us to build a sophisticated web application using the latest technologies (BackBone, Laravel, Coffeescript, Less, message queues, email parsing, Facebook integration) and a top notch team. The results are interesting.
I suspect most people who read this will find it either amazingly expensive or remarkably cheap! I think we did a great job and kept the costs within what we could afford. Luckily, we already have a successful product so we could be our own Angel in this endeavor and not rely on any outside funding. In fact, if we were a true startup we could have done it a lot cheaper, more on that farther down.
These numbers are all ballpark, I looked up the figures, but didn’t add it all up to the penny. These costs are for the last half a year or so that we’ve worked on Snappy. Another important note is that we were not working on Snappy that entire period. There was the continuing work on HelpSpot, support on HelpSpot and Taylor was working on Laravel. I’ve taken that into account a bit in the numbers, again ballpark.
Obviously, this is most software companies largest expense and no exception for us. I couldn’t ask for a better team and wouldn’t trade these guys for anything.
I considered rolling this into payroll, but it’s so shockingly huge I decided to break it out!
Other overhead: $20,000
This is just “stuff”. We have a small office, hardware, 401K matches, time off, software, etc.
Design: $50,000: Focus Lab
I spoke previously about my desire to implement a great design and to have a consistent brand experience for Snappy from logo through the app. This was the most (by several times) we ever paid an outside company. I think, if you have the means, it’s money well spent.
Note: this ballpark design cost posted with the permission of Focus Lab
Snappy is wired together with a lot of other great services. The monthly fees for these are shockingly affordable.
Business Insurance: $1500*
This is also an estimate. We currently pay about $2,000 in errors and omissions insurance. The extra product, extra revenue (hopefully) and being a product where we host the data will add to the cost. This number could be higher, I’m being optimistic!
So far we’ve spent very little on servers. We already have an existing infrastructure and partnership with EngineHosting. They’ve just been running everything on a custom VM cluster for us during the beta and it’s been very smooth so far.
A note here, that we did initial play with AWS when we thought we may use some of their other infrastructure pieces but it was so complicated, expensive and slow that in the end it just made sense to use a more traditional (I guess if you can call a VM cluster traditional) setup.
Yeah, so before Snappy was called Snappy it was going to be called SnapReply. I need to write up that change in direction another time, but the short version of this story is that we paid $2,000 to a domain squatter for a domain we’re not using 🙂
What’s it all mean?
That its not nearly as expensive to build a top notch product as all this money flying around makes it seem. Sure, with some VC money we could have hired more people but that doesn’t always equate to things getting done faster. In fact, it can have the opposite effect. Especially in a new product where there’s a lot of wandering around. Trying something, seeing if it works, reworking it and so on.
This isn’t to say there’s not a place for VC or even that under some scenarios I might not take some myself. But it’s so affordable to build a great app these days that it just makes sense to build it and get it rolling on your own if at all possible. See if it works, see if people like it and if you then want to take on some investors to scale it, pursue that from a more strategically advantageous position. Or don’t and just be profitable!
Update: The kids have started a Hacker News thread. I’ll follow up on comments over there