Archive for August, 2007

The cost of building a web application in Australia

One of the great things about the Internet is that people have started sharing some great information about their products and processes. It is now possible to find out what some of the leading web applications cost to build, and therefore make a good guess as to what it is going to cost you to build one.Some costs of other web applications are shown below (US$), with the numbers taken from Read Write Web.

Dropsend: $48,012
Freshbooks: $20,000
Maya’s Mom: $70,000
Mobissimo: $60,000
Wesabe: $200,000

The numbers above suggest that angel financing (or your own cash) should be all that is required to get a prototype out the door. Anyone asking for 7 figures off a VC without having anything to show for it is trying to run before they are walking. Any VC happy to give out that kind of money without first seeing a prototype, well my contact details are here!

If you ask Guy Kawasaki however, building a web app can be done even cheaper. In a tribute to that post, the following numbers are supplied from our experience in building TrickyTix. All financial figures are in Australian dollars.

  1. 2 – The number of external developers we added to our existing team of 5 in order to create the right set of skills.
  2. 0 – The number of finalised business models we had before commencing development (although we did have a broad understanding of how we intended to make money)
  3. 0 – The number of documents (functional specification, scope of work, business requirements, business plans, change management requests, risk registers,) we wrote before starting building our app.
  4. $6,500 – Legal costs for trademarks and terms and conditions (x 2). We didn’t skimp on this, and went and found ourselves a firm that specialises in IT, IP and fast-starting companies.
  5. $500 – We spent approximately $500 registering domain names.
  6. $2,500 – Accountant fees to be spent on setting up two new companies. One to run operations, the other to hold the IP.
  7. $60 – We used Basecamp to project manage the entire development, as well as manage task allocation.
  8. 4 – By the time we push the prototype out to test, 4 months will have been spent building the app.
  9. $1,000 – Time spent developing the logo, colour palette and typography. All done in-house.
  10. $4,200 – Money spent on User interface and the design of both the website and the application. Again, all done in-house.
  11. $6,500 – Development costs for the front-end of the application. As with any modern web app a significant portion of that was spent on javascript, with 1 specialist contractor brought in to assist where required.
  12. $9,000 – Where there is a front-end, there is usually a backend. Think databases, think scripting, think geek. Half the costs here were spent getting in a specialist contractor, and the other half completed in-house.
  13. $0 – When you use open source software, the license fees are pretty sweet.
  14. $450 per month – Dedicated server to host the application. There are cheaper options, but this should suffice for now.

There are other fees still to come (merchant account for one), but it won’t be much higher than what is listed here. In the final wrap up, we end up with:

$30,710 to develop a prototype web application in Australia.

I have tried to account for in-house development at wage cost rather than actual consulting prices, but to be honest there is probably still a lot of “sweat equity” unaccounted for.

So what do you think – has the price come out lower or higher than expected? Anything you think we have failed to take into account?

Visit the Anthill forum

Comments (4)

So close I can smell it

I haven’t blogged for a while, as I was waiting for the newly minted Australian Anthill website to go live. Now it is here, I hope to be able to share our ups and downs as we attempt to grow a brand new business.So where is the prototype I hear you ask?

It is close. Real close. If it was a freshly baked loaf of bread, you would be able to smell it rising in the oven right about now.

While our regular consulting work has taken priority many times over the development of Trickytix, if we are honest with ourselves we have also neglected to properly embrace the notion of constraints.

he team at 37Signals put it best in their e-book Getting Real:

There’s never enough to go around. Not enough time. Not enough money. Not enough people.

That’s a good thing.

Instead of freaking out about these constraints, embrace them. Let them guide you. Constraints drive innovation and force focus. Instead of trying to remove them, use them to your advantage.

Constraints are limitations that actually improve the likelihood of a successful outcome. While we have embraced a number of important ones (restricted budget, small development team, tight deadlines, no wireframes just jump straight into the User Interface), we may have built “too much” software for the first prototype.

That is not to stay that the first cut will be a perfect, bug free solution (far from it!). But it may do some kind of cool things that it didn’t really need to do for the first launch. Your customers won’t thank you for spending 4 hours to make the background of an element light up when it is dragged and dropped on the screen, if by doing so you miss your ship date.

But enough of that, and back to what I promised the Anthill guys I would blog about. Everyone wants to know what things cost, especially if they are considering making the leap into self-employment themselves.

With that in mind I will get my next post out by the end of this week, with details of what it costs to build a prototype web application in Australia. Development costs, hardware and software, legal fees, accounting fees, trust companies, food, beer and everything in between.

eep in mind we are boot strapping the initial build from our own cash reserves (no VC funding at this stage), but we are an established company and so have tried to do things properly.

Therefore:

  1. If you are a freelancer in your bedroom thinking of building your own app, yes you can do it cheaper than what we have (divide our cost by at least a factor of 5).
  2. If you are a business unit within a large ASX listed company, hire 30 more people, multiply our cost by a factor of 25 and spend the next 18 months building your product. You too will miss ship date.

See you next post.

Leave a Comment